KI-Coding-Tools Changelog Hub

Claude Codev2.1.58Neu

v2.1.58

25. Februar 2026

## What's changed - Expand Remote Control to more users

Codex CLIvrust-v0.105.0Neu

0.105.0

25. Februar 2026

## New Features - The TUI now syntax-highlights fenced code blocks and diffs, adds a `/theme` picker with live preview, and uses better theme-aware diff colors for light and dark terminals. (#11447, #12581) - You can now dictate prompts by holding the spacebar to record and transcribe voice input directly in the TUI. (#3381) - Multi-agent workflows are easier to run and track: `spawn_agents_on_csv` can fan out work from a CSV with built-in progress/ETA, and sub-agents are easier to follow with nicknames, a cleaner picker, and visible child-thread approval prompts. (#10935, #12320, #12327, #12332, #12570, #12767) - The TUI picked up new convenience commands: `/copy` copies the latest complete assistant reply, while `/clear` and `Ctrl-L` clear the screen without losing thread context, with `/clear` also able to start a fresh chat. (#12444, #12520, #12613, #12628) - Approval controls are more flexible: Codex can now ask for extra sandbox permissions for a command, and you can auto-reject specific approval prompt types without turning approvals off entirely. (#11871, #12087) - App-server clients can do more with threads: `thread/list` can search by title, thread status is exposed in read/list responses and notifications, and `thread/resume` returns the latest turn inline so reconnects are less lossy. (#11776, #11786, #12578) ## Bug Fixes - Long links in the TUI stay clickable when wrapped, which also fixes related clipping and layout issues in several views. (#12067) - Several TUI interaction edge cases were fixed: queued-message editing now works in more terminals, follow-up prompts no longer get stuck if you press Enter while a final answer is still streaming, and approval dialogs now respond with the correct request id. (#12240, #12569, #12746) - `@` parsing in the chat composer is more reliable, so commands like `npx -y @scope/pkg@latest` no longer accidentally open the file picker or block submission. (#12643) - App-server websocket handling is more robust: thread listeners survive disconnects, Ctrl-C waits for in-flight turns before restarting, and websocket clients that send `permessage-deflate` can connect successfully. (#12373, #12517, #12629) - Linux sandboxed commands now get a minimal `/dev`, fixing failures in tools that need entropy or other standard device nodes. (#12081) - `js_repl` now reports uncaught kernel failures more clearly, recovers cleanly afterward, and correctly attaches `view_image` results from nested tool calls. (#12636, #12725) ## Documentation - Added a public security policy with Bugcrowd reporting guidance. (#12193) - Updated install and local workflow docs to use `cargo install --locked cargo-nextest` and to avoid routine `--all-features` builds unless you specifically need full feature coverage. (#12377, #12429) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.104.0...rust-v0.105.0 - #12071 Use V2 websockets if feature enabled @pakrym-oai - #12052 feat(core): zsh exec bridge @owenlin0 - #12072 Add message phase to agent message thread item @mousseau-oai - #12077 got rid of experimental_mode for configtoml @won-openai - #12036 codex-api: realtime websocket session.create + typed inbound events @aibrahim-oai - #12073 Add model-visible context layout snapshot tests @charley-oai - #12096 Updated issue labeler script to include safety-check label @etraut-openai - #11944 [js_repl] paths for node module resolution can be specified for js_repl @aaronl-openai - #11798 fix: Restricted Read: /System is too permissive for macOS platform de… @leoshimo-oai - #12015 Enable default status line indicators in TUI config @jif-oai - #12124 feat: phase 1 and phase 2 e2e latencies @jif-oai - #12121 feat: phase 2 usage @jif-oai - #12120 feat: memory usage metrics @jif-oai - #12133 feat: validate agent config file paths @jif-oai - #12137 nit: change model for phase 1 @jif-oai - #12069 Stop filtering model tools in js_repl_tools_only mode @fjord-oai - #12135 feat: better slug for rollout summaries @jif-oai - #12157 Disable collab tools during review delegation @jif-oai - #11802 Fixed a hole in token refresh logic for app server @etraut-openai - #12105 fix: file watcher @jif-oai - #12167 memories: bump rollout summary slug cap to 60 @zuxin-oai - #12177 js_repl: canonicalize paths for node_modules boundary checks @fjord-oai - #12025 app-server support for Windows sandbox setup. @iceweasel-oai - #12187 fix: Remove citation @zuxin-oai - #12180 [apps] Temporary app block. @mzeng-openai - #11786 app-server: expose loaded thread status via read/list and notifications @euroelessar - #12038 state: enforce 10 MiB log caps for thread and threadless process logs @charley-oai - #12203 Stabilize app-server detached review and running-resume tests @charley-oai - #12211 [apps] Update apps allowlist. @mzeng-openai - #12081 fix(linux-sandbox): mount /dev in bwrap sandbox @viyatb-oai - #12164 Update docs links for feature flag notice @etraut-openai - #12231 Adjust memories rollout defaults @jif-oai - #12152 feat: sub-agent injection @jif-oai - #12233 feat: add shell snapshot failure reason @jif-oai - #12193 docs: add codex security policy @viyatb-oai - #12228 feat: add configurable write_stdin timeout @jif-oai - #11787 Adjust MCP tool approval handling for custom servers @colby-oai - #12179 Move previous turn context tracking into ContextManager history @charley-oai - #12244 Restore phase when loading from history @mousseau-oai - #12101 client side modelinfo overrides @sayan-oai - #12251 Add configurable agent spawn depth @jif-oai - #12254 chore: increase stack size for everyone @jif-oai - #12250 feat: no timeout mode on ue @jif-oai - #12255 chore: consolidate new() and initialize() for McpConnectionManager @bolinfest - #12256 Revert "feat: no timeout mode on ue" @jif-oai - #12258 Undo stack size Bazel test hack @charley-oai - #12264 Fix flaky test @jif-oai - #12087 feat: add Reject approval policy with granular prompt rejection controls @bolinfest - #12253 Skip removed features during metrics emission @jif-oai - #12265 Clarify cumulative proposed_plan behavior in Plan mode @charley-oai - #12266 app-server tests: reduce intermittent nextest LEAK via graceful child shutdown @bolinfest - #11953 ws turn metadata via client_metadata @pash-openai - #11778 fix(bazel): replace askama templates with include_str! in memories @rupurt - #11382 Add configurable MCP OAuth callback URL for MCP login @dkumar-oai - #12274 [bazel] Fix proc_macro_dep libs @zbarsky-openai - #12054 feat(config): add permissions.network proxy config wiring @viyatb-oai - #12269 tests(thread_resume): interrupt running turns in resume error-path tests @bolinfest - #12009 Update pnpm versions to fix cve-2026-24842 @mjr-openai - #12080 skill-creator: lazy-load PyYAML in frontmatter parsing @xl-openai - #12271 tests: centralize in-flight turn cleanup helper @bolinfest - #12286 app-server: fix flaky list_apps_returns_connectors_with_accessible_flags test @bolinfest - #11776 app-server: improve thread resume rejoin flow @maxj-oai - #11822 [apps] Store apps tool cache in disk to reduce startup time. @mzeng-openai - #12221 memories: add rollout_summary_file header to raw memories and tune prompts @zuxin-oai - #12309 Set memories phase reasoning effort constants @jif-oai - #12315 chore: nit explorer @jif-oai - #12320 feat: add nick name to sub-agents @jif-oai - #12328 chore: better agent names @jif-oai - #12326 disable collab for phase 2 @jif-oai - #12267 Add MCP server context to otel tool_result logs @colby-oai - #12327 feat: cleaner TUI for sub-agents @jif-oai - #12332 feat: better agent picker in TUI @jif-oai - #12344 feat: do not enqueue phase 2 if not necessary @jif-oai - #12340 fix: simplify macOS sleep inhibitor FFI @yvolovich-cyber - #12347 fix: nick name at thread/read @jif-oai - #12294 Reuse connection between turns @pakrym-oai - #12287 app-server: add JSON tracing logs @maxj-oai - #12140 Refactor network approvals to host/protocol/port scope @viyatb-oai - #12300 js_repl: block wrapped payload prefixes in grammar @fjord-oai - #11368 fix(network-proxy): add unix socket allow-all and update seatbelt rules @viyatb-oai - #12275 js_repl: remove codex.state helper references @fjord-oai - #12289 CODEX-4927: Surface local login entitlement denials in browser @daniel-oai - #12205 ci(bazel): install Node from node-version.txt in remote image @fjord-oai - #12312 feat: add config `allow_login_shell` @jif-oai - #12086 [apps] Implement apps configs. @mzeng-openai - #12353 fix(core): require approval for destructive MCP tool calls @colby-oai - #12218 app-server: harden disconnect cleanup paths @maxj-oai - #12291 core tests: use hermetic mock server in review suite @viyatb-oai - #12374 [apps] Enforce simple logo url format. @mzeng-openai - #12377 docs: use --locked when installing cargo-nextest @derekf-oai - #12370 Add ability to attach extra files to feedback @pakrym-oai - #12306 Move sanitizer into codex-secrets @viyatb-oai - #12379 clarify model_catalog_json only applied on startup @sayan-oai - #12307 Show model/reasoning hint when switching modes @charley-oai - #12403 [apps] Fix gateway url. @mzeng-openai - #12240 fix(tui): queued-message edit shortcut unreachable in some terminals @fcoury - #12405 [apps] Bump MCP tool call timeout. @mzeng-openai - #12406 fix: explicitly list name collisions in JSON schema generation @bolinfest - #12301 Add field to Thread object for the latest rename set for a given thread @natea-oai - #12268 Wire realtime api to core @aibrahim-oai - #12415 fix(nix): include libcap dependency on linux builds @rgodha24 - #12416 Add experimental realtime websocket URL override @aibrahim-oai - #12303 Improve Plan mode reasoning selection flow @charley-oai - #12418 Add experimental realtime websocket overrides and text mirroring @aibrahim-oai - #12381 fix: address flakiness in thread_resume_rejoins_running_thread_even_with_override_mismatch @bolinfest - #12422 feat: use OAI Responses API MessagePhase type directly in App Server v2 @bolinfest - #12376 test(app-server): wait for turn/completed in turn_start tests @yvolovich-cyber - #12408 ignore v1 in JSON schema codegen @bolinfest - #12314 fix(core) Filter non-matching prefix rules @dylan-hurd-oai - #12429 feat: discourage the use of the --all-features flag @bolinfest - #12252 Fix compaction context reinjection and model baselines @charley-oai - #12427 chore: move config diagnostics out of codex-core @bolinfest - #12430 Collapse waited message @pakrym-oai - #12432 chore: remove codex-core public protocol/shell re-exports @bolinfest - #12434 fix: codex-arg0 no longer depends on codex-core @bolinfest - #12435 refactor(core): move embedded system skills into codex-skills crate @bolinfest - #12440 chore: delete empty codex-rs/code file @bolinfest - #12441 Delete AggregatedStream @pakrym-oai - #11293 feat(linux-sandbox): implement proxy-only egress via TCP-UDS-TCP bridge @viyatb-oai - #12410 profile-level model_catalog_json overrie @sayan-oai - #12420 Prevent replayed runtime events from forcing active status @etraut-openai - #12428 Prefer v2 websockets if available @pakrym-oai - #12419 Improve token usage estimate for images @etraut-openai - #12474 fix: make skills loader tests hermetic with ~/.agents skills @bolinfest - #12473 core: preserve constrained approval/sandbox policies in TurnContext @bolinfest - #12067 fix(tui): preserve URL clickability across all TUI views @fcoury - #12469 Route inbound realtime text into turn start or steer @aibrahim-oai - #12479 Revert "Route inbound realtime text into turn start or steer" @aibrahim-oai - #12475 fix: make realtime conversation flake test order-insensitive @bolinfest - #12476 Make shell detection tests robust to Nix shell paths @rupurt - #11447 feat(tui): syntax highlighting via syntect with theme picker @fcoury - #12373 app-server: retain thread listener across disconnects @maxj-oai - #12444 feat(tui) /clear @won-openai - #12485 fix(core) exec policy parsing 3 @dylan-hurd-oai - #12423 Send events to realtime api @aibrahim-oai - #12364 feat: monitor role @jif-oai - #12313 Handle orphan exec ends without clobbering active exploring cell @jif-oai - #12455 feat(tui): support Alt-d delete-forward-word @dougEfresh - #12480 Revert "Revert "Route inbound realtime text into turn start or steer"" @aibrahim-oai - #12509 Sort themes case-insensitively in picker @etraut-openai - #12511 Add C# syntax option to highlight selections @etraut-openai - #12513 Add PR babysitting skill for this repo @etraut-openai - #12518 test: vendor zsh fork via DotSlash and stabilize zsh-fork tests @bolinfest - #12553 Return image content from view_image @pakrym-oai - #12555 refactor: decouple MCP policy construction from escalate server @bolinfest - #12559 chore: nit name @jif-oai - #12562 chore: awaiter @jif-oai - #12565 chore: add doc to memories @jif-oai - #12568 chore: phase 2 name @jif-oai - #12571 fix: TUI constraint @jif-oai - #12570 feat: keep dead agents in the agent picker @jif-oai - #12575 feat: agent nick names to model @jif-oai - #12500 feat: add uuid helper @jif-oai - #12580 chore: rename memory feature flag @jif-oai - #12541 Allow exec resume to parse output-last-message flag after command @etraut-openai - #12579 feat: role metrics multi-agent @jif-oai - #12141 feat: land sqlite @jif-oai - #12576 chore: better bazel test logs @jif-oai - #12028 remove feature flag collaboration modes @aibrahim-oai - #12520 tweaked /clear to support clear + new chat, also fix minor bug for macos terminal @won-openai - #12556 refactor: normalize unix module layout for exec-server and shell-escalation @bolinfest - #12421 app-server: box request dispatch future to reduce stack pressure @charley-oai - #12528 chore(deps): bump libc from 0.2.180 to 0.2.182 in /codex-rs @dependabot - #12529 chore(deps): bump syn from 2.0.114 to 2.0.117 in /codex-rs @dependabot - #12583 Use Arc-based ToolCtx in tool runtimes @bolinfest - #12540 fix: add ellipsis for truncated status indicator @sayan-oai - #12609 fix(tui): recover on owned wrap mapping mismatch @fcoury - #12569 fix(tui): queue steer Enter while final answer is still streaming to prevent dead state @guidedways - #12530 chore(deps): bump owo-colors from 4.2.3 to 4.3.0 in /codex-rs @dependabot - #12549 fix: show command running in background terminal in details under status indicator @sayan-oai - #3381 voice transcription @nornagon-openai - #12619 Handle realtime spawn_transcript delegation @aibrahim-oai - #11408 Update models.json @github-actions - #12629 app-server: fix connecting via websockets with `Sec-WebSocket-Extensions: permessage-deflate` @JaviSoto - #12632 refactor: delete exec-server and move execve wrapper into shell-escalation @bolinfest - #12638 refactor: decouple shell-escalation from codex-core @bolinfest - #12357 feat(core): persist network approvals in execpolicy @viyatb-oai - #12648 fix(exec) Patch resume test race condition @dylan-hurd-oai - #12049 Support implicit skill invocation analytics events @alexsong-oai - #12647 Avoid `AbsolutePathBuf::parent()` panic under `EMFILE` by skipping re-absolutization @etraut-openai - #12633 Add skill approval event/response @pakrym-oai - #12650 chore: rm hardcoded PRESETS list @sayan-oai - #12652 Simplify skill tracking @pakrym-oai - #12653 memories: tighten consolidation prompt schema and indexing guidance @zuxin-oai - #12667 feat: mutli agents persist config overrides @jif-oai - #12635 memories: tighten memory lookup guidance and citation requirements @zuxin-oai - #12663 fix: replay after `/agent` @jif-oai - #11258 Send warmup request @pakrym-oai - #12688 feat: use process group to kill the PTY @jif-oai - #11871 feat(core) Introduce Feature::RequestPermissions @dylan-hurd-oai - #12628 ctrl-L (clears terminal but does not start a new chat) @won-openai - #9859 feat(network-proxy): add MITM support and gate limited-mode CONNECT @viyatb-oai - #12649 feat: run zsh fork shell tool via shell-escalation @bolinfest - #12643 Fix @mention token parsing in chat composer @etraut-openai - #12658 fix: also try matching namespaced prefix for modelinfo candidate @sayan-oai - #11766 feat(sleep-inhibitor): add Linux and Windows idle-sleep prevention @yvolovich-cyber - #12581 feat(tui): add theme-aware diff backgrounds with capability-graded palettes @fcoury - #12697 Ensure shell command skills trigger approval @pakrym-oai - #12707 refactor: remove unused seatbelt unix socket arg @bolinfest - #12687 Add TUI realtime conversation mode @aibrahim-oai - #12639 Honor `project_root_markers` when discovering `AGENTS.md` @etraut-openai - #10935 Agent jobs (spawn_agents_on_csv) + progress UI @daveaitel-openai - #12700 revert audio scope @nornagon-openai - #12711 fix: temp remove citation @zuxin-oai - #12613 feat(tui) - /copy @won-openai - #12695 Add app-server event tracing @pakrym-oai - #12717 Raise image byte estimate for compaction token accounting @etraut-openai - #12724 fix: make EscalateServer public and remove shell escalation wrappers @bolinfest - #12517 codex-rs/app-server: graceful websocket restart on Ctrl-C @maxj-oai - #12636 fix(js_repl): surface uncaught kernel errors and reset cleanly @fjord-oai - #12729 fix: clarify the value of SkillMetadata.path @bolinfest - #12719 feat: pass helper executable paths via Arg0DispatchPaths @bolinfest - #12720 add AWS_LC_SYS_NO_JITTER_ENTROPY=1 to release musl build step to unblock releases @sayan-oai - #12725 Fix js_repl view_image attachments in nested tool calls @fjord-oai - #12656 chore: change catalog mode to enum @sayan-oai - #12731 chore: migrate additional permissions to PermissionProfile @celia-oai - #12407 tests(js_repl): stabilize CI runtime test execution @fjord-oai - #12737 feat: add experimental additionalPermissions to v2 command execution approval requests @celia-oai - #12372 feat: update Docker image digest to reflect #12205 @fjord-oai - #12746 fix: chatwidget was not honoring approval_id for an ExecApprovalRequestEvent @bolinfest - #12185 tests(js_repl): remove node-related skip paths from js_repl tests @fjord-oai - #12358 feat(ui): add network approval persistence plumbing @viyatb-oai - #12730 feat: zsh-fork forces scripts/**/* for skills to trigger a prompt @bolinfest - #12750 fix: keep shell escalation exec paths absolute @bolinfest - #12753 Surface skill permission profiles in zsh-fork exec approvals @celia-oai - #12319 feat: add service name to app-server @jif-oai - #12692 fix: flaky test due to second-resolution for thread ordering @jif-oai - #12578 feat: add search term to thread list @jif-oai - #12660 Support external agent config detect and import @alexsong-oai - #12756 feat: record whether a skill script is approved for the session @bolinfest - #12767 Display pending child-thread approvals in TUI @jif-oai - #12768 feat: add large stack test macro @jif-oai - #12666 feat: adding stream parser @jif-oai - #12761 feat: record memory usage @jif-oai - #12772 nit: migration @jif-oai - #12352 otel: add host.name resource attribute to logs/traces via gethostname @mcgrew-oai - #12770 chore: unify max depth parameter @jif-oai - #12787 feat: fix sqlite home @jif-oai

OpenCodev1.2.14Neu

v1.2.14

25. Februar 2026

## Core - Add message delete endpoint (@shantur) ## TUI - Consume stdout concurrently with process exit in auth login (@Ayushlm10) **Thank you to 2 community contributors:** - @Ayushlm10: - fix: consume stdout concurrently with process exit in auth login (#15058) - @shantur: - feat(core): add message delete endpoint (#14417)

OpenCodev1.2.13Neu

v1.2.13

25. Februar 2026

No notable changes

OpenCodev1.2.12Neu

v1.2.12

25. Februar 2026

## Core - Synchronize changes - Temporarily disable plan enter tool to prevent unintended mode switches during task execution - Migrate Bun.spawn to Process utility with timeout and cleanup - Disable Bun config cache in CI - Await git ID cache write in project module - Import custom tools via file URL ## TUI - Add Go SDK code generation script - Show LSP errors for apply_patch tool ## Desktop - Enhance Windows app resolution and UI loading states (@neriousy) - Update desktop README for accuracy **Thank you to 1 community contributor:** - @neriousy: - feat(desktop): enhance Windows app resolution and UI loading states (#13320)

Claude Codev2.1.56Neu

v2.1.56

25. Februar 2026

## What's changed - VS Code: Fixed another cause of "command 'claude-vscode.editor.openLast' not found" crashes

OpenClawv2026.2.24Neu

openclaw 2026.2.24

25. Februar 2026

### Änderungen - Auto-Reply/Abort-Verknüpfungen: Ausklappen eigenständige Stopp-Phrasen (`stop openclaw`, `stop action`, `stop run`, `stop agent`, `please stop` und verwandte Varianten), nachfolgende Zeichensetzung akzeptieren (z. B. `STOP OPENCLAW!!!`), mehrsprachige Stopp-Schlüsselwörter (einschließlich ES/FR/ZH/HI/AR/JP/DE/PT/RU-Formen) hinzufügen und exakten `do not do that` als Stopp-Auslöser behandeln, während strikte eigenständige Übereinstimmung beibehalten wird. (#25103) Danke an @steipete und @vincentkoc. - Android/App UX: eine native Vier-Schritte-Onboarding-Flow liefern, Nach-Onboarding in eine Fünf-Tab-Shell verschieben (Connect, Chat, Voice, Screen, Settings), eine vollständige Connect-Einrichtung/Manual-Modus Screen hinzufügen und Android-Chat-/Settings-Oberflächen für das neue Navigationsmodell aktualisieren. - Talk/Gateway-Konfig: provisioning-neutrale Talk-Konfiguration mit Legacy-Kompatibilität hinzufügen und Gateway Talk ElevenLabs-Konfigurationsmetadaten für Setup- bzw. Status-Oberflächen freigeben. - Security/Audit: `security.trust_model.multi_user_heuristic` hinzufügen, um wahrscheinlich geteilten Benutzerzugang zu kennzeichnen, und das Vertrauen-Modell des persönlichen Assistenten zu erläutern, mit Verschärfungsleitfäden für absichtliche Multi-User-Setups (`sandbox.mode="all"`, arbeitsbereichsgebundene FS, reduzierter Tool-Surface, keine persönlichen/ privaten Identitäten auf geteilten Laufzeiten). - Abhängigkeiten: zentrale Laufzeit- und Tooling-Pakete im Arbeitsbereich aktualisieren (Bedrock SDK, pi runtime stack, OpenAI, Google auth, oxlint/oxfmt), wobei gezielt `@buape/carbon` unverändert pinned bleibt. ### Breaking - **BREAKING:** Heartbeat-Zustellung blockiert nun direkte/DM-Ziele, wenn bei der Ziel-Parsing eine direkte Chat-Identifikation erkannt wird (z. B. `user:<id>`, Telegram-Benutzer-Chat-IDs oder WhatsApp direkte Nummern/JIDs). Heartbeat-Läufe werden weiterhin ausgeführt, aber direkte Nachrichten-Zustellungen werden übersprungen; nur Nicht-DM-Ziele (z. B. Kanal/Gruppe) können ausgehende Heartbeat-Nachrichten erhalten. - **BREAKING:** Security/Sandbox: blockiere standardmäßig Docker `network: "container:<id>"` Namespace-Beitrittsmodus für Sandbox- und Sandbox-Browser-Container. Um dieses Verhalten beizubehalten, setze `agents.defaults.sandbox.docker.dangerouslyAllowContainerNamespaceJoin: true` (Break-Glass). Danke an @tdjackey für den Bericht. ### Fixes - Routing/Sitzungs-Isolation: harte Follow-up-Routing-Behandlung, sodass explizite Cross-Channel-Origin-Antworten nie beim Route-Fail auf den aktiven Dispatcher zurückfallen; überschüssige Warteschlangen-Metadaten (`channel`/`to`/`thread`) beim Follow-up-Drain beibehalten; Ursprungskanal-Kontext gegenüber internen Provider-Tags für Embedded-Followups bevorzugen. Verhindert, dass Webchat/Control-UI-Kontext Discord-auf Discord-Zuordnungen in geteilten Sitzungen kapert. (#25864) Danke @Gamedesigner. - Security/Routing: Closed-Fehlerbehandlung für geteilte Sitzungen bei Cross-Channel-Antworten, indem Outbound-Zielauflösung an Metadaten des aktuellen Turn-Quellkanals gebunden wird (statt veralteter Session-Route-Fallbacks), und diese Turn-Quell-Felder durch Gateway + Befehls-Planer mit Regressionstestung weitergeleitet werden. (#24571) Danke @brandonwise. - Heartbeat-Routing: Verhindern von Heartbeat-Lecks/Spam in Discord und anderen Direktnachrichten-Zielen durch Blockieren direkter Chat-Heartbeat-Ziele und interne Keepalive-Prompts/ Cron-Exec-Anweisungen. (#25871) - Heartbeat Defaults/Prompts: impliziten Ziel-Heartbeat von `last` zu `none` wechseln (Opt-in für externe Lieferung) und interne Cron/Exec-Formulierungen verwenden, wenn Lieferung deaktiviert ist, damit Hintergrundprüfungen das Relay-Verhalten der Nutzer nicht beeinflussen. (#25871, #24638, #25851) - Auto-Replies/Heartbeat-Queueing: Heartbeat-Läufe fallen lassen, wenn eine Sitzung bereits einen aktiven Lauf hat, statt eine veraltete Follow-up-Warteschlange zu erstellen; vermeidet Duplikate bei der Antwortverzweigung nach Drain der Queue. (#25610, #25606) Danke an @mcaxtr. - Cron/Heartbeat-Lieferung: Vermeide das Erben des gecachten Session-`lastThreadId` für Zielauflösung im Heartbeat-Modus, außer ein Thread/Thema ist explizit angefordert; Ankündigungs-Modus Cron- und Heartbeat-Lieferungen bleiben auf Top-Ebene statt in aktiven Konversationsthreads. (#25730) Danke @markshields-tl. - Messaging-Tool-Deduplizierung: Ursprungs-Kanal-Metadaten als maßgeblich behandeln für die Unterdrückung derselben Zieladresse bei `message.send` in proaktiven Läufen (Heartbeat/Cron/Exec-Event), inklusive synthetischer Provider-Kontexte; so verursachen `delivery-mirror` Transkript-Einträge keine doppelten Telegram-Sendungen mehr. (#25835) Danke @jadeathena84-arch. - Channels/Typing Keepalive: Typing-Callbacks bei Keepalive-Intervallen bei langen Antworten aktualisieren; Keepalive-Timer bei Idle/Cleanup über Core + Extensions-Dispatchers-Callsites löschen, damit Typing-Indikatoren nicht mitten in der Inferenz ablaufen. (#25886, #25882) Danke @stakeswky. - Agents/Model-Fallback: Wenn ein Lauf derzeit ein konfiguriertes Fallback-Modell nutzt, weiter durch die konfigurierte Fallback-Kette laufen statt direkt auf Primary-Only zu fallen; verhindert Sackgassen-Fehler, wenn Primary im Cooldown bleibt. (#25922, #25912) Danke @Taskle. - Gateway/Models: explizite `agents.defaults.models` Allowlist-Referenzen auch dann berücksichtigen, wenn das Bundle-Modell-Katalog-Daten veraltet sind; fehlende Allowlist-Einträge in `models.list synthetisieren` und Auswahl von `sessions.patch`/`/model` für diese Referenzen ohne falsche `model not allowed`-Fehler erlauben. (#20291) Danke @kensipe, @nikolasdehor und @vincentkoc. - Control UI/Agents: `agents.defaults.model.fallbacks` in die Overview-Fallback-Eingabe übernehmen, wenn kein per-Agent-Modell-Eintrag existiert, explizite per-Agent-Fallback-Overrides (einschließlich leerer Listen) aber beibehalten. (#25729, #25710) Danke @Suko. - Automation/Subagent/Cron Zuverlässigkeit: `ANNOUNCE_SKIP` in `sessions_spawn`-Abschluss-/Direktankündigungen beachten; transiente direkte Ankündigungs-Wiederholungen bei Kanal-Unverfügbarkeit (z. B. WhatsApp Listener-Reconnect-Fenster) hinzufügen; `cron` im `coding` Tool-Profil aufnehmen, damit Cron-Aktionen bei expliziter Gateway-Policy-Ausnahme ausgeführt werden können. (#25800, #25656, #25842, #25813, #25822, #25821) Danke an @astra-fer, @aaajiao, @dwight11232-coder, @kevinWangSheng, @widingmarcus-cyber und @stakeswky. - Discord/Voice Zuverlässigkeit: Laufzeitabhängige DAVE-Abhängigkeit wiederherstellen (`@snazzah/davey`), konfigurierbare DAVE-Beitrittsoptionen (`channels.discord.voice.daveEncryption` und `channels.discord.voice.decryptionFailureTolerance`) hinzufügen, Sprach-Listener/Sitzungs-Teardown bereinigen, gegen veraltete Verbindungs-Ereignisse absichern und kontrolliertes Wiederbeitreten nach wiederholten Decrypt-Fehlern auslösen, um STT-Stabilität bei DAVE-Empfangsfehlern zu verbessern. (#25861, #25372, #24883, #24825, #23890, #23105, #22961, #23421, #23278, #23032) - Discord/Block-Streaming: Block-Stream-Antworten wiederherstellen, indem nur Begründungs-Payloads unterdrückt werden (nicht alle `block` Payloads), fehlende Discord-Antworten im Modus `channels.discord.streaming=block` korrigieren. (#25839, #25836, #25792) Danke an @pewallin. - Discord/Proxy + Reaktionen + Model-Picker: Thread-Channel-Proxy abrufen in eingehende Mediendownloads/Sticker

Claude Codev2.1.55Neu

v2.1.55

25. Februar 2026

## What's changed - Fixed BashTool failing on Windows with EINVAL error

Gemini CLIv0.30.0Neu

Release v0.30.0

25. Februar 2026

## Änderungen seit dem letzten Update * feat(ux): Textumbruchfähigkeiten für Markdown-Tabellen hinzugefügt von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/18240 * Revertiere "fix(mcp): sicherstellen, dass MCP-Transport geschlossen wird, um Speicherlecks zu verhindern" von @skeshive in https://github.com/google-gemini/gemini-cli/pull/18771 * chore(release): Version auf 0.30.0-nightly.20260210.a2174751d erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18772 * chore: ungenutzte und nicht aufgeführte Abhängigkeiten in packages/core aufräumen von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/18762 * chore(core): Text der Activate-Skill-Eingabeaufforderung direkter formulieren von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/18605 * Add autoconfigure memory usage setting to the dialog von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/18510 * fix(core): Rennbedingung in der Policy-Persistierung verhindern von @braddux in https://github.com/google-gemini/gemini-cli/pull/18506 * fix(evals): False-Positive im hierarchischen Speichertest verhindern von @Abhijit-2592 in https://github.com/google-gemini/gemini-cli/pull/18777 * test(evals): alle `save_memory`-Evals als `USUALLY_PASSES` kennzeichnen aufgrund von Zuverlässigkeitsproblemen von @jerop in https://github.com/google-gemini/gemini-cli/pull/18786 * feat(cli): Einstellung zum Verstecken der Shortcuts-Hinweis-UI hinzugefügt von @LyalinDotCom in https://github.com/google-gemini/gemini-cli/pull/18562 * feat(core): formelle 5-Phasen-Sequenzplanungs-Worklow von @jerop in https://github.com/google-gemini/gemini-cli/pull/18759 * Einführung von Limits für Suchergebnisse von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/18767 * fix(cli): Schließen der Debug-Konsole nach automatischem Öffnen über Flimmern zulassen von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/18795 * feat(masking): Ausgabe-Maskierung von Tools standardmäßig aktivieren von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18564 * perf(ui): Tabellenrendering durch Memoisierung gestylter Zeichen optimieren von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/18770 * feat: Mehrzeilige Textantworten im AskUser-Tool von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/18741 * perf(cli): Große Debug-Logs zuschneiden und Nachrichtenverlauf begrenzen von @mattKorwel in https://github.com/google-gemini/gemini-cli/pull/18663 * fix(core): vollständige MCP-Erkennung, wenn konfigurierte Server übersprungen werden von @LyalinDotCom in https://github.com/google-gemini/gemini-cli/pull/18586 * fix(core): CLI-Version zwischenspeichern, um Konsistenz während Sitzungen sicherzustellen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/18793 * fix(cli): doppelte Darstellung in shpool beheben und VSCode-Lint-Warnungen adressieren von @braddux in https://github.com/google-gemini/gemini-cli/pull/18704 * feat(plan): Plan-Modus-Policy-Overrides dokumentieren und validieren von @jerop in https://github.com/google-gemini/gemini-cli/pull/18825 * Fix Drucktaste zum Beenden des Auswahlmodus durch eine beliebige Taste von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/18421 * fix(cli): F12-Verhalten aktualisieren, Drawer nur zu öffnen, wenn Browser scheitert von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/18829 * feat(plan): Fähigkeiten im Plan-Modus aktivieren lassen von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/18817 * docs(plan): Dokumentation zu Plan-Modus-Werkzeugen hinzufügen von @jerop in https://github.com/google-gemini/gemini-cli/pull/18827 * Entferne experimentellen Hinweis in Extensions-Einstellungen-Dokumentation von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/18822 * Update Prompt- und Grep-Werkdefinitions zur Begrenzung der Kontextgröße von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/18780 * docs(plan): Dokumentation für `ask_user`-Tool hinzufügen von @jerop in https://github.com/google-gemini/gemini-cli/pull/18830 * Revert unbeabsichtigter Credential-Exposition von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/18840 * feat(core): interne Utility-Modelle auf Gemini 3 aktualisieren von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/18773 * feat(a2a): Value-Resolver für Auth-Credential-Auflösung hinzufügen von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/18653 * Removed getPlainTextLength von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/18848 * Weitere Tweaks am Grep-Prompt von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/18846 * refactor(cli): Reactive useSettingsStore-Hook von @psinha40898 in https://github.com/google-gemini/gemini-cli/pull/14915 * fix(mcp): Sicherstellen, dass stdio MCP-Serverausführung die Umgebungsvariable GEMINI_CLI=1 gesetzt hat. von @richieforeman in https://github.com/google-gemini/gemini-cli/pull/18832 * fix(core): Headless-Modus-Erkennung für Flags und Abfrageargumente verbessern von @galz10 in https://github.com/google-gemini/gemini-cli/pull/18855 * refactor(cli): UI vereinfachen und Legacy-Dialoglogik entfernen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18566 * feat(cli): --allowed-tools und excludeTools zugunsten der Policy-Engine veralten lassen von @Abhijit-2592 in https://github.com/google-gemini/gemini-cli/pull/18508 * fix(workflows): Maintainer-Erkennung für automatisierte PR-Aktionen verbessern von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/18869 * refactor(cli): useToolScheduler konsolidieren und Legacy-Implementierung löschen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18567 * Update Changelog für v0.28.0 und v0.29.0-preview0 von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/18819 * fix(core): sicherstellen, dass Sub-Agenten registriert werden, unabhängig von tools.allowed von @mattKorwel in https://github.com/google-gemini/gemini-cli/pull/18870 * Anzeige einer Benachrichtigung, wenn es Konflikte mit einem Extensions-Befehl gibt, von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/17890 * fix(cli): '?' Shortcuts-Hilfetext beim Tastenkurzschluss und aktiven Zustand ausblenden von @LyalinDotCom in https://github.com/google-gemini/gemini-cli/pull/18583 * fix(core): bedingte Policy-Regeln priorisieren und Plan-Modus härten von @Abhijit-2592 in https://github.com/google-gemini/gemini-cli/pull/18882 * feat(core): Plan-Modus-Systemprompt für agentengesteuerte Ausführung verfeinern von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/18799 * feat(plan): Metriken für Nutzung des `AskUser`-Tools erstellen von @Adib234 in https://github.com/google-g

Claude Codev2.1.53Neu

v2.1.53

25. Februar 2026

## What's changed - Fixed a UI flicker where user input would briefly disappear after submission before the message rendered - Fixed bulk agent kill (ctrl+f) to send a single aggregate notification instead of one per agent, and to properly clear the command queue - Fixed graceful shutdown sometimes leaving stale sessions when using Remote Control by parallelizing teardown network calls - Fixed `--worktree` sometimes being ignored on first launch - Fixed a panic ("switch on corrupted value") on Windows - Fixed a crash that could occur when spawning many processes on Windows - Fixed a crash in the WebAssembly interpreter on Linux x64 & Windows x64 - Fixed a crash that sometimes occurred after 2 minutes on Windows ARM64

OpenCodev1.2.11Neu

v1.2.11

24. Februar 2026

## Core - Add workspace-serve command (experimental) - ACP both live and load share synthetic pending status preceding actual data (@noamzbr) - Replace structuredClone with spread operator for process.env in tests - Add 50ms tolerance for NTFS mtime precision in Windows FileTime assertions - Replace Unix-only test assumptions with cross-platform alternatives - Use path.sep in discovery test for cross-platform path matching - Normalize backslash paths in config rel() and file ignore on Windows - Fix plugin resolution with createRequire fallback on Windows - Harden preload cleanup against Windows EBUSY errors - Normalize git excludesFile path for Windows in tests - Stream bash output and add synthetic pending events to ACP (@noamzbr) - Add git flags for snapshot operations and fix tests for cross-platform on Windows - Handle CRLF line endings in markdown frontmatter parsing on Windows - Use path.join for cross-platform glob test assertions - Upgrade to Bun 1.3.10 canary and force baseline builds always - Normalize paths at permission boundaries on Windows - Windows path support and canonicalization (@edemaine) - Upgrade OpenTUI to v0.1.81 - Change detection on Windows, especially Cygwin (@edemaine) - Cache platform binary in postinstall for faster startup - Revert caching platform binary in postinstall for faster startup - Cache platform binary in postinstall for faster startup - Publish desktop beta releases to a separate repository - Add experimental endpoint to list all sessions - Fixed terminal issues in the app - Respect info exclude in snapshot staging - Missing plugin dependencies cause TUI to black screen (@elithrar) ## TUI - Support variant parameter in GitHub Actions and OpenCode GitHub run command (@elithrar) ## Desktop - Ignore stale part deltas in the application - Fix bug where lines remain highlighted after canceling a comment (@neriousy) - Replace error handling with serverErrorMessage utility and add ConfigInvalidError checks (@OpeOginni) - Preserve native path separators in file path helpers - Remove file tree tooltips - Update createOpenReviewFile test to match new call order - Wait for loadFile to complete before opening file tab - Windows E2E test failures due to IPv6 networking issues resolved - Correct inverted chevron direction in todo list (@kevinWangSheng) - Feed customization options - Add beta icon to desktop application - E2E test updated to current version - Remove double-border in share button - Better sound effect disabling UX - Add custom scroll view to app - Show and hide reasoning summaries in the app - Stay pinned with auto-scroll on todos, questions, and permissions - Bring back -i flag in sidecar arguments for desktop - Large text pasted into prompt input no longer causes main thread to lock ## SDK - Scripts using Turbo commands would not run on Windows **Thank you to 10 community contributors:** - @elithrar: - fix(cli): missing plugin deps cause TUI to black screen (#14432) - fix(github): support variant in github action and opencode github run (#14431) - @tuhin-cmd: - docs: add Bangla README translation (#14331) - @Seungjun0906: - docs(ko): improve wording in gitlab, ide, index, keybinds, and lsp docs (#14517) - @github-actions[bot]: - Update VOUCHED list - @pirrozani: - docs(tui): correct typo in TUI documentation (#14604) - @edemaine: - fix(desktop): change detection on Windows, especially Cygwin (#13659) - fix: Windows path support and canonicalization (#13671) - @kevinWangSheng: - fix(app): correct inverted chevron direction in todo list (#14628) - @noamzbr: - feat: ACP - stream bash output and synthetic pending events (#14079) - fix: ACP both live and load share synthetic pending status preceeding… (#14916) - @OpeOginni: - refactor: replace error handling with serverErrorMessage utility and checks for if error is ConfigInvalidError (#14685) - @neriousy: - fix(app): on cancel comment unhighlight lines (#14103)

Claude Codev2.1.52Neu

v2.1.52

24. Februar 2026

## What's changed - VS Code: Fixed extension crash on Windows ("command 'claude-vscode.editor.openLast' not found")

OpenClawv2026.2.23Neu

openclaw 2026.2.23

24. Februar 2026

### Änderungen - Providers/Kilo Gateway: erste‑Klasse‑Unterstützung des `kilocode` Providers hinzufügen (Auth, Onboarding, implizite Provider-Erkennung, Modell-Standards, Transcript-/Cache-ttl-Handling und Docs) mit Default‑Modell `kilocode/anthropic/claude-opus-4.6`. (#20212) Danke an @jrf0110 und @markijbema. - Providers/Vercel AI Gateway: Claude‑Abkürzungsmodelle (`vercel-ai-gateway/claude-*`) durch Normalisierung auf kanonische Anthropic‑geroutete Modell‑IDs akzeptieren. (#23985) Danke an @sallyom, @markbooch und @vincentkoc. - Docs/Prompt caching: dedizierte Prompt‑Caching‑Referenz hinzufügen, die `cacheRetention`, Merge‑Vorrang pro‑Agent `params`, Bedrock/OpenRouter‑Verhalten sowie Cache‑ttl + Heartbeat‑Tuning abdeckt. Danke an @svenssonaxel. - Gateway/HTTP security headers: optionale Unterstützung `gateway.http.securityHeaders.strictTransportSecurity` einführen, um `Strict-Transport-Security` für direkte HTTPS-Deployments auszugeben, mit Laufzeitverkabelung, Validierung, Tests und Härte-Dokumentation. - Sessions/Cron: Session‑Wartung verstärken mit `openclaw sessions cleanup`, zielgerichtete Speicherung pro Agent, Disk‑Budget‑Kontrollen (`session.maintenance.maxDiskBytes` / `highWaterBytes`) sowie sicherere Transcript-/Archiv‑Bereinigung + Laufprotokoll‑Aufbewahrung. (#24753) Danke an @gumadeiras. - Tools/web_search: Unterstützung für `provider: "kimi"` (Moonshot) mit Schlüssel‑/Config‑Schema-Verkabelung und einem korrigierten zweistufigen `$web_search` Tool‑Flow, der Tool‑Ergebnisse vor der finalen Synthese echot, einschließlich Zitierextraktion aus Suchergebnissen. (#16616, #18822) Danke an @adshine. - Media understanding/Video: nativer Moonshot Video‑Provider hinzufügen und Moonshot in die automatische Video‑Schlüssel­erkennung aufnehmen, plus Refactor der Videoausführung, um `entry/config/provider` BaseURL+Header‑Priorität zu honorieren (entsprechend dem Audio‑Verhalten). (#12063) Danke an @xiaoyaner0201. - Agents/Config: per‑Agent Overrides `params` unterstützen, die über Modellerwartungen gemerged werden (einschließlich `cacheRetention`), damit Mixed‑Traffic‑Agents Cache-Verhalten unabhängig abstimmen können. (#17470, #17112) Danke an @rrenamed. - Agents/Bootstrap: Bootstrap‑Datei‑Snapshots pro Session‑Key cachen und bei Session‑Reset/Löschen löschen, wodurch Prompt‑Cache‑Invalidationen aus in‑Session Writes zu AGENTS.md/MEMORY.md reduziert werden. (#22220) Danke an @anisoptera. ### Breaking - **BREAKING:** Browser‑SSRF‑Policy verwendet standardmäßig den Trusted‑Network‑Modus (`browser.ssrfPolicy.dangerouslyAllowPrivateNetwork=true`, wenn nicht gesetzt), und die kanonische Config nutzt `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork` statt `browser.ssrfPolicy.allowPrivateNetwork`. `openclaw doctor --fix` migriert den Legacy‑Schlüssel automatisch. ### Fixes - Security/Config: sensible wirkende dynamische Catchall‑Schlüssel in `config.get` Snapshots redigieren (z. B. `env.*` und `skills.entries.*.env.*`) und die Round‑trip‑Wiederherstellung für diese redaktierten Sentinel‑Werte beibehalten. Danke an @merc1305. - Tests/Vitest: lokale Parallel‑Worker‑Defaults nach Host‑Speicher stufen, Gateway seriell default bei Hosts mit wenig RAM belassen und einen Low‑Profile‑Fallback‑Befehl für speicherbeschränkte Land/Gate‑Läufe dokumentieren, um lokale OOMs zu verhindern. (#24719) Danke an @ngutman. - WhatsApp/Group policy: Filterung des `groupAllowFrom`‑Absenders bei `groupPolicy: "allowlist"` ohne explizite `groups` korrigieren – zuvor wurden alle Gruppenmeldungen auch für erlaubte Absender blockiert. (#24670) - Agents/Context pruning: `cache-ttl`-Berechtigung auf Moonshot/Kimi und ZAI/GLM‑Provider erweitern (einschließlich OpenRouter‑Modell‑Refs), sodass `contextPruning.mode: "cache-ttl"` nicht mehr stillschweigend für diese Sessions übersprungen wird. (#24497) Danke an @lailoo. - Doctor/Memory: Abfrage der default‑Agenten‑Speicherbereitschaft Gateways‑seit während `openclaw doctor` (statt aus generischer Gateway‑Gesundheit abzuleiten) prüfen; warnen, wenn Gateway‑Memory‑Probe nicht verfügbar oder bereit ist, während `openclaw configure`‑Remediationsempfehlungen beibehalten werden. (#22327) Danke an @therk. - Sessions/Store: inbound gemischte Keys für Metadaten und Routenupdates kanonisieren; legacy‑Entries in Kleinschreibweise migrieren, um Duplikat-Sessions und fehlende TUI/WebUI‑Historie zu verhindern. (#9561) Danke an @hillghost86. - Telegram/Reactions: Fehlersicherheit bei Reaktionen lockern (Policy/Token/Emoji/API), snake_case `message_id` zulassen und auf inbound Message‑ID‑Kontext zurückgreifen, wenn explicites `messageId` weggelassen wird, damit DM‑Reaktionen stabil bleiben ohne Regenerationsschleifen. (#20236, #21001) Danke an @PeterShanxin und @vincentkoc. - Telegram/Polling: Persistierte Polling‑Offsets auf Bot‑Identität beschränken und einen einzigen wartenden Runner‑Stop‑Pfad bei Abbruch/Ne Versuch verwenden, um Offset‑Durchblutungen über Tokens hinweg zu verhindern und Überschneidungen von Pollern bei Neustart/Fehlerbehebung zu vermeiden. (#10850, #11347) Danke an @talhaorak, @anooprdawar und @vincentkoc. - Telegram/Reasoning: Wenn `/reasoning off` aktiv ist, delivery‑Segmente mit reiner Reasoning‑Logik unterdrücken und rohes Fallback‑Resend von unterdrücktem `Reasoning:`/`<think>`-Text blockieren, um internes Reasoning in Legacy‑Sessions zu verhindern und trotzdem Antworten bereitzustellen. (#24626, #24518) - Agents/Reasoning: Wenn modell‑default Thinking aktiv ist (z. B. `thinking=low`), Auto‑Reasoning deaktiviert halten, es sei denn explizit aktiviert, um Leakage von `Reasoning:`‑Denkschleifen in Kanalantworten zu verhindern. (#24335, #24290) Danke an @Kay-051. - Agents/Reasoning: vermeiden, Reasoning‑required‑Fehler als Context‑Overflows zu klassifizieren, damit diese Fehlschläge nicht mehr in eine Kompaktions‑Wiederherstellung gezwungen werden. (#24593) Danke an @vincentkoc. - Agents/Models: codieren Konfig‑Grenzwerte für `agents.defaults.model` / `agents.defaults.imageModel` als `string | {primary,fallbacks}`, explizite vs. effektive Modellauflösung trennen und `models status --agent`‑Quelle so korrigieren, dass standardvererbte Agents als `defaults` gekennzeichnet werden, während Laufzeitselektion weiterhin Defaults‑Fallback respektiert. (#24210) Danke an @bianbiandashen. - Agents/Compaction: Parameter `agentDir` in manuelle `/compact`-Befehlsläufe einpflegen, damit Compaction‑Autorisierung/Profilauflösung auf den aktiven Agenten beschränkt bleibt. (#24133) Danke an @Glucksberg. - Agents/Compaction: Modellmetadaten durch die eingebettete Runtime weiterreichen, damit Safeguard‑Zusammenfassungen laufen können, wenn `ctx.model` nicht verfügbar ist, und wiederholte `"Summary unavailable due to context limits"`‑Fallback‑Zusammenfassungen vermieden werden. (#3479) Danke an @battman21, @hanxiao und @vincentkoc. - Agents/Compaction: Safeguard‑Compaction abbrechen, wenn Zusammenfassungsgenerierung nicht laufen kann (fehlendes Modell/ API‑Key oder Zusammenfassungsfehler); Verlauf beibehalten statt auf Fallback-Text `"Summary unavailable"` zu reduzieren. (#10711) Danke an @DukeDeSouth und @vincentkoc. - Agents/Tools: `session_status` lesen Transkript‑basierte Nutzung mid-turn und tail‑read Session‑Logs für kontextbasierte Berichte, ohne Full‑Log‑Scans. (#22387) Danke an @1ucian. - Agents/Overflow: zusätzliche Provider‑Context‑Overflow‑Fehlerformen erkennen (einschließlich `input length` + Varianten `max_tokens` exceed-context), damit Fehler durch Kompakt

Gemini CLIv0.29.7Neu

Release v0.29.7

24. Februar 2026

## Was geändert wurde * fix(patch): cherry-pick d96bd05 to release/v0.29.6-pr-19867 to patch version v0.29.6 and create version 0.29.7 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/20111 **Ganzes Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.6...v0.29.7

Claude Codev2.1.51Neu

v2.1.51

24. Februar 2026

## What's changed - Added `claude remote-control` subcommand for external builds, enabling local environment serving for all users. - Updated plugin marketplace default git timeout from 30s to 120s and added `CLAUDE_CODE_PLUGIN_GIT_TIMEOUT_MS` to configure. - Added support for custom npm registries and specific version pinning when installing plugins from npm sources - BashTool now skips login shell (`-l` flag) by default when a shell snapshot is available, improving command execution performance. Previously this required setting `CLAUDE_BASH_NO_LOGIN=true`. - Fixed a security issue where `statusLine` and `fileSuggestion` hook commands could execute without workspace trust acceptance in interactive mode. - Tool results larger than 50K characters are now persisted to disk (previously 100K). This reduces context window usage and improves conversation longevity. - Fixed a security issue where HTTP hooks could interpolate arbitrary environment variables from header values. Env var interpolation now requires an explicit `allowedEnvVars` list in the hook configuration. - Fixed a bug where duplicate `control_response` messages (e.g. from WebSocket reconnects) could cause API 400 errors by pushing duplicate assistant messages into the conversation. - Added `CLAUDE_CODE_ACCOUNT_UUID`, `CLAUDE_CODE_USER_EMAIL`, and `CLAUDE_CODE_ORGANIZATION_UUID` environment variables for SDK callers to provide account info synchronously, eliminating a race condition where early telemetry events lacked account metadata. - Fixed slash command autocomplete crashing when a plugin's SKILL.md description is a YAML array or other non-string type - HTTP hooks are now routed through the sandbox network proxy when sandboxing is enabled, enforcing the domain allowlist. HTTP hooks are not supported for SessionStart/Setup events. - The `/model` picker now shows human-readable labels (e.g., "Sonnet 4.5") instead of raw model IDs for pinned model versions, with an upgrade hint when a newer version is available.

Gemini CLIv0.29.6Neu

Release v0.29.6

23. Februar 2026

## Was sich geändert hat * fix(patch): cherry-pick aa9163d to release/v0.29.5-pr-19991 [CONFLICTS] by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/20039 **Voller Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.5...v0.29.6

OpenClawv2026.2.22Neu

openclaw 2026.2.22

23. Februar 2026

### Änderungen - Provider/Mistral: Unterstützung des Mistral-Anbieters hinzugefügt, einschließlich Memory Embeddings und Sprachunterstützung. (#23845) Danke an @vincentkoc. - Update/Kern: Optional integrierter automatischerUpdater für Paketinstallationen (`update.auto.*`) hinzugefügt, standardmäßig ausgeschaltet, mit stabilem Rollout-Verzögerung+Jitter und Beta-Stundentakt. - CLI/Update: `openclaw update --dry-run` zum Vorschau von Channel/Tag/Ziel/Neu-Start-Aktionen, ohne Konfiguration, Installation, Plugin-Synchronisierung oder Neustart zu verändern. - Config/UI: tag-bezogene Einstellungsfilterung hinzugefügt und Konfig-Label/ Hilfetexte erweitert, damit Felder im Dashboard-Konfigurationsbildschirm leichter zu finden und zu verstehen sind. - Kanäle/Synology Chat: nativer Synology Chat-Kanal-Plugin mit Webhook-Eingang, Direct-Message-Routing, ausgehende Send/Medien-Unterstützung, Konten-spezifische Konfiguration und DM-Richtlinien-Kontrollen. (#23012) - iOS/Talk: TTS-Segmente vorabrufen und erwartete Sprachunterbrechungsfehler unterdrücken für flüssigeres Talk-Wiedergabe. (#22833) Danke @ngutman. - Memory/FTS: Spanische und Portugiesische Stop-Wörter-Filterung für Abfrage-Erweiterung im reinen FTS-Suchmodus hinzugefügt, verbessert Gesprächserinnerung für beide Sprachen. Danke @vincentkoc. - Memory/FTS: Japanisch-bezogene Tokenisierung und Stop-Wörter-Filterung für Abfrage-Erweiterung im reinen FTS-Suchmodus (einschließlich gemischter Skripte wie ASCII + Katakana). Danke @vincentkoc. - Memory/FTS: Koreanische Stop-Wörter-Filterung und partikelsensitive Keyword-Extraktion (einschließlich gemischter Koreanisch/Englisch-Stämme) für Abfrage-Erweiterung im reinen FTS-Suchmodus. (#18899) Danke @ruypang. - Memory/FTS: Arabische Stop-Wörter-Filterung für Abfrage-Erweiterung im reinen FTS-Suchmodus zur Verringerung Gesprächs-Füller in Arabisch-Speichern. Danke @vincentkoc. - Discord/Allowlist: kanonisieren gelöste Discord-Allowlist-Namen zu IDs und Auflösung-Fluss für klareres fehlerfreies Verhalten. - Channels/Config: Kanal-Vorschau-Streaming-Konfiguration mit gemeinsamem Resolver und kanonischem Migrationspfad vereinheitlicht. - Gateway/Auth: Abgleich Der Credential-Quelle bei Aufruf/Probe/Status/Auth über gemeinsam genutzte Resolver-Helfer, tabellebene Parität über Gateway-Einstiegspunkte hinweg. - Gateway/Auth: Refaktorierung der Gateway-Credential-Auflösung und Websocket-Handshake-Pfade, Nutzung gemeinsamer typisierter Auth-Kontexte, inklusive expliziter `auth.deviceToken`-Unterstützung in Connect-Frames und Tests. - Skills: Bundle-Job entfernt `food-order` Skill aus diesem Repo; Installation/Verwaltung erfolgt künftig über ClawHub. - Docs/Subagents: Thread-bound Session guidance kanal-zuerst statt Discord-spezifisch, explizite Angabe von Thread-Unterstützung-Kanälen. (#23589) Danke @osolmaz. ### Fehlerbehebungen - Security/CLI: Sensible Werte in `openclaw config get`-Ausgabe vor dem Ausgeben von Config-Pfaden redigieren, Credential-Leckage in Terminalausgabe/History verhindern. (#13683) Danke @SleuthCo. - Install/Discord Voice: `@discordjs/opus` optional machen, sodass Openclaw-Installation/Update nicht mehr scheitert, wenn native Opus-Builds fehlschlagen; `opusscript` bleibt Runtime-Fallback-Decoder für Discord-Voice-Flows. (#23737, #23733, #23703) Danke @jeadland, @Sheetaa, @Breakyman. - Docker/Setup: `$OPENCLAW_CONFIG_DIR/identity` während `docker-setup.sh` vorab erstellen, damit CLI-Befehle, die Geräte-Identität benötigen (z. B. `devices list`), keine EACCES-Fehler bei restriktiven Bind-Mounts verursachen. (#23948) Danke @ackson-beep. - Exec/Hintergrund: Standard-Exec-Timeout nicht mehr auf Hintergrund-Sitzungen anwenden, wenn kein explizites Timeout gesetzt ist, damit langlaufende Hintergrundjobs nicht am Standard-Timeout beendet werden. (#23303) - Slack/Threading: Sitzungen: Eltern-Sitzung Forking und Thread-History-Kontext aktiv halten, indem Gate-Konzepte für erste Runde entfernt werden; Thread-History-Fetch und Antwort-Prompt-Kontext-Injektion angepasst. (#23843, #23090) Danke @vincentkoc und @Taskle. - Slack/Threading: beachte `replyToMode`, wenn Slack Top-level `thread_ts` automatisch füllt, und ignorieren Inline-Direktiven `replyToId`, wenn `replyToMode` auf `off` steht, damit Thread-Forcings nur bei expliziter Konfiguration aktiviert bleiben. (#23839, #23320, #23513) Danke @vincentkoc und @dorukardahan. - Slack/Extension: Weiterleitung von `message read` `threadId` zu `readMessages` und Nutzung von Delivery-Context `threadId` als Outbound-`thread_ts`-Fallback, damit Antworten/Lesen in dem richtigen Slack-Thread bleiben. (#22216, #22485, #23836) Danke @vincentkoc, @lan17, @dorukardahan. - Slack/Upload: aufbereitete User-IDs (U-IDs) zu DM-Channel-IDs via `conversations.open` auflösen, bevor `files.uploadV2` aufgerufen wird; `chat.postMessage` toleriert User-IDs direkt, aber `files.uploadV2` → `completeUploadExternal` validiert `channel_id` gegen `^[CGDZ][A-Z0-9]{8,}$`, verhindert Fehlermeldungen `invalid_arguments`. - Webchat/Chat: assisstant `final` Payload-Messages direkt auf Chat-Status anwenden, damit gesendete Turns sofort gerendert werden, ohne auf vollständige Historie zu warten. (#14928) Danke @BradGroux. - Webchat/Chat: bei Out-of-Band final events (z. B. Tool-Call Seitenläufe) angegebene finale Assistant-Payloads direkt anhängen, statt eine transient History-Resets zu erzwingen. (#11139) Danke @AkshayNavle. - Webchat/Performance: neu laden `chat.history` nur nach finalen Ereignissen, wenn das finale Payload keine renderbare Assistant-Nachricht enthält, um teure Voll-History-Refreshes bei normalen Turns zu vermeiden. (#20588) Danke @amzzzzzzz. - Webchat/Sitzungen: externe Sitzungsrouting-Metadaten beibehalten, wenn interne `chat.send` Turns unter `webchat` laufen, damit explizite kanal-bezogene Sitzungen (z. B. Telegram) nicht nach `webchat` neu geschrieben werden und Folge-Lieferungen fehlgeleitet werden. (#23258) Danke @binary64. - Webchat/Sitzungen: vorhandenes Sitzungs-`label` über `/new` und `/reset` Rollovers beibehalten, damit Reset-Sitzungen in der Sitzungsverlauf-Liste auffindbar bleiben. (#23755) Danke @ThunderStormer. - Gateway/Chat UI: Inline-Antwort/Audio-Direktiven-Tags aus nicht-Streaming-finalen Webchat-Übertragungen entfernen (einschließlich `chat.inject`), während leere Nachrichten-Inhalte erhalten bleiben. (#23298) Danke @SidQin-cyber. - Chat/UI: Inline-Antwort-/Audio-Direktiven-Tags (`[[reply_to_current]]`, `[[reply_to:<id>]]`, `[[audio_as_voice]]`) aus angezeigtem Chat-Verlauf, Live-Chat-Ereignis-Ausgabe und Sitzungs-Vorschau-Schnipseln entfernen, damit Steuerelement-Tags nicht mehr in sichtbare Oberflächen gelangen. - Gateway/Chat UI: Nicht-Streaming-finales `chat.send`/`chat.inject` Payload-Text mit demselben Envelope/Untrusted-Context-Stripping wie `chat.history` bereinigen, um Wrapper-Markup wie `<<<EXTERNAL_UNTRUSTED_CONTENT...>>>` in der Control UI zu vermeiden. (#24012) Danke @_mittelaltergouda. - Telegram/Media: sende eine nutzerseitige Telegram-Antwort, wenn Medien-Download fehlschlägt (nicht Größenfehler), statt Nachricht still zu verwerfen. - Telegram/Webhook: Webhook-Monitoring so lange aktiv halten, bis Gateway-Abbruchsignale eintreten, falsche Kanalabbrüche vermeiden und Schleifen bei automatischem Webhook-Neustart verhindern. - Telegram/Polling: wiederhole bei

OpenClawv2026.2.21Neu

openclaw 2026.2.21

21. Februar 2026

### Änderungen - Models/Google: Gemini 3.1-Unterstützung hinzufügen (`google/gemini-3.1-pro-preview`). - Providers/Onboarding: Volcano Engine (Doubao) und BytePlus Providers/Modelle hinzufügen (einschließlich Codierungsvarianten), Onboarding-Auth-Optionen für interaktive + nicht-interaktive Flows verknüpfen und Docs an `volcengine-api-key` ausrichten. (#7967) Danke an @funmore123. - Channels/CLI: pro Konto/Kanal `defaultTo` Outbound-Routing-Fallback hinzufügen, damit `openclaw agent --deliver` senden kann, wenn kein explizites `--reply-to` angegeben ist, wenn ein Default-Ziel konfiguriert ist. (#16985) Danke an @KirillShchetinin. - Channels: ermöglichen per-Kanal-Modell-Overrides über `channels.modelByChannel` und diese in /status notieren. Danke an @thewilloftheshadow. - Telegram/Streaming: Vorschau-Streaming-Konfiguration auf `channels.telegram.streaming` (Boolean) vereinfachen, legacy `streamMode`-Werte auto-zuordnen, und Block-vs-Partial-Vorschau-Verzweigungen entfernen. (#22012) Danke an @obviyus. - Discord/Streaming: Stream-Vorschau-Modus für Live-Entwürfe mit Teil-/Block-Optionen und konfigurierbare Chunking hinzufügen. Danke an @thewilloftheshadow. Inspiration @neoagentic-ship-it. - Discord/Telegram: konfigurierbare Lifecycle-Status-Reaktionen für queued/thinking/tool/done/error Phasen mit gemeinsamem Controller und Emoji/Timing-Overrides. Danke an @wolly-tundracube und @thewilloftheshadow. - Discord/Voice: Sprachkanal-Beitreten-Verlassen/Status über `/vc` hinzufügen, plus Auto-Join-Konfiguration für Echtzeit-Sprachunterhaltungen. Danke an @thewilloftheshadow. - Discord: konfigurierbare ephemeral Defaults für Slash-Command-Antworten hinzufügen. (#16563) Danke an @wei. - Discord: Unterstützung beim Aktualisieren von Forum-`available_tags` über Channel-Edit-Aktionen zur Forum-Tag-Verwaltung. (#12070) Danke an @xiaoyaner0201. - Discord: Inklusive Channel-Themen im vertrauenswürdigen inbound Metadata bei neuen Sessions. Danke an @thewilloftheshadow. - Discord/Subagents: thread-bound Subagent-Sitzungen auf Discord mit per-Thread Fokus/Listensteuerung und thread-bound Fortsetzungsrouting für gespawnte Hilfsagenten. (#21805) Danke an @onutc. - iOS/Chat: Chat-UI-Lärm bereinigen durch Entfernen inbound untrusted Metadata/Timestamp-Präfixe, Tool-Ausgaben zu kompakten Zusammenfassungen/Fehlern formatieren, Composer während des Tippens komprimieren und Tap-to-Dismiss Tastatur im Chat-View unterstützen. (#22122) Danke an @mbelinky. - iOS/Watch: gespiegelt Watch-Prompt-Benachrichtigungen in iOS Quick-Reply-Handling bridgen, inkl.Queued-Action-Handoff bis App-Modell-Initialisierung. (#22123) Danke an @mbelinky. - iOS/Gateway: Hintergrund-Wake und Reconnect-Verhalten stabilisieren mit Hintergrund-Reconnect-Suppression/Lease-Windows, BGAppRefresh-Wake-Fallback, Location-Wake-Hook-Drosselung, APNs-Wake-Retry+Nudge-Instrumentierung. (#21226) Danke an @mbelinky. - Auto-reply/UI: Modell-Fallback-Lifecycle-Visibility in verbose Logs, /status Active-Model-Kontext mit Fallback-Grund, kohärente WebUI-Fallback-Indikatoren. (#20704) Danke an @joshavant. - MSTeams: Duplizierte gesendete Nachrichten-Cache-Speicherung durch Entfernen doppelter pro-Nachricht Set-Speicherung und Verwenden von Zeitstempel-Map-Schlüsseln als einzige Mitgliedschaftsquelle. (#22514) Danke an @TaKO8Ki. - Agents/Subagents: Standard-Spawn-Tiefe des Subagents verwendet jetzt gemeinsam genutzte `maxSpawnDepth=2`, wodurch Default-Depth-1-Orchestrator-Spawn ermöglicht wird, während Tiefen-Richtlinienprüfungen konsistent bleiben. (#22223) Danke an @tyler6204. - Security/Agents: Owner-ID-Obfuskation mittels dediziertem HMAC-Secret aus der Konfiguration (`ownerDisplaySecret`) und Hashing-Verhalten anpassen, damit Obfuskation unabhängig von Gateway-Token-Behandlung ist. (#7343) Danke an @vincentkoc. - Security/Infra: Gateway-Lock und Tool-Call-Synthetic-IDs von SHA-1 auf SHA-256 umstellen, unveränderte Verkürzungs-Längen, um Hash-Basis zu stärken bei ansonsten deterministischem Verhalten. (#7343) Danke an @vincentkoc. - Dependencies/Tooling: nicht-blockierende Dead-Code-Scans in CI via Knip/ts-prune/ts-unused-exports hinzufügen, um ungenutzte Abhängigkeiten/Exporte früher aufzudecken. (#22468) Danke an @vincentkoc. - Dependencies/Unused Dependencies: entferne oder scope ungenutzte Root- und Extension-Dependencies (`@larksuiteoapi/node-sdk`, `signal-utils`, `ollama`, `lit`, `@lit/context`, `@lit-labs/signals`, `@microsoft/agents-hosting-express`, `@microsoft/agents-hosting-extensions-teams`, sowie plugin-local `openclaw` devDeps in `extensions/open-prose`, `extensions/lobster`, und `extensions/llm-task`). (#22471, #22495) Danke an @vincentkoc. - Dependencies/A2UI: Abhängigkeitsauflösung nach Root-Bereinigung härten (Auflösung von `lit`, `@lit/context`, `@lit-labs/signals`, `signal-utils` aus Workspace/Root) und Bündelungs-Fallback-Verhalten vereinfachen, inkl. `pnpm dlx rolldown`-Kompatibilität. (#22481, #22507) Danke an @vincentkoc. ### Fixes - Security/Agents: obere Retry-Schleife des embedded Pi-Runners mit höherem, last-gehobeltem dynamischen Limit (32-160 Versuche) begrenzen und expliziten `retry_limit`-Fehlerpayload zurückgeben, wenn Wiederholungen nie konvergieren, um unendliche interne Retry-Zyklen zu verhindern (`GHSA-76m6-pj3w-v7mf`). - Telegram: Duplikate von Bot-Token-Besitz bei Start/Status erkennen, sekundäre Konten als nicht konfiguriert markieren und Duplicate-Account-Startup-blockieren, bevor Polling beginnt, um endlose `getUpdates`-Konflikt-Schleifen zu vermeiden. - Agents/Tool-Images: Quelldateinamen in `agents/tool-images` Resize-Logs aufnehmen, damit Komprimierungs-Ereignisse auf spezifische Dateien zurückverfolgt werden können. - Providers/OAuth: Qwen- und Chutes-Refresh-Behandlung härten durch Validierung von Refresh-Response-Ablaufwerten und Beibehaltung vorheriger Refresh-Tokens, wenn Anbieter leere Felder zurückgeben, mit Regression-Abdeckung für leere Token-Antworten. - Models/Kimi-Coding: fehlendes implizites Provider-Template für `kimi-coding` mit korrekter `anthropic-messages` API-Type und Basis-URL hinzufügen, um 403-Fehler bei Verwendung von Kimi for Coding zu beheben. (#22409) - Auto-reply/Tools: `senderIsOwner` durch embedded Queued/Followup Runner-Parameter weiterleiten, damit Owner-only-Tools für autorisierte Sender verfügbar bleiben. (#22296) Danke an @hcoj. - Discord: Modell-Auswahlrichter-Back-Navigation wiederherstellen, wenn Provider fehlt, und den Discord-Picker-Flow dokumentieren. (#21458) Danke an @pejmanjohn und @thewilloftheshadow. - Memory/QMD: pro-Agent memorySearch.enabled=false während Gateway QMD-Startup berücksichtigen, Multi-Collection QMD-Suchen in je-Collection-Anfragen aufteilen (`search`/`vsearch`/`query`), bevorzugt Sammlungs-Hinweis-Dokumentenauflösung, Neustart-Updates bei transienten Locks/Timeouts erneut versuchen, `qmd embed` im BM25-Only Suchmodus überspringen und Embedding-Läufe global mit Failure-Backoff serialisieren, um CPU-Stürme bei Multi-Agenten-Hosts zu verhindern. (#20581, #21590, #20513, #20001, #21266, #21583, #20346, #19493) Danke an @danielrevivo, @zanderkrause, @sunyan034-cmd, @tilleulenspiegel, @dae-oss, @adamlongcreativellc, @jonathanadams96, und @kiliansitel. - Memory/Builtin: automatische Sync-Races mit Manager

Claude Codev2.1.50Neu

v2.1.50

20. Februar 2026

## What's changed - Added support for `startupTimeout` configuration for LSP servers - Added `WorktreeCreate` and `WorktreeRemove` hook events, enabling custom VCS setup and teardown when agent worktree isolation creates or removes worktrees. - Fixed a bug where resumed sessions could be invisible when the working directory involved symlinks, because the session storage path was resolved at different times during startup. Also fixed session data loss on SSH disconnect by flushing session data before hooks and analytics in the graceful shutdown sequence. - Linux: Fixed native modules not loading on systems with glibc older than 2.30 (e.g., RHEL 8) - Fixed memory leak in agent teams where completed teammate tasks were never garbage collected from session state - Fixed `CLAUDE_CODE_SIMPLE` to fully strip down skills, session memory, custom agents, and CLAUDE.md token counting - Fixed `/mcp reconnect` freezing the CLI when given a server name that doesn't exist - Fixed memory leak where completed task state objects were never removed from AppState - Added support for `isolation: worktree` in agent definitions, allowing agents to declaratively run in isolated git worktrees. - `CLAUDE_CODE_SIMPLE` mode now also disables MCP tools, attachments, hooks, and CLAUDE.md file loading for a fully minimal experience. - Fixed bug where MCP tools were not discovered when tool search is enabled and a prompt is passed in as a launch argument - Improved memory usage during long sessions by clearing internal caches after compaction - Added `claude agents` CLI command to list all configured agents - Improved memory usage during long sessions by clearing large tool results after they have been processed - Fixed a memory leak where LSP diagnostic data was never cleaned up after delivery, causing unbounded memory growth in long sessions - Fixed a memory leak where completed task output was not freed from memory, reducing memory usage in long sessions with many tasks - Improved startup performance for headless mode (`-p` flag) by deferring Yoga WASM and UI component imports - Fixed prompt suggestion cache regression that reduced cache hit rates - Fixed unbounded memory growth in long sessions by capping file history snapshots - Added `CLAUDE_CODE_DISABLE_1M_CONTEXT` environment variable to disable 1M context window support - Opus 4.6 (fast mode) now includes the full 1M context window - VSCode: Added `/extra-usage` command support in VS Code sessions - Fixed memory leak where TaskOutput retained recent lines after cleanup - Fixed memory leak in CircularBuffer where cleared items were retained in the backing array - Fixed memory leak in shell command execution where ChildProcess and AbortController references were retained after cleanup

OpenCodev1.2.10Neu

v1.2.10

20. Februar 2026

## Desktop - Don't spawn sidecar if default is localhost server ## SDK - Build SDK to dist/ instead of dist/src **Thank you to 1 community contributor:** - @rmk40: - docs: clarify tool name collision precedence (#14313)

OpenCodev1.2.9Neu

v1.2.9

20. Februar 2026

## Core - Add missing id, sessionID, and messageID to MCP tool attachments (@NatChung) - Remove unnecessary deep clones from session loop and LLM stream - Remove User-Agent header assertion from LLM test to fix failing test ## TUI - Use structuredClone instead of remeda's clone for better performance and native support (@mhart) ## Desktop - Restore settings header mask **Thank you to 2 community contributors:** - @mhart: - Use structuredClone instead of remeda's clone (#14351) - @NatChung: - fix: add missing id/sessionID/messageID to MCP tool attachments (#14345)

Claude Codev2.1.49Neu

v2.1.49

19. Februar 2026

## What's changed - Fixed Ctrl+C and ESC being silently ignored when background agents are running and the main thread is idle. Pressing twice within 3 seconds now kills all background agents. - Fixed prompt suggestion cache regression that reduced cache hit rates. - Fixed `plugin enable` and `plugin disable` to auto-detect the correct scope when `--scope` is not specified, instead of always defaulting to user scope - Simple mode (`CLAUDE_CODE_SIMPLE`) now includes the file edit tool in addition to the Bash tool, allowing direct file editing in simple mode. - Permission suggestions are now populated when safety checks trigger an ask response, enabling SDK consumers to display permission options - Sonnet 4.5 with 1M context is being removed from the Max plan in favor of our frontier Sonnet 4.6 model, which now has 1M context. Please switch in /model. - Fixed verbose mode not updating thinking block display when toggled via `/config` — memo comparators now correctly detect verbose changes - Fixed unbounded WASM memory growth during long sessions by periodically resetting the tree-sitter parser - Fixed potential rendering issues caused by stale yoga layout references - Improved performance in non-interactive mode (`-p`) by skipping unnecessary API calls during startup - Improved performance by caching authentication failures for HTTP and SSE MCP servers, avoiding repeated connection attempts to servers requiring auth - Fixed unbounded memory growth during long-running sessions caused by Yoga WASM linear memory never shrinking - SDK model info now includes `supportsEffort`, `supportedEffortLevels`, and `supportsAdaptiveThinking` fields so consumers can discover model capabilities. - Added `ConfigChange` hook event that fires when configuration files change during a session, enabling enterprise security auditing and optional blocking of settings changes. - Improved startup performance by caching MCP auth failures to avoid redundant connection attempts - Improved startup performance by reducing HTTP calls for analytics token counting - Improved startup performance by batching MCP tool token counting into a single API call - Fixed `disableAllHooks` setting to respect managed settings hierarchy — non-managed settings can no longer disable managed hooks set by policy (#26637) - Fixed `--resume` session picker showing raw XML tags for sessions that start with commands like `/clear`. Now correctly falls through to the session ID fallback. - Improved permission prompts for path safety and working directory blocks to show the reason for the restriction instead of a bare prompt with no context

OpenCodev1.2.8Neu

v1.2.8

19. Februar 2026

## Core - Support adaptive thinking for Claude Sonnet 4.6 (@tctev) ## TUI - Add custom tool and MCP call responses that are visible and collapsible (@yanosh-k) ## Desktop - Black screen on launch with sidecar server fixed - Clear todos on abort **Thank you to 2 community contributors:** - @yanosh-k: - feat(tui): add custom tool and mcp call responses visible and collapsable (#10649) - @tctev: - feat(opencode): support adaptive thinking for claude sonnet 4.6 (#14283)

OpenCodev1.2.7Neu

v1.2.7

19. Februar 2026

## Core - Fixed terminal rendering and interaction issues in the application - Normalize file status paths relative to instance directory (@shantur) - Migrate from Bun.Glob to npm glob package - Bump AI SDK packages for Google, Google Vertex, Anthropic, Bedrock, and provider utils - Add support for medium reasoning with Gemini 3.1 - Remove use of Bun.file - Text files misclassified as binary - Fetch default server at top level in desktop application - Terminal rework in the app - Bake in the AWS and Google authentication packages - Token substitution in OPENCODE_CONFIG_CONTENT now works correctly (@ariane-emory) - Revert migration from Bun.file() to Filesystem module - Migrate project.ts from Bun.file() to Filesystem/stat modules - Migrate read tool from Bun.file() to Filesystem module - Migrate write tool from Bun.file() to Filesystem module - Migrate Edit tool from Bun.file() to Filesystem module - Migrate remaining tool files from Bun.file() to Filesystem/stat modules - Migrate storage.ts from Bun.file()/Bun.write() to Filesystem module - Migrate src/storage/json-migration.ts from Bun.file() to Filesystem module - Migrate MCP auth module from Bun file APIs to Filesystem module - Migrate storage database from Bun.file() to statSync for file existence checks - Migrate session prompt module from Bun.file() to Filesystem/stat modules - Fix crash in `opencode run` and show errored tool calls in output - Migrate skill discovery to use Filesystem module instead of Bun file APIs - Migrate session instruction handling from Bun.file() to Filesystem module - Migrate provider.ts from Bun.file() to Filesystem module - Migrate shell.ts from Bun.file() to statSync for improved file system operations - Migrate log utility from Bun.file() to Node.js fs module for better compatibility - Migrate models.ts from Bun.file()/Bun.write() to Filesystem module - Use HashiCorp releases API for installing terraform-ls (@edubxb) - Migrate LSP server from Bun.file()/Bun.write() to Filesystem module - Migrate session command from Bun.file() to statSync for improved file system operations - Migrate agent.ts from Bun.file() to Filesystem module - Migrate auth module from Bun.file()/Bun.write() to Filesystem module - Pass sessionID and callID to shell.env hook input (@tesdal) - Fix terminal cross-talk issue in the application - Update SST version - Migrate src/global/index.ts from Bun.file() to Filesystem module - Emit PROMPT_TOO_LARGE error when GitHub context overflows (@elithrar) - Migrate src/bun/index.ts from Bun.file()/Bun.write() to Filesystem module - Migrate config/markdown.ts from Bun.file() to Filesystem module - Migrate file/index.ts from Bun.file() to Filesystem module - Migrate format/formatter.ts from Bun.file() to Filesystem module - Allow readJson to be called without explicit type parameter - Migrate file/ripgrep.ts from Bun APIs to Filesystem module - Migrate index.ts from Bun.file() to Filesystem module - Add Julia language server support (@zarly) - Bump GitLab AI provider to 3.6.0 to add Sonnet 4.6 support (@vglafirov) - Add centralized filesystem module for Bun.file migration - Fix Clojure syntax highlighting (@finalfantasia) - Ensure explore subagent prompts for external directory permission instead of auto-denying - Don't autoload kilo - Add Kilo as a native provider (@Nomadcxx) - Simplify redundant ternary in updateMessage (@yikayiyo) - Ensure Read tool uses fs/promises for all file system operations - Make read tool more memory efficient - Surface plugin auth providers in the login picker (@anoldguy) - Invalidate OAuth credentials when OAuth provider indicates they are invalid (@GreenStage) - Don't fetch models.dev on completion (@gigamonster256) - Recover state after SSE reconnect and harden SSE streams - Keep message part order stable when files resolve asynchronously - Drop IDs from attachments in tools and assign them in prompt.ts instead ## TUI - Improve GitHub action branch detection and handle 422 errors (@elithrar) - Ensure onExit callback fires after terminal output is written - Migrate TUI thread module from Bun.file() to Filesystem module - Migrate agent command from Bun.file()/Bun.write() to Filesystem module - Migrate import command from Bun.file() to Filesystem module - Update pasteImage to only increment count when the previous attachment is an image (@OpeOginni) - Migrate editor.ts from Bun.file()/Bun.write() to Filesystem module - Migrate clipboard.ts from Bun.file() to Filesystem module - Migrate CLI run command from Bun.file() to Filesystem/stat modules - Session list --max-count parameter now correctly limits the number of sessions displayed (@mharris717) - Exit cleanly without hanging after session ends - Style scrollbox for permission and sidebar (@akronb) - Increase button heights and improve permission prompt layout alignment - Display new session banner with logo and project details in TUI ## Desktop - Update Japanese translations for WSL integration (@taroj1205) - Made localhost URLs work correctly in isLocal function - Navigate to last session when navigating to a project - Fix typecheck errors in app - Deduplicate allServers list in app - Adjust session turn horizontal padding - Tighten prompt dock padding in app - Fixed sidecar spawning a window on Windows - Delay prompt mode toggle tooltip - Shorten prompt mode toggle tooltips in the app - Reduce review panel padding - Tweak search button style in UI - Expanded color state on titlebar buttons - Tweak hover and active styles for title bar buttons - Share button now has a border - Adjust file tree background color - Handle sidecar key in projectsKey for desktop projects - Fixed desktop app incorrectly identifying local servers - Refactor server management backend - Use group-hover for file tree icon color swap at all nesting levels - Simplify mode toggle icon styling in TUI - Clean up desktop implementation - Temporarily disable WSL support in desktop application - Use radio group in prompt input - Simplify prompt mode toggle icon colors via CSS and tighten message timeline padding in TUI - Fix prompt input quirks in app - Terminal disconnect and resync functionality fixed - Replicate tauri-plugin-shell logic in desktop application - Improve modified file visibility and button spacing in TUI - Show monochrome file icons by default in tree view, revealing colors on hover to reduce visual clutter - Fix share button text styling to use consistent 12px regular font weight - Add warning icon to permission requests for better visibility - Extract dock prompt shell component - UI no longer flashes when switching tabs (@neriousy) - Avoid sidecar health-check timeout on shell startup in desktop app (@ysm-dev) - Increase prompt mode toggle height for better clickability - Add more end-to-end tests for desktop application (@neriousy) - Update magnifying-glass icon in UI - Tighten titlebar action padding - Refine titlebar search and open padding - Center titlebar search and soften keybind styling - Align titlebar search text size - Match titlebar active background to hover - Use weak borders in titlebar actions - Reduce titlebar right padding - Keep file tree toggle visible - Adjust icon button spacing in UI - Session timeline and turn handling reworked in app - Keep Escape handling local to prompt input on macOS desktop (@itskritix) - Hide server CLI window on Windows ## SDK - Fix nested exports transformation in SDK publish script **Thank you to 25 community contributors:** - @itskritix: - fix(app): keep Escape handling local to prompt input on macOS desktop (#13963) - @vynnlee: - docs(ko): improve Korean translation accuracy and clarity in Zen docs (#13951) - @chenmijiang: - docs: improve zh-cn and zh-tw documentation translations (#13942) - @hmu332233: - fix(docs): correct reversed meaning in Korean plugins logging section (#13945) - @neriousy: - feat(desktop): more e2e tests (#13975) - fix(app): ui flashing when switching tabs (#13978) - @ysm-dev: - fix(desktop): avoid sidecar health-check timeout on shell startup (#13925) - @alexcarpenter: - fix: Homepage video section layout shift (#13987) - @gigamonster256: - fix: don't fetch models.dev on completion (#13997) - @GreenStage: - fix: Invalidate oauth credentials when oauth provider says so (#14007) - @anoldguy: - feat: surface plugin auth providers in the login picker (#13921) - @akronb: - fix(tui): style scrollbox for permission and sidebar (#12752) - @yikayiyo: - refactor: simplify redundant ternary in updateMessage (#13954) - @Nomadcxx: - feat: add Kilo as a native provider (#13765) - @finalfantasia: - fix(opencode): fix Clojure syntax highlighting (#13453) - @mharris717: - fix(cli): session list --max-count not honored, shows too few sessions (#14162) - @vglafirov: - feat: GitLab Duo - bump gitlab-ai-provider to 3.6.0 (adds Sonnet 4.6) (#14115) - @zarly: - feat: add Julia language server support (#14129) - @elithrar: - fix(github): emit PROMPT_TOO_LARGE error on context overflow (#14166) - fix(github): action branch detection and 422 handling (#14322) - @OpeOginni: - fix(opencode): update pasteImage to only increment count when the previous attachment is an image too (#14173) - @tesdal: - feat(plugin): pass sessionID and callID to shell.env hook input (#13662) - @edubxb: - fix(lsp): use HashiCorp releases API for installing terraform-ls (#14200) - @ariane-emory: - fix: token substitution in OPENCODE_CONFIG_CONTENT (alternate take) (#14047) - @Seungjun0906: - docs(ko): improve wording in ecosystem, enterprise, formatters, and github docs (#14220) - @shantur: - fix(core): normalize file.status paths relative to instance dir (#14207) - @taroj1205: - feat(i18n): update Japanese translations to WSL integration (#13160)

Gemini CLIv0.29.5Neu

Release v0.29.5

19. Februar 2026

**Vollständiges Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.4...v0.29.5

Gemini CLIv0.29.4Neu

Release v0.29.4

19. Februar 2026

**Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.3...v0.29.4

OpenClawv2026.2.19Neu

openclaw 2026.2.19

19. Februar 2026

### Änderungen - iOS/Watch: ein Apple-Watch- companion MVP mit Watch-Posteingang-UI, Watch-Benachrichtigungs-Relay-Handling und Gateway-Bachstellenflächen für Watch-Status/Senden-Flows hinzufügen. (#20054) Danke @mbelinky. - iOS/Gateway: disconnectete iOS-Knoten via APNs aufwachen vor `nodes.invoke` und Gateway-Sitzungen beim Still-Push-Wake automatisch wieder verbinden, um Invoke-Fehler zu reduzieren, wenn die App im Hintergrund ist. (#20332) Danke @mbelinky. - Gateway/CLI: Paired-Device-Hygiene-Flows mit `device.pair.remove` hinzufügen, plus `openclaw devices remove` und geschützte `openclaw devices clear --yes [--pending]`-Befehle zum Entfernen gepaarter Einträge und optional Ablehnung ausstehender Anfragen. (#20057) Danke @mbelinky. - iOS/APNs: Push-Registrierung und Signierungs-Konfiguration für Node-Zustellung hinzufügen. (#20308) Danke @mbelinky. - Gateway/APNs: Push-Test-Pipeline für APNs-Zustellungsprüfung in Gateway-Flows hinzufügen. (#20307) Danke @mbelinky. - Security/Audit: Befunde `gateway.http.no_auth` hinzufügen, wenn `gateway.auth.mode="none"` Gateway-HTTP-APIs erreichbar lässt, mit Loopback-Warnung und Remote-Exposure-Kritikalität, plus Regression-Abdeckung und Dokumentations-Updates. - Skills: Coding-Agent-Fähigkeitsleitfaden härten, indem Shell-Command-Beispiele entfernt werden, die untrusted Issue-Text direkt in Befehlsstrings interpolieren. - Dev-Tooling: Ausrichtung des lokalen/CI-Formatverhaltens von `oxfmt`. (#12579) Danke @vincentkoc. ### Fixes - Agents/Streaming: Partielle Streaming-Sitzung während Reasoning-Streams aktiv halten, native `thinking_*`-Streams-Ereignisse konsistent behandeln, gemischte Reasoning-End-Signale deduplizieren und veraltete mutierende Tool-Fehler nach Erfolg des Gleichziel-Neuversuchs bereinigen. (#20635) Danke @obviyus. - iOS/Screen: Lebenszyklus-Eigentümerschaft von `WKWebView` in den `ScreenWebView`-Koordinator verschieben und expliziten Attach/Detach-Flow verwenden, um Risiko von Gesten-/Lebenszyklus-Abstürzen zu reduzieren (`__NSArrayM insertObject:atIndex:`-Pfad) während Bildschirm-Tab-Updates. (#20366) Danke @ngutman. - iOS/Onboarding: Pairing-Status-Flackern beim Auto-Resume verhindern, stabile Zustandsübergänge für Resume beibehalten. (#20310) Danke @mbelinky. - iOS/Onboarding: Pairing- und Reconnect-Verhalten stabilisieren durch Zurücksetzen veralteter Pairing-Anforderungszustände bei manuellem Retry, Trennung beider Operator- und Node-Gateways bei Operator-Fehlern und Vermeidung doppelter Pairing-Schleifen durch Attachment der Operator-Transport-Identität. (#20056) Danke @mbelinky. - iOS/Signing: Lokale Auto-Auswahl Overrides für Signing-Team während Projekterzeugung wiederherstellen, indem `.local-signing.xcconfig` in aktives Signing-Config eingespeist wird und `OPENCLAW_DEVELOPMENT_TEAM` im Local-Signing-Setup ausgegeben wird. (#19993) Danke @ngutman. - Telegram: Nachrichten-ähnliche eingehende Verarbeitung vereinheitlichen, sodass `message` und `channel_post` denselben Dedupe/Access/Media-Pipeline teilen und behaviorally konsistent bleiben. (#20591) Danke @obviyus. - Telegram/Agents: Ausführung/bash Tool-Fehler-Warnungen hinter VR-Modus verstecken, damit Standard-Telegram-Antworten sauber bleiben, während verbose-Sitzungen Diagnostics aufzeigen. (#20560) Danke @obviyus. - Telegram/Cron/Heartbeat: Explizite Telegram-Themenziele in Cron- und Heartbeat-Zustellung berücksichtigen (`<chatId>:topic:<threadId>`), damit geplanter Versand im konfigurierten Thema landet statt im zuletzt aktiven Thread. (#19367) Danke @Lukavyi. - Gateway/Daemon: `TMPDIR` in installierte Service-Umgebungen weiterleiten, damit macOS LaunchAgent-Gateway-Läufe zuverlässig SQLite-Temp-/Journal-Dateien öffnen können, statt mit `SQLITE_CANTOPEN` zu scheitern. (#20512) Danke @Clawborn. - Agents/Billing: aktives Modell, das einen Abrechnungsfehler verursacht hat, in nutzerorientierte Abrechnungsnachrichten aufnehmen (z. B. `OpenAI (gpt-5.3)`) über Payload, Failover und Lifecycle-Fehlerpfade hinweg, damit Nutzer genau erkennen können, welcher Schlüssel Guthaben benötigt. (#20510) Danke @echoVic. - Gateway/TUI: `agents.defaults.blockStreamingDefault` für `chat.send` berücksichtigen, indem die harte Überschreibung der Block-Streaming-Deaktivierung entfernt wird, damit Antworten im konfigurierten Block-Modus geliefert werden. (#19693) Danke @neipor. - UI/Sessions: canonicaler Main-Session-Key-Alias in Chat-UI-Flows akzeptieren, damit Main-Session-Routing konsistent bleibt. (#20311) Danke @mbelinky. - OpenClawKit/Protocol: JSON-Boolean-Literals (`true`/`false`) beim Bridging durch `AnyCodable` beibehalten, damit Apple-Client-RPC-Parameter Booleans nicht erneut als `1`/`0` kodieren. Danke @mbelinky. - Commands/Doctor: Embedding-Provider-Warnungen überspringen, wenn `memory.backend` `qmd` ist, da QMD Embeddings intern verwaltet und `memorySearch`-Provider nicht benötigt. (#17263) Danke @miloudbelarebia. - Canvas/A2UI: Bündel-Asset-Auflösung und Empty-State-Behandlung verbessern, damit UI-Fallbacks zuverlässig gerendert werden. (#20312) Danke @mbelinky. - Commands/Doctor: Verhindern, ungültige Configs mit neuen Defaults von `gateway.auth.token` während der Reparatur umzuschreiben, und nur schreiben, wenn reale Config-Änderungen erkannt werden, um versehentliche Token-Duplikation und Backup-Churn zu vermeiden. - Gateway/Auth: Standardmäßig unresolved Gateway-Auth auf Token-Modus setzen mit Start-up-Auto-Generierung/-Persistierung von `gateway.auth.token`, während explizites `gateway.auth.mode: "none"` für beabsichtigte Open-Loopback-Setups erlaubt bleibt. (#20686) danke @gumadeiras. - Channels/Matrix: Erkennung von Erwähnungen für `formatted_body` Matrix-zu-Links korrigieren, indem Matrix.to-Erwähnungsformate konsistent behandelt werden. (#16941) Danke @zerone0x. - Heartbeat/Cron: Intervall-Herzschläge überspringen, wenn `HEARTBEAT.md` fehlt oder leer ist und keine getaggten Cron-Ereignisse in der Warteschlange sind, während Fallback für getaggte Erinnerungen beibehalten wird. (#20461) danke @vikpos. - Browser/Relay: Bereits laufenden Extension-Relais erneut verwenden, wenn der Relay-Port von einem anderen OpenClaw-Prozess belegt ist, jedoch bei Port-Konflikten außerhalb des Relays fehlschlagen, um unverbundene Listener nicht zu verdecken. (#20035) Danke @mbelinky. - Scripts: clawdock-Hilfskommando-Unterstützung aktualisieren, um `docker-compose.extra.yml` dort zu verwenden, wo verfügbar. (#17094) Danke @zerone0x. - Lobster/Config: Lobster-Executable-Pfad-Überschreibungen entfernen (`lobsterPath`), PATH-basierte Ausführung erzwingen und fokussierte Windows-Wrapper-Lösungen-Tests hinzufügen, um shell-freies Verhalten stabil zu halten. - Gateway/WebChat: Blockieren von `sessions.patch` und `sessions.delete` für WebChat-Clients, damit Session-Store-Manipulationen auf Nicht-WebChat-Operator-Flows beschränkt bleiben. Danke @allsmog für den Report. - Gateway: Launchctl GUI-Domänen-Bootstrap-Fehler auf macOS klarstellen. (#13795) Danke @vincentkoc. - Lobster/CI: flaky Test Windows cmd-Shim-Skripte lösen. (#20833) Danke @vincentkoc. - Browser/Relay: Gateway-Token-Auth auf `/extension` und `/cdp` erzwingen und Chrome-Extension-Setup auf die Verwendung eines einzigen `gateway.auth.token`-Inputs für Relay-Authentifizierung ausrichten. Danke @tdjackey für den Bericht. - Gateway/Hooks: Boot-Checks pro konfigurierbarem Agenten-Umfang durchführen, einschließlich Auflösung des Session-Keys pro Agent, Regression-Abdeckung von Startup-Hooks und Logging von Boot

Gemini CLIv0.29.3Neu

Release v0.29.3

19. Februar 2026

## Änderungen * fix(patch): cherry-pick c43500c to release/v0.29.2-pr-19502 to patch version v0.29.2 and create version 0.29.3 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/19522 **Voller Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.2...v0.29.3

Gemini CLIv0.29.2

Release v0.29.2

19. Februar 2026

## Änderungen * fix(patch): cherry-pick 261788c to release/v0.29.1-pr-19453 to patch version v0.29.1 and create version 0.29.2 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/19491 **Vollständiger Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.1...v0.29.2

Gemini CLIv0.29.1

Release v0.29.1

18. Februar 2026

## Was geändert wurde * fix(patch): cherry-pick 858918f to release/v0.29.0-pr-19460 to patch version v0.29.0 and create version 0.29.1 by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/19480 **Vollständiges Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.29.0...v0.29.1

Claude Codev2.1.47

v2.1.47

18. Februar 2026

## What's changed - Fixed FileWriteTool line counting to preserve intentional trailing blank lines instead of stripping them with `trimEnd()`. - Fixed Windows terminal rendering bugs caused by `os.EOL` (`\r\n`) in display code — line counts now show correct values instead of always showing 1 on Windows. - Improved VS Code plan preview: auto-updates as Claude iterates, enables commenting only when the plan is ready for review, and keeps the preview open when rejecting so Claude can revise. - Fixed a bug where bold and colored text in markdown output could shift to the wrong characters on Windows due to `\r\n` line endings. - Fixed compaction failing when conversation contains many PDF documents by stripping document blocks alongside images before sending to the compaction API (anthropics/claude-code#26188) - Improved memory usage in long-running sessions by releasing API stream buffers, agent context, and skill state after use - Improved startup performance by deferring SessionStart hook execution, reducing time-to-interactive by ~500ms. - Fixed an issue where bash tool output was silently discarded on Windows when using MSYS2 or Cygwin shells. - Improved performance of `@` file mentions - file suggestions now appear faster by pre-warming the index on startup and using session-based caching with background refresh. - Improved memory usage by trimming agent task message history after tasks complete - Improved memory usage during long agent sessions by eliminating O(n²) message accumulation in progress updates - Fixed the bash permission classifier to validate that returned match descriptions correspond to actual input rules, preventing hallucinated descriptions from incorrectly granting permissions - Fixed user-defined agents only loading one file on NFS/FUSE filesystems that report zero inodes (anthropics/claude-code#26044) - Fixed plugin agent skills silently failing to load when referenced by bare name instead of fully-qualified plugin name (anthropics/claude-code#25834) - Search patterns in collapsed tool results are now displayed in quotes for clarity - Windows: Fixed CWD tracking temp files never being cleaned up, causing them to accumulate indefinitely (anthropics/claude-code#17600) - Use `ctrl+f` to kill all background agents instead of double-pressing ESC. Background agents now continue running when you press ESC to cancel the main thread, giving you more control over agent lifecycle. - Fixed API 400 errors ("thinking blocks cannot be modified") that occurred in sessions with concurrent agents, caused by interleaved streaming content blocks preventing proper message merging. - Simplified teammate navigation to use only Shift+Down (with wrapping) instead of both Shift+Up and Shift+Down. - Fixed an issue where a single file write/edit error would abort all other parallel file write/edit operations. Independent file mutations now complete even when a sibling fails. - Added `last_assistant_message` field to Stop and SubagentStop hook inputs, providing the final assistant response text so hooks can access it without parsing transcript files. - Fixed custom session titles set via `/rename` being lost after resuming a conversation (anthropics/claude-code#23610) - Fixed collapsed read/search hint text overflowing on narrow terminals by truncating from the start. - Fixed an issue where bash commands with backslash-newline continuation lines (e.g., long commands split across multiple lines with `\`) would produce spurious empty arguments, potentially breaking command execution. - Fixed built-in slash commands (`/help`, `/model`, `/compact`, etc.) being hidden from the autocomplete dropdown when many user skills are installed (anthropics/claude-code#22020) - Fixed MCP servers not appearing in the MCP Management Dialog after deferred loading - Fixed session name persisting in status bar after `/clear` command (anthropics/claude-code#26082) - Fixed crash when a skill's `name` or `description` in SKILL.md frontmatter is a bare number (e.g., `name: 3000`) — the value is now properly coerced to a string (anthropics/claude-code#25837) - Fixed /resume silently dropping sessions when the first message exceeds 16KB or uses array-format content (anthropics/claude-code#25721) - Added `chat:newline` keybinding action for configurable multi-line input (anthropics/claude-code#26075) - Added `added_dirs` to the statusline JSON `workspace` section, exposing directories added via `/add-dir` to external scripts (anthropics/claude-code#26096) - Fixed `claude doctor` misclassifying mise and asdf-managed installations as native installs (anthropics/claude-code#26033) - Fixed zsh heredoc failing with "read-only file system" error in sandboxed commands (anthropics/claude-code#25990) - Fixed agent progress indicator showing inflated tool use count (anthropics/claude-code#26023) - Fixed image pasting not working on WSL2 systems where Windows copies images as BMP format (anthropics/claude-code#25935) - Fixed background agent results returning raw transcript data instead of the agent's final answer (anthropics/claude-code#26012) - Fixed Warp terminal incorrectly prompting for Shift+Enter setup when it supports it natively (anthropics/claude-code#25957) - Fixed CJK wide characters causing misaligned timestamps and layout elements in the TUI (anthropics/claude-code#26084) - Fixed custom agent `model` field in `.claude/agents/*.md` being ignored when spawning team teammates (anthropics/claude-code#26064) - Fixed plan mode being lost after context compaction, causing the model to switch from planning to implementation mode (anthropics/claude-code#26061) - Fixed `alwaysThinkingEnabled: true` in settings.json not enabling thinking mode on Bedrock and Vertex providers (anthropics/claude-code#26074) - Fixed `tool_decision` OTel telemetry event not being emitted in headless/SDK mode (anthropics/claude-code#26059) - Fixed session name being lost after context compaction — renamed sessions now preserve their custom title through compaction (anthropics/claude-code#26121) - Increased initial session count in resume picker from 10 to 50 for faster session discovery (anthropics/claude-code#26123) - Windows: fixed worktree session matching when drive letter casing differs (anthropics/claude-code#26123) - Fixed `/resume <session-id>` failing to find sessions whose first message exceeds 16KB (anthropics/claude-code#25920) - Fixed "Always allow" on multiline bash commands creating invalid permission patterns that corrupt settings (anthropics/claude-code#25909) - Fixed React crash (error #31) when a skill's `argument-hint` in SKILL.md frontmatter uses YAML sequence syntax (e.g., `[topic: foo | bar]`) — the value is now properly coerced to a string (anthropics/claude-code#25826) - Fixed crash when using `/fork` on sessions that used web search — null entries in search results from transcript deserialization are now handled gracefully (anthropics/claude-code#25811) - Fixed read-only git commands triggering FSEvents file watcher loops on macOS by adding --no-optional-locks flag (anthropics/claude-code#25750) - Fixed custom agents and skills not being discovered when running from a git worktree — project-level `.claude/agents/` and `.claude/skills/` from the main repository are now included (anthropics/claude-code#25816) - Fixed non-interactive subcommands like `claude doctor` and `claude plugin validate` being blocked inside nested Claude sessions (anthropics/claude-code#25803) - Windows: Fixed the same CLAUDE.md file being loaded twice when drive letter casing differs between paths (anthropics/claude-code#25756) - Fixed inline code spans in markdown being incorrectly parsed as bash commands (anthropics/claude-code#25792) - Fixed teammate spinners not respecting custom spinnerVerbs from settings (anthropics/claude-code#25748) - Fixed shell commands permanently failing after a command deletes its own working directory (anthropics/claude-code#26136) - Fixed hooks (PreToolUse, PostToolUse) silently failing to execute on Windows by using Git Bash instead of cmd.exe (anthropics/claude-code#25981) - Fixed LSP `findReferences` and other location-based operations returning results from gitignored files (e.g., `node_modules/`, `venv/`) (anthropics/claude-code#26051) - Moved config backup files from home directory root to `~/.claude/backups/` to reduce home directory clutter (anthropics/claude-code#26130) - Fixed sessions with large first prompts (>16KB) disappearing from the /resume list (anthropics/claude-code#26140) - Fixed shell functions with double-underscore prefixes (e.g., `__git_ps1`) not being preserved across shell sessions (anthropics/claude-code#25824) - Fixed spinner showing "0 tokens" counter before any tokens have been received (anthropics/claude-code#26105) - VSCode: Fixed conversation messages appearing dimmed while the AskUserQuestion dialog is open (anthropics/claude-code#26078) - Fixed background tasks failing in git worktrees due to remote URL resolution reading from worktree-specific gitdir instead of the main repository config (anthropics/claude-code#26065) - Fixed Right Alt key leaving visible `[25~` escape sequence residue in the input field on Windows/Git Bash terminals (anthropics/claude-code#25943) - The `/rename` command now updates the terminal tab title by default (anthropics/claude-code#25789) - Fixed Edit tool silently corrupting Unicode curly quotes (\u201c\u201d \u2018\u2019) by replacing them with straight quotes when making edits (anthropics/claude-code#26141) - Fixed OSC 8 hyperlinks only being clickable on the first line when link text wraps across multiple terminal lines.

Codex CLIvrust-v0.104.0

0.104.0

18. Februar 2026

## New Features - Added `WS_PROXY`/`WSS_PROXY` environment support (including lowercase variants) for websocket proxying in the network proxy. (#11784) - App-server v2 now emits notifications when threads are archived or unarchived, enabling clients to react without polling. (#12030) - Protocol/core now carry distinct approval IDs for command approvals to support multiple approvals within a single shell command execution flow. (#12051) ## Bug Fixes - `Ctrl+C`/`Ctrl+D` now cleanly exits the cwd-change prompt during resume/fork flows instead of implicitly selecting an option. (#12040) - Reduced false-positive safety-check downgrade behavior by relying on the response header model (and websocket top-level events) rather than the response body model slug. (#12061) ## Documentation - Updated docs and schemas to cover websocket proxy configuration, new thread archive/unarchive notifications, and the command approval ID plumbing. (#11784, #12030, #12051) ## Chores - Made the Rust release workflow resilient to `npm publish` attempts for an already-published version. (#12044) - Standardized remote compaction test mocking and refreshed related snapshots to align with the default production-shaped behavior. (#12050) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.103.0...rust-v0.104.0 - #11784 feat(network-proxy): add websocket proxy env support @viyatb-oai - #12044 don't fail if an npm publish attempt is for an existing version. @iceweasel-oai - #12040 tui: exit session on Ctrl+C in cwd change prompt @charley-oai - #12030 app-server: Emit thread archive/unarchive notifications @euroelessar - #12061 Chore: remove response model check and rely on header model for downgrade @shijie-oai - #12051 feat(core): plumb distinct approval ids for command approvals @owenlin0 - #12050 Unify remote compaction snapshot mocks around default endpoint behavior @charley-oai

OpenClawv2026.2.17

openclaw 2026.2.17

18. Februar 2026

### Änderungen - Agents/Anthropic: opt-in 1M Context Beta-Header-Unterstützung für Opus/Sonnet über Modell `params.context1m: true` hinzufügen (entspricht `anthropic-beta: context-1m-2025-08-07`). - Agents/Models: Anthropic Sonnet 4.6 (`anthropic/claude-sonnet-4-6`) über Aliases/Standardwerte hinweg unterstützen mit Forward-Compatibility-Fallback, wenn Upstream-Kataloge noch nur Sonnet 4.5 expose. - Commands/Subagents: `/subagents spawn` für deterministische Subagenten-Aktivierung aus Chat-Befehlen hinzufügen. (#18218) Danke an @JoshuaLelon. - Agents/Subagents: Eine akzeptierte Antwort-Hinweis für `sessions_spawn` hinzufügen, der erklärt, dass Polling-Subagenten für One-off-Aufrufe deaktiviert sind. Danke an @tyler6204. - Agents/Subagents: Spawning-Subagenten-Task-Nachrichten mit Kontext kennzeichnen, um Quelleninformation in nachgelagerter Verarbeitung zu bewahren. Danke an @tyler6204. - iOS/Share: eine iOS-Share-Erweiterung hinzufügen, die geteilte URL/Text/Bild-Inhalte direkt an Gateway `agent.request` weiterleitet, mit Delivery-Route-Fallback und optionalen Empfangsbestätigungen. (#19424) Danke an @mbelinky. - iOS/Talk: einen `Background Listening`-Schalter hinzufügen, der Talk Mode aktiv hält, wenn die App im Hintergrund läuft (standardmäßig ausgeschaltet aus Battery-Schutzgründen). Danke an @zeulewan. - iOS/Talk: einen `Voice Directive Hint`-Schalter für Talk-Mode-Eingabeaufforderungen hinzufügen, damit Nutzer Anweisungen zum Sprachwechsel von ElevenLabs deaktivieren können, um Tokens zu sparen, wenn nicht benötigt. (#18250) Danke an @zeulewan. - iOS/Talk: Barge-in-Verhalten härten, indem Unterbrechungen beim Sprechen deaktiviert werden, wenn Output-Route Lautsprecher/Receiver ist, um Fehlunterbrechungen durch lokales TTS-Verbleich zu reduzieren. Danke an @zeulewan. - Slack: native Einzel-Nachrichten-Streaming mit Slack `chat.startStream`/`appendStream`/`stopStream` hinzufügen; Antworten-Threading mit `replyToMode` beibehalten, Streaming standardmäßig aktiviert, und bei Streaming-Fehlern auf normale Delivery zurückfallen. (#9972) Danke an @natedenh. - Slack: konfigurierbare Streaming-Modi für Entwurf-Vorschauen hinzufügen. (#18555) Danke an @Solvely-Colin. - Telegram/Agents: Inline-Button `style`-Unterstützung (`primary|success|danger`) über das Nachrichten-Tool-Schema, Telegram-Aktions-Parsing, Send-Pipeline und Laufzeit-Hinweise integrieren. (#18241) Danke an @obviyus. - Telegram: Benutzer-Message-Reaktionen als System-Events darstellen, konfigurierbarer Bereich `channels.telegram.reactionNotifications`. (#10075) Danke an @Glucksberg. - iMessage: `replyToId` bei ausgehenden Text/Mediatransfers unterstützen und führende `[[reply_to:<id>]]`-Tags normalisieren, damit Antworten auf die beabsichtigte iMessage zielen. Danke an @tyler6204. - Tool Display/Web UI: intent-first Tool-Detail-Ansichten und Exec-Summary hinzufügen. (#18592) Danke an @xdLawless2. - Discord: native `/exec`-Befehlsoptionen (host/security/ask/node) freischalten, sodass Discord-Slash-Befehle Autocomplete und strukturierte Eingaben erhalten. Danke an @thewilloftheshadow. - Discord: wiederverwendbare interaktive Komponenten mit `components.reusable=true` zulassen, sodass Buttons, Selektionen und Formulare mehrfach vor Ablauf verwendet werden können. Danke an @thewilloftheshadow. - Discord: pro-Button `allowedUsers`-Whitelist hinzufügen, um zu beschränken, wer Buttons klicken darf. Danke an @thewilloftheshadow. - Cron/Gateway: pro-Job-Webhook-Lieferung (`delivery.mode = "webhook"`) von Ankündigungszustellung trennen, gültige HTTP(S)-Webhook-URLs erzwingen und ein temporäres Legacy-Fallback `notify + cron.webhook` für gespeicherte Jobs beibehalten. (#17901) Danke an @advaitpaliwal. - Cron/CLI: deterministische standardmäßige Streckung für wiederkehrende Cron-Schedules zur vollen Stunde (inkl. 6-Feld-Sekunden-Cron), Bestehende Jobs auto-migrationsartig auf persistierte `schedule.staggerMs` migrieren und `openclaw cron add/edit --stagger <Dauer>` plus `--exact` Overrides für genaue Timing-Steuerung pro Job hinzufügen. - Cron: pro Lauf modelle Nutzungs-Telemetrie in Cron-Lauf-Logs/Webhooks protokollieren und ein lokales Nutzungsbericht-Skript zum Aggregieren der Token-Nutzung pro Job hinzufügen. (#18172) Danke an @HankAndTheCrew. - Tools/Web: Whitelists-Listen-URLs für `web_search` und `web_fetch` hinzufügen. (#18584) Danke an @smartprogrammer93. - Browser: `extraArgs`-Konfig für benutzerdefinierte Chrome-Startargumente hinzufügen. (#18443) Danke an @JayMishra-source. - Voice Call: inbound Begrüßungs-TTS vorab-cachen für schnellere erste Wiedergabe. (#18447) Danke an @JayMishra-source. - Skills: kompakte Skill-Dateipfade `<location>` im System-Prompt durch Ersetzen von Home-Verzeichnis-Prefixen durch `~` zusammenführen und gezielte Kompaktionstests für Prompt-Serialisierung hinzufügen. (#14776) Danke an @bitfish3. - Skills: Präzision der Skill-Beschreibung-Routing-Grenzen mit expliziter Anleitung „Use when“/„NOT for“ für coding-agent/github/weather verfeinern und PTY/Browser-Fallback-Wortlaut klären. (#14577) Danke an @DylanWoodAkers. - Auto-reply/Prompts: vertrauenswürdige eingehende `message_id` in Konversations-Metadaten-Payloads für nachgelagerte Targeting-Workflows einbeziehen. Danke an @tyler6204. - Auto-reply: `sender_id` in vertrauenswürdigen eingehenden Metadaten einbeziehen, damit Moderations-Workflows den Absender gezielt ansteuern können, ohne auf untrusted Text angewiesen zu sein. (#18303) Danke an @crimeacs. - UI/Sessions: doppelte Anzeige von Session-Prefixen vermeiden (z. B. `Subagent Subagent ...`). Danke an @tyler6204. - Agents/Z.AI: `tool_stream` standardmäßig aktivieren für Echtzeit-Tool-Aufruf-Streaming, mit Opt-out über `params.tool_stream: false`. (#18173) Danke an @tianxiao1430-jpg. - Plugins: Vorab-Overrides für `before_agent_start` Modell/Provider vor Auflösung hinzufügen. (#18568) Danke an @natefikru. - Mattermost: Emoji-Reaktionsaktionen plus Reaktions-Ereignis-Benachrichtigungen hinzufügen, einschließlich eines expliziten booleschen `remove`-Flags, um versehentliche Entfernungen zu vermeiden. (#18608) Danke an @echo931. - Memory/Search: FTS-Fallback plus Abfrage-Erweiterung für Memory-Suche hinzufügen. (#18304) Danke an @irchelper. - Agents/Models: pro-Modell-Override `thinkingDefault` in Model-Konfiguration unterstützen. (#18152) Danke an @wu-tian807. - Agents: `llms.txt`-Entdeckung im Standardverhalten aktivieren. (#18158) Danke an @yolo-maxi. - Extensions/Auth: OpenAI Codex CLI-Auth-Provider-Integration hinzufügen. (#18009) Danke an @jiteshdhamaniya. - Feishu: Bitable Create-App/Create-Field-Tools für Automatisierungs-Workflows hinzufügen. (#17963) Danke an @gaowanqi08141999. - Docker: optionaler Build-Arg `OPENCLAW_INSTALL_BROWSER`, um Chromium + Xvfb vorzuinstallieren und Laufzeit-Playwright-Installationen zu vermeiden. (#18449) ### Fixes - Agents/Image: Resize-Diagnostiken auf eine Zeile pro Bild zusammenführen und sichtbare Pixel-/Byte-Größen im Log-Eintrag für schnellere Triagierung einbeziehen. - Agents/Subagents: vorauseilende Absicherung des kombinierten Tool-Result-Context vor Modellaufrufen durch Kürzen zu großer Ausgaben und Kompaktieren ältester Tool-Result-Nachrichten, um Kontextfenster-Überläufe zu vermeiden. Danke an @tyler6204. - Agents/Subagents: explizite Subagent-Richtlinien hinzufügen, um aus `[compacted: tool output removed to free context]`/`[truncated: output exceeded

Gemini CLIv0.29.0

Release v0.29.0

18. Februar 2026

## Was sich geändert hat * fix: Entferne `ask_user`-Werkzeug in nicht-interaktiven Modi von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/18154 * fix(cli): Erlaube eingeschränkte .env-Ladung in untrusted sandboxed Ordnern von @galz10 in https://github.com/google-gemini/gemini-cli/pull/17806 * Ermutige Agenten, Ecosystem-Tools zu nutzen, um Arbeiten zu erledigen von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/17881 * feat(plan): Vereinheitliche Workflow-Ortung im System-Prompt zur Optimierung des Cachings von @jerop in https://github.com/google-gemini/gemini-cli/pull/18258 * feat(core): aktiviere getUserTierName in der Konfiguration von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/18265 * feat(core): Standard-Execution-Limits für Subagenten hinzufügen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18274 * Fix-Problem, bei dem der Agent bei interaktiven Befehlen hängenbleibt. von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/18272 * chore(release): Version auf 0.29.0-nightly.20260203.71f46f116 erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18243 * feat(core): harte Umgehung der Richtlinie für lokale Subagenten entfernen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18153 * feat(plan): Implementiere `plan`-Befehl von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/17698 * feat: Erhöhe Beschriftung für `ask_user`-Label auf 16 Zeichen von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/18320 * Füge Informationen über den Lebenszyklus der Agentenfähigkeiten hinzu und kläre Metadaten des Docs-Writers-Fähigkeiten. von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/18234 * feat(core): füge `enter_plan_mode`-Werkzeug hinzu von @jerop in https://github.com/google-gemini/gemini-cli/pull/18324 * Stoppe Anzeige einer Fehlermeldung in `/plan` von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/18333 * fix(hooks): entferne unnötige Logging-Ausgaben bei Hook-Registrierung von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18332 * fix(mcp): sicherstellen, dass MCP-Transport geschlossen wird, um Speicherlecks zu verhindern von @cbcoutinho in https://github.com/google-gemini/gemini-cli/pull/18054 * feat(skills): Verknüpfung für Agentenfähigkeiten implementieren von @MushuEE in https://github.com/google-gemini/gemini-cli/pull/18295 * Changelogs für 0.27.0 und 0.28.0-preview0 von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/18336 * chore: Dokumente korrigieren, da Fähigkeiten und Hooks stabil sind von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/18358 * feat(admin): Admin-Whitelist für MCP-Server-Konfigurationen implementieren von @skeshive in https://github.com/google-gemini/gemini-cli/pull/18311 * fix(core): Retry-Logik für transiente SSL/TLS-Fehler (#17318) von @ppgranger in https://github.com/google-gemini/gemini-cli/pull/18310 * Add support for /extensions config command von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/17895 * fix(core): handle non-kompatible mcpbridge-Antworten von Xcode 26.3 von @peterfriese in https://github.com/google-gemini/gemini-cli/pull/18376 * feat(cli): Unterstützungen für W, B, E Vim-Bewegungen und Operator-Unterstützung von @ademuri in https://github.com/google-gemini/gemini-cli/pull/16209 * fix: Windows-spezifische Agent-Qualität & System-Prompt von @scidomino in https://github.com/google-gemini/gemini-cli/pull/18351 * feat(plan): Unterstütze `replace`-Werkzeug im Plan-Modus zum Bearbeiten von Plänen von @jerop in https://github.com/google-gemini/gemini-cli/pull/18379 * Verbesserung der Anleitungen für Memory-Tool und Evaluations-Tests von @alisa-alisa in https://github.com/google-gemini/gemini-cli/pull/18091 * fix(cli): Farbverlinkungs-Erfolgsmeldung für Extension-Link grün von @MushuEE in https://github.com/google-gemini/gemini-cli/pull/18386 * rückgängig machen von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/18147 * feat(plan): Anleitung zur Iteration über genehmigte Pläne vs. Erstellung neuer Pläne von @jerop in https://github.com/google-gemini/gemini-cli/pull/18346 * feat(plan): Behebe ungültige Tool-Aufrufe im Plan-Modus von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/18352 * feat(plan): Integriere Planungsartefakte und Tools in primäre Arbeitsabläufe von @jerop in https://github.com/google-gemini/gemini-cli/pull/18375 * Fix Berechtigungsprüfung von @scidomino in https://github.com/google-gemini/gemini-cli/pull/18395 * ux(polish) Autocomplete im Eingabe-Prompt von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/18181 * fix: lösche unendliche Schleife bei Verwendung von "Mit externem Editor ändern" von @ppgranger in https://github.com/google-gemini/gemini-cli/pull/17453 * feat: Erweiterung von verify-release auf macOS und Windows von @yunaseoul in https://github.com/google-gemini/gemini-cli/pull/18145 * feat(plan): Unterstützung von MCP-Servern im Plan-Modus von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/18229 * chore: Aktualisiere Ordner-Vertrauens-Meldungen von @galz10 in https://github.com/google-gemini/gemini-cli/pull/18402 * feat(plan): Erzeuge Metrik für Ausführung von Plänen, die im Plan-Modus generiert wurden von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/18236 * perf(ui): Optimierung von stripUnsafeCharacters mit Regex von @gsquared94 in https://github.com/google-gemini/gemini-cli/pull/18413 * feat(context): Implementiere Beobachtungs-Maskierung für Tool-Ausgaben von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18389 * feat(core,cli): Implementiere speichernde Ausgabe von Sitzungen und Aufräumen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/18416 * Kürze temporäres Verzeichnis von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/17901 * feat(plan): Füge Verhaltens-Evalien für Plan-Modus hinzu von @jerop in https://github.com/google-gemini/gemini-cli/pull/18437 * Füge Extension-Registerkarten-Client hinzu von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/18396 * Extension-Konfiguration standardmäßig aktivieren von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/18447 * Automatisch Change-Logs beim Release generieren von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/18401 * Entferne previewFeatures und setze Gemini 3 als Standard von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/18414 * feat(admin): MCP-Whitelist auf Extensions & Gemini MCP-Liste

Codex CLIvrust-v0.103.0

0.103.0

17. Februar 2026

## New Features - App listing responses now include richer app details (`app_metadata`, branding, and labels), so clients can render more complete app cards without extra requests. (#11706) - Commit co-author attribution now uses a Codex-managed `prepare-commit-msg` hook, with `command_attribution` override support (default label, custom label, or disable). (#11617) ## Bug Fixes - Removed the `remote_models` feature flag to prevent fallback model metadata when it was disabled, improving model selection reliability and performance. (#11699) ## Chores - Updated Rust dependencies (`clap`, `env_logger`, `arc-swap`) and refreshed Bazel lock state as routine maintenance. (#11888, #11889, #11890, #12032) - Reverted the Rust toolchain bump to `1.93.1` after CI breakage. (#11886, #12035) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.102.0...rust-v0.103.0 - #11699 chore: rm remote models fflag @sayan-oai - #11706 [apps] Expose more fields from apps listing endpoints. @mzeng-openai - #11890 chore(deps): bump arc-swap from 1.8.0 to 1.8.2 in /codex-rs @dependabot - #11886 chore(deps): bump rust-toolchain from 1.93.0 to 1.93.1 in /codex-rs @dependabot - #12032 chore: just bazel-lock-update @bolinfest - #11888 chore(deps): bump clap from 4.5.56 to 4.5.58 in /codex-rs @dependabot - #11889 chore(deps): bump env_logger from 0.11.8 to 0.11.9 in /codex-rs @dependabot - #11617 Use prompt-based co-author attribution with config override @gabec-openai - #12035 Revert "chore(deps): bump rust-toolchain from 1.93.0 to 1.93.1 in /co…dex-rs (#11886)" @etraut-openai

Continue.devv1.2.16-vscode

v1.2.16-vscode

17. Februar 2026

## Was sich geändert hat * .continue/agents von @sestinj in https://github.com/continuedev/continue/pull/10233 * .continue/agents p2 von @sestinj in https://github.com/continuedev/continue/pull/10235 * feat: eingeschränkte MCP App-Unterstützung von @RomneyDa in https://github.com/continuedev/continue/pull/10132 * feat(cli): curl/irl Installationsskripte und Dokumentation von @RomneyDa in https://github.com/continuedev/continue/pull/9835 * feat: CLI-Befehl zum Hinzufügen von cn-Checks von @sestinj in https://github.com/continuedev/continue/pull/10330 * feat: Aktivierung der Agenten-Modus-Unterstützung für LM Studio-Modelle von @veeceey in https://github.com/continuedev/continue/pull/10327 * docs: Standardmodell auf Claude Opus 4.6 aktualisieren von @BekahHW in https://github.com/continuedev/continue/pull/10270 * [Snyk] Behebung von 1 Sicherheitslücke von @sestinj in https://github.com/continuedev/continue/pull/10386 * fix: Unicode-Validierung im model apikey oder Headern von @uinstinct in https://github.com/continuedev/continue/pull/10380 * docs: GitHub-Trigger-Ereignistypen für Workflows hinzufügen von @BekahHW in https://github.com/continuedev/continue/pull/10213 * docs: Workflows-Dokumentation - MVP aktualisieren von @BekahHW in https://github.com/continuedev/continue/pull/10369 * fix: Windows-Remote-Erkennung für Terminal-Befehle generalisieren von @shanevcantwell in https://github.com/continuedev/continue/pull/10391 * [Snyk] Sicherheitsupgrade axios von 1.13.2 auf 1.13.5 von @sestinj in https://github.com/continuedev/continue/pull/10423 * [Snyk] Sicherheitsupgrade @openapitools/openapi-generator-cli von 2.27.0 auf 2.28.3 von @sestinj in https://github.com/continuedev/continue/pull/10421 * feat: Best Practices für React konsolidieren von @RomneyDa in https://github.com/continuedev/continue/pull/10417 * Rücknahme von "[Snyk] Fix for 1 vulnerabilities" von @RomneyDa in https://github.com/continuedev/continue/pull/10424 * feat: Erstellen/Lese-Regel-Tools standardmäßig deaktivieren von @RomneyDa in https://github.com/continuedev/continue/pull/10543 * chore: Pre-Release-Bumps vscode 32 jetbrains 61 von @RomneyDa in https://github.com/continuedev/continue/pull/10563 * docs: Einführungstext hinzufügen, Beispiel und Textbausteine aktualisieren von @TyDunn in https://github.com/continuedev/continue/pull/10565 * docs: READMEs für neue Produkt-Richtung aktualisieren von @TyDunn in https://github.com/continuedev/continue/pull/10566 * [WIP] Aktualisierung der Docs, um Checks widerzuspiegeln von @sestinj in https://github.com/continuedev/continue/pull/10489 * docs: Check-Best-Practices-Beispiele verbessern und Skill-Links hinzufügen von @sestinj in https://github.com/continuedev/continue/pull/10591 ## Neue Mitwirkende * @veeceey hat seinen ersten Beitrag in https://github.com/continuedev/continue/pull/10327 geleistet **Voller Changelog**: https://github.com/continuedev/continue/compare/v1.2.15-vscode...v1.2.16-vscode

Codex CLIvrust-v0.102.0

0.102.0

17. Februar 2026

## New Features - Added a more unified permissions flow, including clearer permissions history in the TUI and a slash command to grant sandbox read access when directories are blocked. (#11633, #11512, #11550, #11639) - Introduced structured network approval handling, with richer host/protocol context shown directly in approval prompts. (#11672, #11674) - Expanded app-server fuzzy file search with explicit session-complete signaling so clients can stop loading indicators reliably. (#10268, #11773) - Added customizable multi-agent roles via config, including migration toward the new multi-agent naming/config surface. (#11917, #11982, #11939, #11918) - Added a `model/rerouted` notification so clients can detect and render model reroute events explicitly. (#12001) ## Bug Fixes - Fixed remote image attachments so they persist correctly across resume/backtrack and history replay in the TUI. (#10590) - Fixed a TUI accessibility regression where animation gating for screen reader users was not consistently respected. (#11860) - Fixed app-server thread resume behavior to correctly rejoin active in-memory threads and tighten invalid resume cases. (#11756) - Fixed `model/list` output to return full model data plus visibility metadata, avoiding unintended server-side filtering. (#11793) - Fixed several `js_repl` stability issues, including reset hangs, in-flight tool-call races, and a `view_image` panic path. (#11932, #11922, #11800, #11796) - Fixed app integration edge cases in mention parsing and app list loading/filtering behavior. (#11894, #11518, #11697) ## Documentation - Updated contributor guidance to require snapshot coverage for user-visible TUI changes. (#10669) - Updated docs/help text around Codex app and MCP command usage. (#11926, #11813) ## Chores - Improved developer log tooling with new `just log --search` and `just log --compact` modes. (#11995, #11994) - Updated vendored `rg` and tightened Bazel/Cargo lockfile sync checks to reduce dependency drift. (#12007, #11790) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.101.0...rust-v0.102.0 - #10268 app-server: add fuzzy search sessions for streaming file search @nornagon-openai - #11547 Parse first order skill/connector mentions @canvrno-oai - #11227 feat(app-server): experimental flag to persist extended history @owenlin0 - #10672 Add js_repl host helpers and exec end events @fjord-oai - #11512 add a slash command to grant sandbox read access to inaccessible directories @iceweasel-oai - #11631 chore(core) Deprecate approval_policy: on-failure @dylan-hurd-oai - #11636 Better error message for model limit hit. @xl-openai - #11633 feat: introduce Permissions @bolinfest - #10669 docs: require insta snapshot coverage for UI changes @joshka-oai - #11645 fix: skip review_start_with_detached_delivery_returns_new_thread_id o… @owenlin0 - #11639 [feat] add seatbelt permission files @celia-oai - #11622 Remove absolute path in rollout_summary @wendyjiao-openai - #10671 Add js_repl_tools_only model and routing restrictions @fjord-oai - #11657 app-server tests: disable shell_snapshot for review suite @bolinfest - #11646 app-server: stabilize detached review start on Windows @bolinfest - #11638 fix(app-server): surface more helpful errors for json-rpc @owenlin0 - #11417 [apps] Add is_enabled to app info. @mzeng-openai - #11630 Add new apps_mcp_gateway @canvrno-oai - #11656 Persist complete TurnContextItem state via canonical conversion @charley-oai - #11510 Remove git commands from dangerous command checks @joshka-oai - #11668 feat(shell-tool-mcp): add patched zsh build pipeline @nornagon-openai - #11275 Added a test to verify that feature flags that are enabled by default are stable @etraut-openai - #11651 Add cwd as an optional field to thread/list @acrognale-oai - #11660 chore(approvals) More approvals scenarios @dylan-hurd-oai - #11518 [apps] Fix app loading logic. @mzeng-openai - #11679 fix: dont show NUX for upgrade-target models that are hidden @sayan-oai - #11515 Point Codex App tooltip links to app landing page @joshka-oai - #11671 chore(core) Restrict model-suggested rules @dylan-hurd-oai - #11703 fix(ci) lock rust toolchain at 1.93.0 to unblock @dylan-hurd-oai - #11662 feat(network-proxy): structured policy signaling and attempt correlation to core @viyatb-oai - #11709 fix(shell-tool-mcp) build dependencies @dylan-hurd-oai - #11618 feat: add token usage on memories @jif-oai - #11722 Lower missing rollout log level @jif-oai - #11712 chore: streamline phase 2 @jif-oai - #11731 feat: memories config @jif-oai - #11736 feat: increase windows workers stack @jif-oai - #11739 feat: add slug in name @jif-oai - #11745 chore: move explorer to spark @jif-oai - #11748 Fix memories output schema requirements @jif-oai - #11669 core: limit search_tool_bm25 to Apps and clarify discovery guidance @apanasenko-oai - #11755 app-server-test-client websocket client and thread tools @maxj-oai - #11663 fix: reduce flakiness of compact_resume_after_second_compaction_preserves_history @bolinfest - #11667 sandbox NUX metrics update @iceweasel-oai - #11695 Updated app bug report template @etraut-openai - #11477 feat: switch on dying sub-agents @jif-oai - #11711 feat(tui): prevent macOS idle sleep while turns run @yvolovich-cyber - #11686 Report syntax errors in rules file @etraut-openai - #11763 Update read_path prompt @zuxin-oai - #11772 chore: mini @jif-oai - #11697 [apps] Improve app listing filtering. @mzeng-openai - #11666 Add js_repl kernel crash diagnostics @fjord-oai - #11687 support app usage analytics @alexsong-oai - #11769 Improve GitHub issue deduplication reliability by introducing a stage… @etraut-openai - #11770 fix(nix): use correct version from Cargo.toml in flake build @rupurt - #11677 turn metadata: per-turn non-blocking @pash-openai - #11692 rmcp-client: fix auth crash @maxj-oai - #10590 tui: preserve remote image attachments across resume/backtrack @charley-oai - #11782 turn metadata followups @pash-openai - #11773 [app-server] add fuzzyFileSearch/sessionCompleted @nornagon-openai - #11756 codex-rs: fix thread resume rejoin semantics @maxj-oai - #11793 fix: send unfiltered models over model/list @sayan-oai - #11799 fix(protocol): make local image test Bazel-friendly @joshka-oai - #11796 Fix js_repl view_image test runtime panic @fjord-oai - #11800 Fix js_repl in-flight tool-call waiter race @fjord-oai - #11658 feat(skills): add permission profiles from openai.yaml metadata @celia-oai - #11790 bazel: enforce MODULE.bazel.lock sync with Cargo.lock @joshka-oai - #11803 add perf metrics for connectors load @alexsong-oai - #11659 Handle model-switch base instructions after compaction @charley-oai - #11813 Fixed help text for `mcp` and `mcp-server` CLI commands @etraut-openai - #11672 feat(core): add structured network approval plumbing and policy decision model @viyatb-oai - #11674 feat(tui): render structured network approval prompts in approval overlay @viyatb-oai - #11550 feat(tui) Permissions update history item @dylan-hurd-oai - #11767 fix(core): add linux bubblewrap sandbox tag @viyatb-oai - #11534 Add process_uuid to sqlite logs @charley-oai - #11487 core: snapshot tests for compaction requests, post-compaction layout, some additional compaction tests @charley-oai - #11690 fix: show user warning when using default fallback metadata @sayan-oai - #11780 chore(tui): reduce noisy key logging @apanasenko-oai - #11884 fix: only emit unknown model warning on user turns @sayan-oai - #11893 bazel: fix snapshot parity for tests/*.rs rust_test targets @apanasenko-oai - #11759 feat: use shell policy in shell snapshot @jif-oai - #11615 Allow hooks to error @gt-oai - #11918 chore: rename collab feature flag key to multi_agent @jif-oai - #11924 nit: memory storage @jif-oai - #11917 feat: add customizable roles for multi-agents @jif-oai - #11926 docs: mention Codex app in README intro @vb-openai - #11900 feat: drop MCP managing tools if no MCP servers @jif-oai - #11939 Rename collab modules to multi agents @jif-oai - #11894 [apps] Fix app mention syntax. @mzeng-openai - #11866 chore(core) rm Feature::RequestRule @dylan-hurd-oai - #11948 add(feedback): over-refusal / safety check @fouad-openai - #11860 Fixed screen reader regression in CLI @etraut-openai - #11964 add(core): safety check downgrade warning @fouad-openai - #11951 fix(core) exec_policy parsing fixes @dylan-hurd-oai - #11932 fix: js_repl reset hang by clearing exec tool calls without waiting @jif-oai - #11974 Hide /debug slash commands from popup menu @jif-oai - #11922 fix: race in js repl @jif-oai - #11969 fix(ci) Fix shell-tool-mcp.yml @dylan-hurd-oai - #11908 Exit early when session initialization fails @jif-oai - #11986 nit: wording multi-agent @jif-oai - #11995 feat: add `--search` to `just log` @jif-oai - #11994 feat: add `--compact` mode to `just log` @jif-oai - #11833 Don't allow model_supports_reasoning_summaries to disable reasoning @etraut-openai - #11807 Centralize context update diffing logic @charley-oai - #12007 Update vendored rg to the latest stable version (15.1) @etraut-openai - #11970 Protect workspace .agents directory in Windows sandbox @etraut-openai - #12005 Add /statusline tooltip entry @jif-oai - #11982 feat: move agents config to main config @jif-oai - #11224 chore: clarify web_search deprecation notices and consolidate tests @sayan-oai - #12001 Feat: add model reroute notification @shijie-oai - #11801 Add remote skill scope/product_surface/enabled params and cleanup @xl-openai

Claude Codev2.1.45

v2.1.45

17. Februar 2026

## What's changed - Added support for Claude Sonnet 4.6 - Added support for reading `enabledPlugins` and `extraKnownMarketplaces` from `--add-dir` directories - Added `spinnerTipsOverride` setting to customize spinner tips — configure `tips` with an array of custom tip strings, and optionally set `excludeDefault: true` to show only your custom tips instead of the built-in ones - Added `SDKRateLimitInfo` and `SDKRateLimitEvent` types to the SDK, enabling consumers to receive rate limit status updates including utilization, reset times, and overage information - Fixed Agent Teams teammates failing on Bedrock, Vertex, and Foundry by propagating API provider environment variables to tmux-spawned processes (anthropics/claude-code#23561) - Fixed sandbox "operation not permitted" errors when writing temporary files on macOS by using the correct per-user temp directory (anthropics/claude-code#21654) - Fixed Task tool (backgrounded agents) crashing with a `ReferenceError` on completion (anthropics/claude-code#22087) - Fixed autocomplete suggestions not being accepted on Enter when images are pasted in the input - Fixed skills invoked by subagents incorrectly appearing in main session context after compaction - Fixed excessive `.claude.json.backup` files accumulating on every startup - Fixed plugin-provided commands, agents, and hooks not being available immediately after installation without requiring a restart - Improved startup performance by removing eager loading of session history for stats caching - Improved memory usage for shell commands that produce large output — RSS no longer grows unboundedly with command output size - Improved collapsed read/search groups to show the current file or search pattern being processed beneath the summary line while active - [VSCode] Improved permission destination choice (project/user/session) to persist across sessions

Claude Codev2.1.44

v2.1.44

16. Februar 2026

## What's changed - Fixed auth refresh errors

OpenCodev1.2.6

v1.2.6

16. Februar 2026

## Core - Add dfmt formatter support for D language files (@burner) - Bump GitLab provider and auth plugin for mid-session token refresh (@vglafirov) - Remove unnecessary per-message title LLM calls (@rmk40) - Prioritize user-defined variables over environment variables in Google Vertex AI configuration - Add OpenAI-compatible endpoint support for Google Vertex provider (@leehack) - Add Venice support for temperature, topP, topK, and smallOption parameters (@dpuyosa) - Add cljfmt formatter support for Clojure files (@finalfantasia) ## TUI - Make use of server directory path for file references in prompts (@OpeOginni) - Add database migration command to convert JSON storage to SQLite - Add --continue and --fork flags to attach command - Fixed inaccurate tips in TUI (@imanolmzd-svg) ## Desktop - Normalize Linux Wayland/X11 backend and decoration policy (@bnema) - Use process-wrap library instead of manual job object handling in desktop (@Brendonovich) **Thank you to 12 community contributors:** - @finalfantasia: - feat(opencode): add `cljfmt` formatter support for Clojure files (#13426) - @pkx07: - fix(website): correct zh-CN translation of proprietary terms in zen.mdx (#13734) - @Brendonovich: - desktop: use process-wrap instead of manual job object (#13431) - @dpuyosa: - feat(opencode): Add Venice support in temperature, topP, topK and smallOption (#13553) - @bnema: - fix(desktop): normalize Linux Wayland/X11 backend and decoration policy (#13143) - @leehack: - feat: add openai-compatible endpoint support for google-vertex provider (#10303) - @hobostay: - fix(docs): correct critical translation errors in Russian zen page (#13830) - @rmk40: - fix(core): remove unnecessary per-message title LLM calls (#13804) - @imanolmzd-svg: - fix (tui): Inaccurate tips (#13845) - @vglafirov: - fix: bump GitLab provider and auth plugin for mid-session token refresh (#13850) - @OpeOginni: - fix(tui): make use of server dir path for file references in prompts (#13781) - @burner: - feat(opencode): add `dfmt` formatter support for D language files (#13867)

OpenClawv2026.2.15

openclaw 2026.2.15

16. Februar 2026

### Changes - Discord: rich interaktive Agenten-Prompts mit Components v2 freischalten (Knöpfe, Auswahlen, Modalfenster und an Dateianhänge gebundene Dateiblöcke) für native Interaktion über Discord. Danke an @thewilloftheshadow. - Discord: UI von Components v2 + Embeds passthrough + UX-Feinabstimmungen zur Ausführungsgenehmigung (CV2-Container, Button-Layout, Discord-Weiterleitung-Skip). Danke an @thewilloftheshadow. - Plugins: expose `llm_input` und `llm_output` Hook-Payloads, damit Erweiterungen Prompt/Input-Kontext und Modell-Ausnutzungshäufigkeiten beobachten können. (#16724) Danke an @SecondThread. - Subagents: verschachtelte Sub-Agents (Sub-Sub-Agents) mit konfigurierbarer Tiefe. Setze `agents.defaults.subagents.maxSpawnDepth: 2`, um Sub-Agents das Spawnen eigener Kinder zu erlauben. Enthält `maxChildrenPerAgent`-Limit (Standard 5), tiefergreifende Tool-Policy und ordnungsgemäße Ankündigungs-Chain-Routing. (#14447) Danke an @tyler6204. - Slack/Discord/Telegram: füge pro-Kanal-Akku-Reaktions-Overrides (Konto/Kanal-Ebene) hinzu, um plattform-spezifische Emoji-Formate zu unterstützen. (#17092) Danke an @zerone0x. - Cron/Gateway: füge fertige Lauf-Webhooks-Benachrichtigungssperre (`notify`) und dedizierte Webhook-Auth-Token-Unterstützung (`cron.webhookToken`) für ausgehende Cron-Webhook-Posts hinzu. (#14535) Danke an @advaitpaliwal. - Channels: dedupliziere Basis-Typen zur Probe-/Tokenauflösung über Core + Extensions hinweg, während die Fehlertypisierung pro Kanal erhalten bleibt. (#16986) Danke an @iyoda und @thewilloftheshadow. ### Fixes - Sicherheit: ersetze veraltete SHA-1 Sandbox-Konfigurations-Hashing durch SHA-256 für deterministische Sandbox-Cache-Identität und -Neuerstellungstests. Danke an @kexinoh. - Sicherheit/Logging: redaktiere Telegram-Bot-Tokens aus Fehlermeldungen und ungekachelten Stack-Traces, um versehentliche Geheimnis-Lecks in Logs zu verhindern. Danke an @aether-ai-agent. - Sandbox/Sicherheit: blockiere gefährliche Sandbox-Docker-Konfigurationen (Bind-Mounts, Host-Netzwerk, unkonfigurierter Seccomp/AppArmor), um Container-Entkommen via Config-Injektion zu verhindern. Danke an @aether-ai-agent. - Sandbox: beibehalte Array-Reihenfolge in Config-Hashing, sodass reihenfolgenabhängige Docker-/Browser-Einstellungen eine korrekte Container-Neuerstellung auslösen. Danke an @kexinoh. - Gateway/Sicherheit: redaktiere sensible Sitzungs-/Pfad-Details aus `status`-Antworten für Nicht-Administratoren; vollständige Details bleiben dem `operator.admin` vorbehalten. (#8590) Danke an @fr33d3m0n. - Gateway/Control UI: bewahre angefragte Operatoren-Scope bei Control-UI-Bypass-Modi (`allowInsecureAuth` / `dangerouslyDisableDeviceAuth`), wenn Geräte-Identität nicht verfügbar ist, um falsche `missing scope`-Fehler bei authentifizierten LAN/HTTP-Operator-Sessions zu verhindern. (#17682) Danke an @leafbird. - LINE/Sicherheit: scheitere beim Webhook-Start, wenn Kanal-Token oder Kanal-Geheimnis fehlen; LINE-Konten gelten nur als konfiguriert, wenn beides vorhanden ist. (#17587) Danke an @davidahmann. - Skills/Sicherheit: beschränke Installer-`download`-`targetDir` auf das pro-Skill-Tools-Verzeichnis, um willkürliche Dateischreibungen zu verhindern. Danke an @Adam55A-code. - Skills/Linux: härte den Go-Installer-Fallback auf apt-basierte Systeme an, sicherer Umgang mit Root/kein-Sudo-Umgebungen, best-effort Aktualisierung des apt-Index, und aussagekräftige Fehler statt Spawn-Fehler. (#17687) Danke an @mcrolly. - Web Fetch/Sicherheit: begrenze die heruntergeladenen Antwortkörpergrößen vor HTML-Parsing, um Speicherüberlauf durch zu große oder tief verschachtelte Seiten zu verhindern. Danke an @xuemian168. - Config/Gateway: mache die Whitelist-Suffix-Matching groß-/klein-schreibunabhängig, während `passwordFile`-Ausnahmen erhalten bleiben, um versehentliche Redaktionen nicht-geheimer Config-Werte wie `maxTokens` und IRC-Passwort-Dateipfade zu vermeiden. (#16042) Danke an @akramcodez. - Dev-Tooling: härte Git `pre-commit` Hook gegen Options-Injektion aus böswilligen Dateinamen (z. B. `--force`) ab, um versehentliches Stagen ignorierter Dateien zu verhindern. Danke an @mrthankyou. - Gateway/Agent: lehne malformed `agent:`-Session-Keys (z. B. `agent:main`) in `agent` und `agent.identity.get` ab, statt sie still auf den Standard-Agenten aufzulösen, um versehentliche Cross-Session-Routing zu verhindern. (#15707) Danke an @rodrigouroz. - Gateway/Chat: härte eingehende `chat.send`-Nachrichtenbehandlung, indem Null-Bytes abgelehnt, unsichere Steuerzeichen entfernt und Unicode zu NFC normalisiert wird, bevor dispatch. (#8593) Danke an @fr33d3m0n. - Gateway/Send: gib einen aktionsfähigen Fehler zurück, wenn `send` auf internes `webchat`-Ziel geht; leite Aufrufer an `chat.send` oder einen deliverable channel weiter. (#15703) Danke an @rodrigouroz. - Control UI: verhindere gespeicherten XSS durch Assistant-Namen/Avatar durch Entfernen von Inline-Skripteinjektion, Serving von Bootstrap-Config als JSON und Durchsetzung von `script-src 'self'`. Danke an @Adam55A-code. - Agents/Sicherheit: bereinige Arbeitsbereich-Pfade, bevor sie in LLM-Prompts eingebettet werden (Unicode-Steuer-/Formatzeichen entfernen), um Instruktions-Injektion durch schädliche Verzeichnisnamen zu verhindern. Danke an @aether-ai-agent. - Agents/Sandbox: kläre Systemprompt-Pfad-Anweisungen, sodass Sandbox-`bash/exec` Container-Pfade (z. B. `/workspace`) nutzt, während Dateitools Host-Bridge-Mapping beibehalten, um Pfad-Fehlversuche beim ersten Versuch wegen host-only absoluten Pfaden im Sandbox-Befehl-Execution zu vermeiden. (#17693) Danke an @app/juniordevbot. - Agents/Context: wende konfigurierte Modell-`contextWindow`-Overrides nach Provider-Discovery an, sodass `lookupContextTokens()` Operator-Konfig-Werte (einschließlich Discovery-Fehlerpfade) respektiert. (#17404) Danke an @michaelbship und @vignesh07. - Agents/Context: leite `lookupContextTokens()` aus auth-verfügbaren Modell-Metadaten ab und halte das kleinste entdeckte Context-Fenster für Duplikat-Modell-IDs, um Cache-Kollisionen zwischen Providern durch überschätzen des Sitzungs-Kontexts zu verhindern. (#17586) Danke an @githabideri und @vignesh07. - Agents/OpenAI: zwinge `store=true` für direkte OpenAI-Responses/Codex-Läufe, um Multi-Turn-Server-seitigen Gesprächszustand beizubehalten, während Proxy-/Non-OpenAI-Endpunkte unverändert bleiben. (#16803) Danke an @mark9232 und @vignesh07. - Memory/FTS: mache `buildFtsQuery` Unicode-bewusst, sodass nicht-ASCII-Anfragen (einschließlich CJK) Schlagwort-Tokens erzeugen statt rein Vektor-Suche zu verwenden. (#17672) Danke an @KinGP5471. - Auto-reply/Compaction: löse Platzhalter `memory/YYYY-MM-DD.md` mit zeitzonenbewussten Laufzeitdaten auf und füge eine `Current time:`-Zeile zu Memory-Flush-Turns hinzu, um falsche Jahres-Memory-Dateinamen zu verhindern, ohne das Systemprompt-Zeit-Varianz zu erzeugen. (#17603, #17633) Danke an @nicholaspapadam-wq und @vignesh07. - Agents: gebe eine explizite Timeout-Antwort zurück, wenn ein eingebetteter Lauf time-out geht, bevor Payloads produziert werden, um stilles Droppen von Turns während langsamer Cache-Refresh-Transaktionen zu verhindern. (#16659) Danke an @liaosvcaf und @vignesh07. - Gruppen-Chats: immer Gruppenkontext (Name, Teilnehmer

OpenCodev1.2.5

v1.2.5

15. Februar 2026

## Core - Ensure SQLite migration logs to stderr instead of stdout ## Desktop - Fixed issue viewing new files opened from the file tree (@shanebishop1) - Only navigate prompt history at input boundaries (@nexxeln) - Add keyboard shortcut Shift+Tab to the application (@neriousy) - Focus window after update and relaunch (@zerone0x) - Add GeistMono Nerd Font to available mono font options (@brandon-julio-t) **Thank you to 7 community contributors:** - @brandon-julio-t: - feat: Add GeistMono Nerd Font to available mono font options (#13720) - @zerone0x: - fix(desktop): focus window after update/relaunch (#13701) - @dector: - docs: add Ukrainian README translation (#13697) - @neriousy: - fix(app): keybind [shift+tab] (#13695) - @nexxeln: - fix(app): only navigate prompt history at input boundaries (#13690) - @shanebishop1: - fix(desktop): issue viewing new files opened from the file tree (#13689) - @alexyaroshuk: - feat(app): localize "free usage exceeded" error & "Add credits" clickable link (#13652)

OpenClawv2026.2.14

openclaw 2026.2.14

15. Februar 2026

### Änderungen - Telegram: Poll-Senden via `openclaw message poll` hinzufügen (Dauer in Sekunden, stumme Übertragung, Anonymitätskontrollen). (#16209) Danke an @robbyczgw-cla. - Slack/Discord: `dmPolicy` + Konfig-Alias `allowFrom` für DM-Zugriffssteuerung hinzufügen; Legacy-Keys `dm.policy` + `dm.allowFrom` bleiben unterstützt und `openclaw doctor --fix` kann sie migrieren. - Discord: Ausführungsgenehmigungs-Promptings auf Zielkanäle oder sowohl DM+Kanal via `channels.discord.execApprovals.target` zulassen. (#16051) Danke an @leonnardo. - Sandbox: `sandbox.browser.binds` hinzufügen, um Browser-Container-Bind-Mounts getrennt von Exec-Containern zu konfigurieren. (#16230) Danke an @seheepeak. - Discord: Debug-Logging für Routing-Entscheidungen der Nachrichten hinzufügen, um `--debug`-Tracing zu verbessern. (#16202) Danke an @jayleekr. ### Fixes - CLI/Plugins: sicherstellen, dass `openclaw message send` nach erfolgreicher Lieferung über plattformgestützte Kanäle beendet wird, damit One-Shot-Sendungen nicht hängen bleiben. (#16491) Danke an @yinghaosang. - CLI/Plugins: registrierte Plugin-`gateway_stop`-Hooks vor dem Beenden von `openclaw message` ausführen (Erfolgs- und Fehlerpfade), damit plugin-basierte Kanäle CLI-Ressourcen für One-Shot bereinigen können. (#16580) Danke an @gumadeiras. - WhatsApp: pro-Konto Overrides von `dmPolicy` berücksichtigen (konto-spezifische Einstellungen haben Vorrang vor Kanalausprägungen für eingehende DMs). (#10082) Danke an @mcaxtr. - Telegram: Wenn `channels.telegram.commands.native` auf `false` steht, Plugin-Befehle aus dem `setMyCommands`-Menü entfernen, Slas-Händler der Plugins bleiben jedoch aufrufbar. (#15132) Danke an @Glucksberg. - LINE: 200 OK für Developers Console-Verifizierungsanfragen (`{"events":[]}`) zurückgeben, ohne `X-Line-Signature`, Signaturen für echte Deliveries bleiben jedoch erforderlich. (#16582) Danke an @arosstale. - Cron: Text-Ausgabe direkt liefern, wenn `delivery.to` gesetzt ist, damit Cron-Empfänger die volle Ausgabe statt Zusammenfassungen erhalten. (#16360) Danke an @thewilloftheshadow. - Cron/Slack: Agenten-Identität (Name und Icon) beibehalten, wenn Cron-Jobs ausgehende Nachrichten liefern. (#16242) Danke an @robbyczgw-cla. - Media: akzeptiere `MEDIA:`-präfixierte Pfade (lasche Whitespace) beim Laden ausgehender Medien, um `ENOENT` für lokal vom Tool zurückgegebene Medienpfade zu verhindern. (#13107) Danke an @mcaxtr. - Agents: sende Tool-Ergebnis-Medien (Screenshots, Bilder, Audio) an Kanäle unabhängig vom verbose-Level. (#11735) Danke an @strelov1. - Agents/Image-Tool: Arbeitsbereich-lokale Bildpfade zulassen, indem aktive Arbeitsbereich-Verzeichnisse in lokalen Mediens-Whitelist aufgenommen werden, und vertrauen auf Sandbox-validierte Pfade in Bild-Laders, um falsche "nicht in einem erlaubten Verzeichnis"-Ablehnungen zu verhindern. (#15541) - Agents/Image-Tool: effektiven Arbeitsbereich-Wurzelpfad in Tool-Verkabelung propagieren, damit Arbeitsbereich-lokale Bildpfade standardmäßig akzeptiert werden, wenn kein expliziter `workspaceDir` gesetzt ist. (#16722) - BlueBubbles: Sender-Identität in Gruppen-Chats-Envelope aufnehmen und sauberen Nachrichtentext an das Agenten-Prompt weitergeben, im Einklang mit iMessage/Signal-Formatierung. (#16210) Danke an @zerone0x. - CLI: lazy Core-Befehlsregistrierung korrigieren, damit Top-Level-Wartungsbefehle (`doctor`, `dashboard`, `reset`, `uninstall`) korrekt aufgelöst werden und kein funktionsloser `maintenance`-Platzhalterbefehl erscheint. - CLI/Dashboard: Wenn `gateway.bind=lan`, localhost-Dashboard-URLs erzeugen, um Browser-Sicherheitskontexte zu erfüllen, while Nicht-LAN-Bind-Verhalten beibehalten. (#16434) Danke an @BinHPdev. - TUI/Gateway: lokale Gateway-Ziel-URL aus Modus `gateway.bind` (Tailnet/lan) ableiten statt hartkodiertem localhost, damit sich `openclaw tui` bei nicht-Loopback-Gateways verbindet. (#16299) Danke an @cortexuvula. - TUI: explizites `--session <key>` in `openclaw tui` auch dann beachten, wenn `session.scope` `global` ist, damit benannte Sessions nicht mehr mit gemeinsamem globale History kollabieren. (#16575) Danke an @cinqu. - TUI: verfügbare Terminalbreite für Anzeige des Sitzungsnamens in durchsuchbaren Auswahllisten nutzen. (#16238) Danke an @robbyczgw-cla. - TUI: Layout der Beschreibungsliste für durchsuchbare Auswahllisten refaktoren und Regressionen für ANSI-Hervorhebung in Breiten testen. - TUI: laufende Streaming-Antworten beibehalten, wenn ein anderer Run gleichzeitig finalisiert, um aktive Runs nicht zu löschen oder History mitten im Stream neu zu laden. (#10704) Danke an @axschr73. - TUI: vor Werkzeug-Streams sichtbaren Text beibehalten, wenn spätere Tool-Grenzen-Deltas vorhergehende Textblöcke vorübergehend weglassen. (#6958) Danke an @KrisKind75. - TUI: ANSI/Steuerzeichen-lastige Historien-Text sanieren, binäre Zeilen redigieren und lange ununterbrochene Tokens vor der Darstellung aufteilen, um Startfehler bei binären Anhängen in der History zu verhindern. (#13007) Danke an @wilkinspoe. - TUI: Rendern des Assistant-Textes im Standard-Terminalforeground (statt fester heller ANSI-Farbe) damit der Kontrast in hellen Themes wie Solarized Light lesbar bleibt. (#16750) Danke an @paymog. - TUI/Hooks: explizite Reset-Grundlage (`new` vs `reset`) durch `sessions.reset` weiterreichen und interne Befehls-Hooks für gateway-getriggerte Resets auslösen, damit `/new` Hook-Workflows in TUI/Webchat funktionieren. - Cron: verhindern, dass `cron list`/`cron status` verspätete wiederkehrende Jobs still überspringen, indem Wartungs-Neuberechnungsemantiken verwendet werden. (#16156) Danke an @zerone0x. - Cron: fehlende/korrupte `nextRunAtMs` für aktualisierte Jobs reparieren, ohne globale Neuberechnung unverbundener Jobs während `cron update`. (#15750) - Cron: überspringen Wiederholungs-Replay bei Startup für Jobs, die mitten im Lauf unterbrochen wurden (veraltete `runningAtMs`-Marker), um Neustart-Schleifen für selbstneu startende Jobs wie Update-Aufgaben zu vermeiden. (#16694) Danke an @sbmilburn. - Discord: bevorzugen Gateway-Guild-ID beim Protokollieren eingehender Nachrichten, damit gecachte Guilds nicht als `guild=dm` erscheinen. Danke an @thewilloftheshadow. - Discord: leere pro-Guild Config-Maps `channels: {}` als kein Kanal-Allowlist interpretieren (nicht ablehnen), sodass Guilds mit `groupPolicy: "open"` auch ohne explizite Kanaleinträge weiterhin Nachrichten erhalten. (#16714) Danke an @xqliu. - Models/CLI: schützen `models status`-Strings vor Zuschneiden von Pfaden, um Abstürze durch fehlerhafte nicht-String-Konfig-Werte zu verhindern. (#16395) Danke an @BinHPdev. - Gateway/Subagents: beibehalten von geparkten Announcement-Items und Statuszustand bei Lieferungsfehlern, erneute Versuche beim Fehler-Drain, vermeiden, dass unbeantwortete Ankündigungen bei Timeout/Fehler verloren gehen. (#16729) Danke an @Clawdette-Workspace. - Gateway/Sessions: aktive eingebettete Runs abbrechen und geparkte Sitzungsarbeit vor `sessions.reset` löschen, unavailable melden, wenn Run nicht rechtzeitig stoppt. (#16576) Danke an @Grynn. - Sessions/Agents: Transkriptpfadauflösung für inkonsistente Agenten-Kontexte härten, explizite Store-Wurzel beibehalten und sicheren Absolutpfad-Fallback in das korrekte Agenten-Sessions-Verzeichnis hinzufügen. (#16288) Danke an @robbyczgw-cla. - Agents: Sicherheits-Timeout um embeddeed `session.compact()` hinzufügen, um festste

OpenCodev1.2.4

v1.2.4

15. Februar 2026

## Core - Add db command for database inspection and querying - Derive all IDs from file paths during JSON migration ## Desktop - Clear notifications action - Fixed stack overflow issue in file tree component

OpenCodev1.2.3

v1.2.3

15. Februar 2026

## Core - Ensure Anthropic models on OpenRouter also have variant support - Add WAL checkpoint on database open - Ensure Vercel variants pass Amazon models under Bedrock key

OpenCodev1.2.2

v1.2.2

14. Februar 2026

## Core - Add comprehensive test coverage for Session.list() filters - Filter sessions at database level to improve session list loading performance - Fix Vercel gateway variants - Bump Vertex AI packages

OpenCodev1.2.1

v1.2.1

14. Februar 2026

v1.2.0 and beyond includes a data migration that will execute on first run. It will migrate all flat files in data directory to a single sqlite database. Depending on how much data you have and speed of computer this can take some time. if anything goes wrong you can retrigger the migration by deleting `~/.local/share/opencode/opencode.db*` files (%APPDATA% on windows) If you have any issues with the migration the original data is not yet deleted and downgrading should work. But please open an issue so we can investigate and include sqlite in the issue title. SDK Users We now have a PartDelta event which sends only incremental changes to parts. This avoids sending the full content of text parts over and over when it is updated ## Core - Show all project sessions from any working directory - Tweak websearch tool description date info to avoid cache busts

OpenCodev1.2.0

v1.2.0

14. Februar 2026

This release includes a data migration that will execute on first run. It will migrate all flat files in data directory to a single sqlite database. Depending on how much data you have and speed of computer this can take some time. if anything goes wrong you can retrigger the migration by deleting ~/.local/share/opencode/opencode.db* files (%APPDATA% on windows) If you have any issues with the migration the original data is not yet deleted and downgrading should work. But please open an issue so we can investigate and include sqlite in the issue title. SDK Users We now have a PartDelta event which sends only incremental changes to parts. This avoids sending the full content of text parts over and over when it is updated ```js PartDelta: BusEvent.define( "message.part.delta", z.object({ sessionID: z.string(), messageID: z.string(), partID: z.string(), field: z.string(), delta: z.string(), }), ) ``` ## Core - Revert to SQLite database implementation - Move timeout configuration from programmatic API to CLI flag (@yanhao98) - Update AI SDK packages and use adaptive reasoning for Claude Opus 4.6 on Vertex, Bedrock, and Anthropic ## TUI - Prevent crash when tool inputs are malformed during opencode run (@0xK3vin) - Add --dir option to run command (@BlankParticle) ## Desktop - Fix Rust compilation issues in desktop application (@Brendonovich) - Remove OPENCODE_SQLITE environment variable from desktop (@Brendonovich) - Terminal resize handling fixed in app - Use prompt_async endpoint to avoid timeout over VPN/tunnel (@eytans) - Remount SDK and sync tree when server URL changes (@ysm-dev) - Sync docs locale cookie on alias redirects (@Seungjun0906) - Performance optimization for showing large diffs and files in desktop (@neriousy) **Thank you to 12 community contributors:** - @BlankParticle: - feat(cli): add --dir option to run command (#12443) - @neriousy: - fix(desktop): performance optimization for showing large diff & files (#13460) - @Seungjun0906: - fix(web): sync docs locale cookie on alias redirects (#13109) - docs(ko): polish Korean phrasing in acp, agents, config, and custom-tools docs (#13446) - @ysm-dev: - fix(app): remount SDK/sync tree when server URL changes (#13437) - @Annopick: - docs: Fix zh-cn translation mistake in tools.mdx (#13407) - @eytans: - fix(web): use prompt_async endpoint to avoid timeout over VPN/tunnel (#12749) - @G36maid: - docs: add pacman installation option for Arch Linux alongside AUR (#13293) - @yanhao98: - fix(test): move timeout config to CLI flag (#13494) - @niushuai1991: - fix: standardize zh-CN docs character set and terminology (#13500) - @kitlangton: - fix(ui): support cmd-click links in inline code (#12552) - @0xK3vin: - fix: prevent opencode run crash on malformed tool inputs (#13051) - @Brendonovich: - desktop: remote OPENCODE_SQLITE env (#13545) - desktop: fix rust

OpenClawv2026.2.13

openclaw 2026.2.13

14. Februar 2026

### Änderungen - Discord: Voice-Nachrichten mit Waveform-Vorschau aus lokalen Audiodateien senden (einschließlich stiller Lieferung). (#7253) Danke @nyanjou. - Discord: konfigurierbaren Presence-Status/Aktivität/Typ/URL hinzufügen (benutzerdefinierter Status standard zu Aktivitätstext). (#10855) Danke @h0tp-ftw. - Slack/Plugins: Thread-Eigentums-Outbound-Gating via `message_sending` Hooks hinzufügen, einschließlich @-Mention Umgehungs-Tracking und Slack Outbound Hook Verkabelung für Cancel/Modify-Verhalten. (#15775) Danke @DarlingtonDeveloper. - Agents: synthetischen Katalog-Support für `hf:zai-org/GLM-5` hinzufügen. (#15867) Danke @battman21. - Skills: doppelten `local-places` Google Places Skill/Proxy entfernen und `goplaces` als einzigen unterstützten Google Places-Pfad beibehalten. - Agents: Vor-Prompt-Kontextdiagnostik (`messages`, `systemPromptChars`, `promptChars`, Anbieter/Modell, Sitzungsdatei) vor eingebetteten Runner-Prompt-Aufrufen hinzufügen, um Overflow-Debugging zu verbessern. (#8930) Danke @Glucksberg. - Onboarding/Providers: erstklassige Hugging Face Inference-Provider-Unterstützung hinzufügen (Provider-Verkabelung, Onboarding-Auth-Auswahl/API-Key-Fluss und Standard-Modell-Auswahl) und Hugging Face Auth-Intent bei Auth-Choice-Remapping beibehalten (`tokenProvider=huggingface` mit `authChoice=apiKey`), while Env-Override-Prompts überspringen, wenn ein explizites Token bereitgestellt wird. (#13472) Danke @Josephrp. ### Behebungen - Outbound: Schreib-vorne-Lieferwarteschlange mit Crash-Recovery-Wiederholungen hinzufügen, um verloren gegangene Outbound-Nachrichten nach Gateway-Neustarts zu verhindern. (#15636) Danke @nabbilkhan, @thewilloftheshadow. - Auto-Antwort/Threading: implicitly Threading beim Antwort-Threading auto-injektieren, sodass `replyToMode` funktioniert ohne modell-emittierte `[[reply_to_current]]`, while `replyToMode: "off"` Verhalten für implizite Slack-Antworten beibehalten und Block-Streaming-Chunks bei `replyToMode: "first"` stabil halten. (#14976) Danke @Diaspar4u. - Outbound/Threading: `replyTo` und `threadId` aus `message send` Tool-Aktionen durch den Core-Outbound-Send-Pfad zu Channel-Adaptern weitergeben, Umleitung von Thread/Reply beibehalten. (#14948) Danke @mcaxtr. - Auto-Antwort/Media: inbound Bilder-Nachrichten (mit Bild) zulassen (kein Caption) um den Agenten zu erreichen statt als leeren Text abzubrechen, und Thread-Kontext in eingereihten/Follow-up-Prompt-Textkörpern für Medienläufe beibehalten. (#11916) Danke @arosstale. - Discord: AutoThread-Antworten in bestehende Threads statt in den Root-Channel routen. (#8302) Danke @gavinbmoore, @thewilloftheshadow. - Web UI: `img` zu DOMPurify erlaubte Tags hinzufügen und `src`/`alt` zu erlaubten Attributen, damit Markdown-Bilder im Webchat gerendert werden statt entfernt. (#15437) Danke @lailoo. - Telegram/Matrix: MP3 und M4A (einschließlich `audio/mp4`) als sprachkompatibel für `asVoice` Routing behandeln, WAV/AAC weiterhin für reguläre Audio-Sends verwenden. (#15438) Danke @azade-c. - WhatsApp: Outbound-Dokumentdateinamen für Web-Session-Dokumentensendungen beibehalten statt immer `"file"`. (#15594) Danke @TsekaLuk. - Telegram: Bot-Menü-Registrierung auf Telegrams 100-Befehls-Limit beschränken mit Overflow-Warnung, während getippte versteckte Befehle weiterhin verfügbar bleiben. (#15844) Danke @battman21. - Telegram: Skill-Befehle auf den aufgelösten Agenten für Standardkonten einschränken, damit `setMyCommands` nicht mehr `BOT_COMMANDS_TOO_MUCH` auslöst, wenn mehrere Agenten konfiguriert sind. (#15599) - Discord: numerische Guild-Allowlist-Einträge nicht falsch routen, indem Eingaben mit prefix `guild:` versehen werden während der Auflösung. (#12326) Danke @headswim. - MS Teams: parse Mention-Entitäten/Text beibehalten beim Anhängen von OneDrive-Fallback-Dateilinks, breitere reale Teams Mention-ID-Formate (`29:...`, `8:orgid:...`) akzeptieren, während Platzhaltermuster weiterhin abgelehnt werden. (#15436) Danke @hyojin. - Media: `text/*` MIME-Typen als Dokumente im Media-Kind Routing klassifizieren, damit Text-Anhänge nicht als unbekannt behandelt werden. (#12237) Danke @arosstale. - Inbound/Web UI: wörtliche `\n`-Sequenzen beim Normalisieren eingehender Texte beibehalten, damit Windows-Pfade wie `C:\\Work\\nxxx\\README.md` nicht beschädigt werden. (#11547) Danke @mcaxtr. - TUI/Streaming: reichere gestreamte Assistant-Text-Bereiche beibehalten, wenn finales Payload vor Tool-Aufruf Textblöcke verliert, while finales Payload weiterhin maßgeblich für Klartext-Updates ist. (#15452) Danke @TsekaLuk. - Providers/MiniMax: impliziten MiniMax API-Key Provider von `openai-completions` zu `anthropic-messages` wechseln mit korrekter Anthropic-kompatibler Basis-URL, Fehler `invalid role: developer (2013)` bei MiniMax M2.5 reparieren. (#15275) Danke @lailoo. - Ollama/Agents: gelöste Modell-/Provider-Basis-URLs für nativen `/api/chat` Streaming verwenden (einschließlich alias Provider), `/v1` Endpunkte normalisieren, Abort + `maxTokens`-Streaming-Optionen für zuverlässige Abbruchvorgänge und Token-Grenzen weiterleiten. (#11853) Danke @BrokenFinger98. - OpenAI Codex/Spark: End-to-End-Unterstützung `gpt-5.3-codex-spark` über Fallback/Thinking/Model-Resolution und `models list` Forward-Compat- Sichtbarkeit implementieren. (#14990, #15174) Danke @L-U-C-K-Y, @loiie45e. - Agents/Codex: `gpt-5.3-codex-spark` in Forward-Compat-Fallback, Live-Modell-Filterung und Thinking-Presets zulassen, Modell-Auswahl-Erkennung für spark korrigieren. (#14990) Danke @L-U-C-K-Y. - Models/Codex: konfigurierte `openai-codex/gpt-5.3-codex-spark` über Forward-Compat-Fallback bei `models list` lösen, damit es nicht fälschlich als fehlend markiert wird, wenn Laufzeitauflösung gelingt. (#15174) Danke @loiie45e. - OpenAI Codex/Auth: OpenClaw OAuth-Profile in `pi` `auth.json` übertragen, damit Modell-Erkennung und Models-List-Registry-Auflösung Codex OAuth-Anmeldeinformationen nutzen können. (#15184) Danke @loiie45e. - Auth/OpenAI Codex: OAuth-Login-Behandlung über Onboarding und `models auth login --provider openai-codex` hinweg teilen, Onboarding beibehalten falls OAuth fehlschlägt, direkte OAuth-Hilfe-Hinweis statt Wizard-Beendung anzeigen. (#15406, Folge zu #14552) Danke @zhiluo20. - Onboarding/Providers: vLLM als Onboarding-Provider mit Modellerkennung, Auth-Profil-Verkabelung und nicht-interaktiver Auth-Choice-Validierung hinzufügen. (#12577) Danke @gejifeng. - Onboarding/CLI: Terminalzustand wiederherstellen, ohne pausierte `stdin` fortzufahren, Onboarding sauber beenden nachdem Web UI gewählt wurde und der Installer zurückkehrt statt stecken zu bleiben. - Signal/Install: automatisches Installieren von `signal-cli` via Homebrew auf Nicht-x64 Linux-Architekturen, Vermeidung von `Exec format error`-Fehlern bei arm64/arm-Hosts. (#15443) Danke @jogvan-k. - macOS Voice Wake: Absturz bei Trigger-Trimmung für CJK/Unicode-Transkripte reparieren, indem Original-String-Bereiche statt transformierter-String-Indizes gematcht und geschnitten werden. (#11052) Danke @Flash-LHR. - Mattermost (Plugin): Websocket-Monitor-Verbindungen mit exponentiellem Backoff erneut versuchen und abort-bewusste Bereinigung, sodass

Claude Codev2.1.42

v2.1.42

13. Februar 2026

## What's changed - Fixed /resume showing interrupt messages as session titles - Fixed Opus 4.6 launch announcement showing for Bedrock/Vertex/Foundry users - Improved error message for many-image dimension limit errors with /compact suggestion

Claude Codev2.1.41

v2.1.41

13. Februar 2026

## What's changed - Fixed AWS auth refresh hanging indefinitely by adding a 3-minute timeout - Added `claude auth login`, `claude auth status`, and `claude auth logout` CLI subcommands - Added Windows ARM64 (win32-arm64) native binary support - Improved `/rename` to auto-generate session name from conversation context when called without arguments - Improved narrow terminal layout for prompt footer - Fixed file resolution failing for @-mentions with anchor fragments (e.g., `@README.md#installation`) - Fixed FileReadTool blocking the process on FIFOs, `/dev/stdin`, and large files - Fixed background task notifications not being delivered in streaming Agent SDK mode - Fixed cursor jumping to end on each keystroke in classifier rule input - Fixed markdown link display text being dropped for raw URL - Fixed auto-compact failure error notifications being shown to users - Fixed permission wait time being included in subagent elapsed time display - Fixed proactive ticks firing while in plan mode - Fixed clear stale permission rules when settings change on disk - Fixed hook blocking errors showing stderr content in UI

OpenCodev1.1.65

v1.1.65

13. Februar 2026

## Core - Revert token substitution in OPENCODE_CONFIG_CONTENT - Ensure @-mentioning a directory uses the read tool instead of deprecated list tool - Add tool.definition hook for plugins to modify tool description and parameters (@spoons-and-mirrors) - Remove worktree delete functionality - Resolve ACP hanging indefinitely in thinking state on Windows (@ASidorenkoCode) ## Desktop - Reconnect event stream on disconnect - Toggle all provider models in settings - Clean up desktop loading page (@Brendonovich) - Notification should navigate to session - Fix prompt input behavior quirks - Failed to create store in app - Only show loading window if SQLite migration is necessary (@Brendonovich) **Thank you to 3 community contributors:** - @ASidorenkoCode: - fix: resolve ACP hanging indefinitely in thinking state on Windows (#13222) - @Brendonovich: - desktop: only show loading window if sqlite migration is necessary - cleanup desktop loading page - @spoons-and-mirrors: - feat(hook): add tool.definition hook for plugins to modify tool description and parameters (#4956)

OpenClawv2026.2.12

openclaw 2026.2.12

13. Februar 2026

### Änderungen - CLI: füge `openclaw logs --local-time` hinzu, um Zeitstempel von Logs in der lokalen Zeitzone anzuzeigen. (#13818) Danke an @xialonglee. - Telegram: rendern Blockzitate als native `<blockquote>`-Tags statt sie zu entfernen. (#14608) - Konfiguration: vermeide das Redaktieren von Feldern ähnlich `maxTokens` während des Snapshot-Redaction der Config, um Round-Trip-Validierungsfehler in `/config` zu verhindern. (#14006) Danke an @constansino. ### Breaking - Hooks: `POST /hooks/agent` lehnt Payload-Überschreibungen des `sessionKey` standardmäßig ab. Um den festen Hook-Kontext beizubehalten, setzen Sie `hooks.defaultSessionKey` (empfohlen mit `hooks.allowedSessionKeyPrefixes: ["hook:"]`). Falls Legacy-Verhalten erforderlich ist, setzen Sie explizit `hooks.allowRequestSessionKey: true`. Danke an @alpernae für den Hinweis. ### Fixes - Gateway/OpenResponses: härtet URL-basierte Verarbeitung von `input_file`/`input_image` mit expliziter SSRF-Verweigerungsrichtlinie, Host-Whitelist (`files.urlAllowlist` / `images.urlAllowlist`), pro-Anfrage URL-Input-Begrenzungen (`maxUrlParts`), blockierte-Abfrage-Audit-Logging und Regression-Abdeckung/Docs-Updates. - Sicherheit: behebt unauthentifizierte Remote-Konfig-Manipulation des Nostr-Profils API. (#13719) Danke an @coygeek. - Sicherheit: entfernt gebündelten bösen Hook. (#14757) Danke an @Imccccc. - Sicherheit/Audit: fügt Hardening-Prüfungen für Hook-Session-Routing hinzu (`hooks.defaultSessionKey`, `hooks.allowRequestSessionKey` und Präfix-Whitelists) und warnt, wenn HTTP-API-Endpunkte explizites Session-Key-Routing zulassen. - Sicherheit/Sandbox: beschränkt abgeglichene Skill-Sync-Ziele auf das Sandbox-Wurzelverzeichnis `skills/` und vermeidet die Verwendung von Frontmatter-gesteuerten Skill-Namen als Dateisystem-Destinationen. Danke an @1seal. - Sicherheit/Web-Tools: behandelt Browser/Web-Inhalte standardmäßig als untrusted (wrappe Ausgaben für Browser-Snapshots/Tabs/Console und strukturierte External-Content-Metadaten für Web-Tools) und entfernt `toolResult.details` aus modellnahen Transkript-/Compaction-Eingaben, um Replay-Risiko von Prompt-Injection zu verringern. - Sicherheit/Hooks: härtere Verifizierung von Webhook- und Geräte-Token durch konstantzeit-vergleichbaren Secrets-Vergleich und pro-Client-Authentifizierungs-Rate-Limiting für Hook-Endpunkte (`429` + `Retry-After`). Danke an @akhmittra. - Sicherheit/Browser: Authentifizierung für Loopback-Browsersteuerungs-HTTP-Routen erforderlich, auto-generiertes `gateway.auth.token` wenn Browser-Steuerung ohne Auth startet, und Sicherheits-Audit-Prüfung für unauthentisierte Browser-Steuerung. Danke an @tcusolle. - Sitzungen/Gateway: härtet Transkriptpfadauflösung und lehnt unsichere Session-IDs/Dateipfade ab, damit Session-Operationen in Agenten-Sessions-Verzeichnissen bleiben. Danke an @akhmittra. - Gateway: erhöhen der WS-Payload-/Buffer-Limits, damit 5.000.000-Byte große Bildanhänge zuverlässig funktionieren. (#14486) Danke an @0xRaini. - Logging/CLI: lokale Zeitzonen-Zeitstempel für Konsolen-Präfixing verwenden, und `±HH:MM`-Offsets einbeziehen beim Verwenden von `openclaw logs --local-time`, um Mehrdeutigkeit zu vermeiden. (#14771) Danke an @0xRaini. - Gateway: aktive Turns vor dem Neustart leeren, um Nachrichtenverlust zu verhindern. (#13931) Danke an @0xRaini. - Gateway: beim Installieren automatisch Auth-Token generieren, um Neustart-Schleifen von launchd zu verhindern. (#13813) Danke an @cathrynlavery. - Gateway: verhindern, dass `undefined`/fehlendes Token in der Auth-Konfiguration auftritt. (#13809) Danke an @asklee-klawd. - Gateway: asynchrones `EPIPE` auf stdout/stderr während des Herunterfahrens behandeln. (#13414) Danke an @keshav55. - Gateway/Control UI: fehlende Dashboard-Ressourcen beim globalen Openclaw-Install über symlink-basierte Node-Manager (nvm/fnm/n/Homebrew) auflösen. (#14919) Danke an @aynorica. - Cron: angefordetes `agentId` für isolierte Job-Auth-Auflösung verwenden. (#13983) Danke an @0xRaini. - Cron: verhindern, dass Cron-Jobs die Ausführung überspringen, wenn `nextRunAtMs` vorrückt. (#14068) Danke an @WalterSumbon. - Cron: übergibt `agentId` an `runHeartbeatOnce` für Haupt-Session-Jobs. (#14140) Danke an @ishikawa-pro. - Cron: Timer neu starten, wenn `onTimer` auslöst, während ein Job noch ausgeführt wird. (#14233) Danke an @tomron87. - Cron: Duplizierte Ausführungen verhindern, wenn mehrere Jobs gleichzeitig ausgelöst werden. (#14256) Danke an @xinhuagu. - Cron: Scheduler-Fehler isolieren, damit ein schlecht laufender Job nicht alle jobs bricht. (#14385) Danke an @MarvinDontPanic. - Cron: verhindern, dass One-shot-`at`-Jobs nach Neustart erneut feuern, wenn Runs übersprungen/fehlerhaft waren. (#13878) Danke an @lailoo. - Heartbeat: verhindern Scheduler-Verzögerungen bei unerwarteten Lauf-Fehlern und vermeiden sofortige Wiederholungs-Schleifen nach `requests-in-flight`-Skips. (#14901) Danke an @joeykrug. - Cron: geerbte gespeicherte Session-Modell-Overrides für isolierte-Agenten-Läufe beachten, während Priorität von `hooks.gmail.model` für Gmail-Hook-Sitzungen beibehalten wird. (#14983) Danke an @shtse8. - Logging/Browser: fallback auf `os.tmpdir()/openclaw`, wenn `/tmp/openclaw` nicht verfügbar ist, für Standard-Log, Browser-Spur und Browser-Download-Temp-Pfade. - WhatsApp: Markdown-Formatierung für Fett/Durchgestrichen in WhatsApp-Formatierung umwandeln. (#14285) Danke an @Raikan10. - WhatsApp: nur-Medien-Sendungen zulassen und führende leere Payloads normalisieren. (#14408) Danke an @karimnaguib. - WhatsApp: Standard-MIME-Typ für Sprachnachrichten, wenn Baileys ihn weglässt. (#14444) Danke an @mcaxtr. - Telegram: keine Textnachrichten im Modell-Auswahlbearbeiter editMessageText behandeln. (#14397) Danke an @0xRaini. - Telegram: REACTION_INVALID als nicht-fatal-warnung anzeigen. (#14340) Danke an @0xRaini. - BlueBubbles: Webhook-Auth-Bypass über Loopback-Proxy-Vertrauen korrigieren. (#13787) Danke an @coygeek. - Slack: Standard-ReplyToMode von "off" auf "all" ändern. (#14364) Danke an @nm-de. - Slack: Kontrollen erkennen, wenn Channel-Nachrichten mit Bot-Erwähnungs-Prefixen beginnen (z. B. `@Bot /new`). (#14142) Danke an @beefiker. - Signal: E.164-Validierung für Signal-Bot-Kontoaufforderung erzwingen, damit falsch eingegebene Nummern früh erkannt werden. (#15063) Danke an @Duartemartins. - Discord: DM-Reaktionen verarbeiten statt sie still zu verwerfen. (#10418) Danke an @mcaxtr. - Discord: Relevanz von replyToMode in Threads beachten. (#11062) Danke an @cordx56. - Heartbeat: Rausfiltern von Rausch-Systemereignissen, damit geplante Erinnerungsbenachrichtigungen nicht feuern, wenn Cron-Läufe nur Heartbeat-Markierungen tragen. (#13317) Danke an @pvtclawn. - Signal: Erwähnungs-Platzhalter als `@uuid`/`@phone` rendern, damit Erwähnungs-Gating und Clawdbot-Zielsetzung funktionieren. (#2013) Danke an @alexgleason. - Discord: leere Inhaltsfelder bei Medien-Nachrichten weglassen, Caption-Whitespace beibehalten. (#9507) Danke an @leszekszpunar. - Onboarding/Anbieter: z-AI-spezifische Auth-Auswahlmöglichkeiten hinzufügen (`zai-coding-global`, `zai-coding-cn`, `zai-global`, `zai

OpenCodev1.1.64

v1.1.64

12. Februar 2026

## Core - Token substitution in OPENCODE_CONFIG_CONTENT environment variable (@ariane-emory) - Look for recent model in fallback in CLI (@smitchaudhary) - Baseline CPU detection fixed (@Hona) - Windows selection behavior and manual Ctrl+C handling (@Hona) - Terminal PTY isolation for app ## TUI - Do not open console on error ## Desktop - Option to turn off sound effects - Normalize OAuth error messages - Suggestion active state broken - Fixed remote HTTP server connections in the app **Thank you to 4 community contributors:** - @amankalra172: - docs: update STACKIT provider documentation with typo fix (#13357) - @Hona: - feat: windows selection behavior, manual ctrl+c (#13315) - fix: baseline CPU detection (#13371) - @smitchaudhary: - fix: look for recent model in fallback in cli (#12582) - @ariane-emory: - fix: token substitution in OPENCODE_CONFIG_CONTENT (#13384)

Codex CLIvrust-v0.101.0

0.101.0

12. Februar 2026

## Bug Fixes - Model resolution now preserves the requested model slug when selecting by prefix, so model references stay stable instead of being rewritten. (#11602) - Developer messages are now excluded from phase-1 memory input, reducing noisy or irrelevant content entering memory. (#11608) - Memory phase processing concurrency was reduced to make consolidation/staging more stable under load. (#11614) ## Chores - Cleaned and simplified the phase-1 memory pipeline code paths. (#11605) - Minor repository maintenance: formatting and test-suite hygiene updates in remote model tests. (#11619) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.100.0...rust-v0.101.0 - #11605 chore: drop and clean from phase 1 @jif-oai - #11602 fix(core) model_info preserves slug @dylan-hurd-oai - #11608 exclude developer messages from phase-1 memory input @wendyjiao-openai - #11591 Add cwd to memory files @wendyjiao-openai - #11614 chore: reduce concurrency of memories @jif-oai - #11619 fix: fmt @jif-oai

OpenCodev1.1.63

v1.1.63

12. Februar 2026

## Core - Improve Codex model list

Codex CLIvrust-v0.100.0

0.100.0

12. Februar 2026

## New Features - Added an experimental, feature-gated JavaScript REPL runtime (`js_repl`) that can persist state across tool calls, with optional runtime path overrides. (#10674) - Added support for multiple simultaneous rate limits across the protocol, backend client, and TUI status surfaces. (#11260) - Reintroduced app-server websocket transport with a split inbound/outbound architecture, plus connection-aware thread resume subscriptions. (#11370, #11474) - Added memory management slash commands in the TUI (`/m_update`, `/m_drop`) and expanded memory-read/metrics plumbing. (#11569, #11459, #11593) - Enabled Apps SDK apps in ChatGPT connector handling. (#11486) - Promoted sandbox capabilities on both Linux and Windows, and introduced a new `ReadOnlyAccess` policy shape for configurable read access. (#11381, #11341, #11387) ## Bug Fixes - Fixed websocket incremental output duplication, prevented appends after `response.completed`, and treated `response.incomplete` as an error path. (#11383, #11402, #11558) - Improved websocket session stability by continuing ping handling when idle and suppressing noisy first-retry errors during quick reconnects. (#11413, #11548) - Fixed stale thread entries by dropping missing rollout files and cleaning stale DB metadata during thread listing. (#11572) - Fixed Windows multi-line paste reliability in terminals (especially VS Code integrated terminal) by increasing paste burst timing tolerance. (#9348) - Fixed incorrect inheritance of `limit_name` when merging partial rate-limit updates. (#11557) - Reduced repeated skill parse-error spam during active edits by increasing file-watcher debounce from 1s to 10s. (#11494) ## Documentation - Added JS REPL documentation and config/schema guidance for enabling and configuring the feature. (#10674) - Updated app-server websocket transport documentation in the app-server README. (#11370) ## Chores - Split `codex-common` into focused `codex-utils-*` crates to simplify dependency boundaries across Rust workspace components. (#11422) - Improved Rust release pipeline throughput and reliability for Windows and musl targets, including parallel Windows builds and musl link fixes. (#11488, #11500, #11556) - Prevented GitHub release asset upload collisions by excluding duplicate `cargo-timing.html` artifacts. (#11564) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.99.0...rust-v0.100.0 - #11383 Do not resend output items in incremental websockets connections @pakrym-oai - #11246 chore: persist turn_id in rollout session and make turn_id uuid based @celia-oai - #11260 feat: support multiple rate limits @xl-openai - #11412 tui: show non-file layer content in /debug-config @bolinfest - #11405 Remove `test-support` feature from `codex-core` and replace it with explicit test toggles @bolinfest - #11428 fix: flaky test @jif-oai - #11429 feat: improve thread listing @jif-oai - #11422 feat: split codex-common into smaller utils crates @bolinfest - #11439 feat: new memory prompts @jif-oai - #11305 Cache cloud requirements @gt-oai - #11452 nit: increase max raw memories @jif-oai - #11455 feat: close mem agent after consolidation @jif-oai - #11454 fix: optional schema of memories @jif-oai - #11449 feat: set policy for phase 2 memory @jif-oai - #11420 chore: rename disable_websockets -> websockets_disabled @sayan-oai - #11402 Do not attempt to append after response.completed @pakrym-oai - #11462 clean: memory rollout recorder @jif-oai - #11381 feat(core): promote Linux bubblewrap sandbox to Experimental @viyatb-oai - #11389 Extract `codex-config` from `codex-core` @bolinfest - #11370 Reapply "Add app-server transport layer with websocket support" @maxj-oai - #11470 feat: panic if Constrained<WebSearchMode> does not support Disabled @bolinfest - #11475 feat: remove "cargo check individual crates" from CI @bolinfest - #11459 feat: memory read path @jif-oai - #11471 chore: clean rollout extraction in memories @jif-oai - #9348 fix(tui): increase paste burst char interval on Windows to 30ms @yuvrajangadsingh - #11464 chore: sub-agent never ask for approval @jif-oai - #11414 Linkify feedback link @pakrym-oai - #11480 chore: update mem prompt @jif-oai - #11485 fix: Constrained import @owenlin0 - #11341 Promote Windows Sandbox @iceweasel-oai - #10674 Add feature-gated freeform js_repl core runtime @fjord-oai - #11419 refactor: codex app-server ThreadState @maxj-oai - #11413 Pump pings @pakrym-oai - #11488 feat: use more powerful machines for building Windows releases @bolinfest - #11479 nit: memory truncation @jif-oai - #11494 Increased file watcher debounce duration from 1s to 10s @etraut-openai - #11335 Add AfterToolUse hook @gt-oai - #11500 feat: build windows support binaries in parallel @bolinfest - #11290 chore(tui) Simplify /status Permissions @dylan-hurd-oai - #11503 Make codex-sdk depend on openai/codex @pakrym-oai - #11474 app-server: thread resume subscriptions @maxj-oai - #11277 Added seatbelt policy rule to allow os.cpus @etraut-openai - #11506 chore: inject originator/residency headers to ws client @apanasenko-oai - #11497 Hydrate previous model across resume/fork/rollback/task start @aibrahim-oai - #11513 feat: try to fix bugs I saw in the wild in the resource parsing logic @bolinfest - #11509 Consolidate search_tool feature into apps @apanasenko-oai - #11388 change model cap to server overload @willwang-openai - #11504 Pre-sampling compact with previous model context @aibrahim-oai - #11516 Clamp auto-compact limit to context window @aibrahim-oai - #11520 Update context window after model switch @aibrahim-oai - #11519 Use slug in tui @pakrym-oai - #11522 fix: add --test_verbose_timeout_warnings to bazel.yml @bolinfest - #11526 fix: remove errant Cargo.lock files @bolinfest - #11521 test(app-server): stabilize app/list thread feature-flag test by using file-backed MCP OAuth creds @bolinfest - #11387 feat: make sandbox read access configurable with `ReadOnlyAccess` @bolinfest - #11486 [apps] Allow Apps SDK apps. @mzeng-openai - #11532 fix compilation @sayan-oai - #11531 Teach codex to test itself @pakrym-oai - #11540 ci: remove actions/cache from rust release workflows @bolinfest - #11542 ci(windows): use DotSlash for zstd in rust-release-windows @bolinfest - #11498 build(linux-sandbox): always compile vendored bubblewrap on Linux; remove CODEX_BWRAP_ENABLE_FFI @viyatb-oai - #11545 fix: make project_doc skill-render tests deterministic @bolinfest - #11543 ci: capture cargo timings in Rust CI and release workflows @bolinfest - #11539 Bump rmcp to 0.15 @gpeal - #11548 Hide the first websocket retry @pakrym-oai - #11551 Add logs to model cache @aibrahim-oai - #11556 Fix rust-release failures in musl linking and release asset upload @bolinfest - #11558 Handle response.incomplete @pakrym-oai - #11557 fix: stop inheriting rate-limit limit_name @xl-openai - #11564 rust-release: exclude cargo-timing.html from release assets @bolinfest - #11546 fix: update memory writing prompt @zuxin-oai - #11448 Fix test flake @gt-oai - #11569 feat: mem slash commands @jif-oai - #11573 Fix flaky pre_sampling_compact switch test @jif-oai - #11571 feat: mem drop cot @jif-oai - #11572 Ensure list_threads drops stale rollout files @jif-oai - #11575 fix: db stuff mem @jif-oai - #11581 nit: upgrade DB version @jif-oai - #11577 feat: truncate with model infos @jif-oai - #11590 chore: clean consts @jif-oai - #11593 feat: metrics to memories @jif-oai - #11579 Fix config test on macOS @gt-oai - #11600 feat: add sanitizer to redact secrets @jif-oai - #11609 chore: drop mcp validation of dynamic tools @jif-oai

OpenCodev1.1.62

v1.1.62

12. Februar 2026

## Core - Return image attachments from webfetch tool - Expose tool arguments in shell hook for plugin visibility ## Desktop - Project icons no longer unload unexpectedly - Preserve undo history for plain-text paste in app (@gitRasheed) - Refactor app for better SolidJS hygiene - More defensive session context metrics handling in app **Thank you to 2 community contributors:** - @dtfiedler: - fix(docs): correct `format` attribute in `StructuredOutputs` (#13340) - @gitRasheed: - fix(app): preserve undo history for plain-text paste (#13351)

OpenCodev1.1.61

v1.1.61

12. Februar 2026

## Core - Allow model configurations without npm/api provider details - Tool outputs are now more LLM-friendly ## Desktop - Add diff virtualization to improve performance for large diffs - Display progress bar for SQLite migrations in desktop app (@Brendonovich) **Thank you to 1 community contributor:** - @Brendonovich: - desktop: sqlite migration progress bar (#13294) - Testing SignPath Integration (#13308)

OpenCodev1.1.60

v1.1.60

12. Februar 2026

## Core - Support Claude agent SDK-style structured outputs in the OpenCode SDK (@K-Mistele) - Support custom API URL per model - Add automatic variant generation for Venice models (@dpuyosa) - Use Promise.all for MCP listTools calls to improve performance - Upgrade OpenTUI to version 0.1.79 - Improve compaction check logic - Make read tool offset 1-indexed to match line numbers and avoid confusion - Add directory reading capability to the read tool ## TUI - Use FFI to resolve Windows raw input and Ctrl+C handling issues (@Hona) - Add toggle to hide session header in TUI (@kitlangton) ## Desktop - Guard randomUUID in insecure browser contexts - Workspace reset functionality fixed (@neriousy) **Thank you to 6 community contributors:** - @neriousy: - fix(app):workspace reset (#13170) - @github-actions[bot]: - Update VOUCHED list - @dpuyosa: - feat(opencode): Venice Add automatic variant generation for Venice models (#12106) - @kitlangton: - feat(tui): add toggle to hide session header (#13244) - @K-Mistele: - feat: support claude agent SDK-style structured outputs in the OpenCode SDK (#8161) - @Hona: - fix(win32): use ffi to get around bun raw input/ctrl+c issues (#13052)

Gemini CLIv0.28.2

Release v0.28.2

11. Februar 2026

## Was sich geändert hat * fix(patch): cherry-pick e9a9474 to release/v0.28.1-pr-18840 [CONFLICTS] by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18847 **Vollständiger Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.28.1...v0.28.2

Codex CLIvrust-v0.99.0

0.99.0

11. Februar 2026

## New Features - Running direct shell commands no longer interrupts an in-flight turn; commands can execute concurrently when a turn is active. (#10513) - Added `/statusline` to configure which metadata appears in the TUI footer interactively. (#10546) - The TUI resume picker can now toggle sort order between creation time and last-updated time with an in-picker mode indicator. (#10752) - App-server clients now get dedicated APIs for steering active turns, listing experimental features, resuming agents, and opting out of specific notifications. (#10721, #10821, #10903, #11319) - Enterprise/admin requirements can now restrict web search modes and define network constraints through `requirements.toml`. (#10964, #10958) - Image attachments now accept GIF and WebP inputs in addition to existing formats. (#11237) - Enable snapshotting of the shell environment and `rc` files (#11172) ## Bug Fixes - Fixed a Windows startup issue where buffered keypresses could cause the TUI sign-in flow to exit immediately. (#10729) - Required MCP servers now fail fast during start/resume flows instead of continuing in a broken state. (#10902) - Fixed a file-watcher bug that emitted spurious skills reload events and could generate very large log files. (#11217) - Improved TUI input reliability: long option labels wrap correctly, Tab submits in steer mode when idle, history recall keeps cursor placement consistent, and stashed drafts restore image placeholders correctly. (#11123, #10035, #11295, #9040) - Fixed model-modality edge cases by surfacing clearer `view_image` errors on text-only models and stripping unsupported image history during model switches. (#11336, #11349) - Reduced false approval mismatches for wrapped/heredoc shell commands and guarded against empty command lists in exec policy evaluation. (#10941, #11397) ## Documentation - Expanded app-server docs and protocol references for `turn/steer`, experimental-feature discovery, `resume_agent`, notification opt-outs, and null `developer_instructions` normalization. (#10721, #10821, #10903, #10983, #11319) - Updated TUI composer docs to reflect draft/image restoration, steer-mode Tab submit behavior, and history-navigation cursor semantics. (#9040, #10035, #11295) ## Chores - Reworked npm release packaging so platform-specific binaries are distributed via `@openai/codex` dist-tags, reducing package-size pressure while preserving platform-specific installs (including `@alpha`). (#11318, #11339) - Pulled in a security-driven dependency update for `time` (RUSTSEC-2026-0009). (#10876) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.98.0...rust-v0.99.0 - #10729 fix(tui): flush input buffer on init to prevent early exit on Windows @Ashutosh0x - #10689 fix: flaky landlock @jif-oai - #10513 Allow user shell commands to run alongside active turns @jif-oai - #10738 nit: backfill stronger @jif-oai - #10246 adding fork information (UI) when forking @pap-openai - #10748 Update explorer role default model @jif-oai - #10425 Include real OS info in metrics. @iceweasel-oai - #10745 feat: resumable backfill @jif-oai - #10758 feat: wire ephemeral in `codex exec` @jif-oai - #10756 chore: handle shutdown correctly in tui @jif-oai - #10637 feat: add memory tool @jif-oai - #10751 feat: repair DB in case of missing lines @jif-oai - #10762 nit: add DB version in discrepancy recording @jif-oai - #10621 Leverage state DB metadata for thread summaries @jif-oai - #9691 Add hooks implementation and wire up to `notify` @gt-oai - #10546 feat(tui): add /statusline command for interactive status line configuration @fcoury - #10752 feat(tui): add sortable resume picker with created/updated timestamp toggle @fcoury - #10769 fix(tui): fix resume_picker_orders_by_updated_at test @owenlin0 - #10423 fix(auth): isolate chatgptAuthTokens concept to auth manager and app-server @owenlin0 - #10775 nit: gpt-5.3-codex announcement @jif-oai - #10782 nit: gpt-5.3-codex announcement 2 @jif-oai - #10711 add sandbox policy and sandbox name to codex.tool.call metrics @iceweasel-oai - #10660 chore: rm web-search-eligible header @sayan-oai - #10783 fix: announcement in prio @jif-oai - #10721 [app-server] Add a method to list experimental features. @mzeng-openai - #10787 chore: limit update to 0.98.0 NUX to < 0.98.0 ver @sayan-oai - #10655 Add analytics for /rename and /fork @pap-openai - #10790 feat: wait for backfill to be ready @jif-oai - #10693 Add app-server transport layer with websocket support @maxj-oai - #10818 other announcement @jif-oai - #10815 Sync app-server requirements API with refreshed cloud loader @xl-openai - #10820 go back to auto-enabling web_search for azure @sayan-oai - #10727 Send beta header with websocket connects @pakrym-oai - #10809 updates: use brew api for version check @magus - #10793 Add stage field for experimental flags. @mzeng-openai - #10821 feat(app-server): turn/steer API @owenlin0 - #10792 Print warning when config does not meet requirements @gt-oai - #10699 feat: expose detailed metrics to runtime metrics @apanasenko-oai - #10784 Gate app tooltips to macOS @aibrahim-oai - #10843 Log an event (info only) when we receive a file watcher event @etraut-openai - #10852 Personality setting is no longer available in experimental menu @etraut-openai - #10840 Removed the "remote_compaction" feature flag @etraut-openai - #10876 sec: fix version of `time` to prevent vulnerability @jif-oai - #10892 nit: test an @jif-oai - #10894 feat: backfill async again @jif-oai - #10902 Handle required MCP startup failures across components @jif-oai - #10851 Removed "exec_policy" feature flag @etraut-openai - #10457 Queue nudges while plan generating @charley-oai - #10822 Add app configs to config.toml @canvrno-oai - #10420 feat(network-proxy): add structured policy decision to blocked errors @viyatb-oai - #10814 fix(linux-sandbox): block io_uring syscalls in no-network seccomp policy @viyatb-oai - #10698 core: preconnect Responses websocket for first turn @joshka-oai - #10574 core: refresh developer instructions after compaction replacement history @charley-oai - #10914 chore(app-server): update AGENTS.md for config + optional collection guidance @owenlin0 - #10928 chore(app-server): add experimental annotation to relevant fields @owenlin0 - #10927 Treat compaction failure as failure state @aibrahim-oai - #10861 Support alternative websocket API @by-openai - #10826 add originator to otel @alexsong-oai - #10855 TUI/Core: preserve duplicate skill/app mention selection across submit + resume @daniel-oai - #10943 app-server: print help message to console when starting websockets server @JaviSoto - #10938 Mark Config.apps as experimental, correct schema generation issue @canvrno-oai - #10947 fix(tui): conditionally restore status indicator using message phase @sayan-oai - #10965 refactor(network-proxy): flatten network config under [network] @viyatb-oai - #10970 Fixed a flaky test @etraut-openai - #10710 Process-group cleanup for stdio MCP servers to prevent orphan process storms @etraut-openai - #10964 feat: add support for allowed_web_search_modes in requirements.toml @bolinfest - #10977 fix: use expected line ending in codex-rs/core/config.schema.json @bolinfest - #10973 Do not poll for usage when using API Key auth @etraut-openai - #10921 Show left/right arrows to navigate in tui request_user_input @charley-oai - #10988 fix: normalize line endings when reading file on Windows @bolinfest - #10903 Add resume_agent collab tool @jif-oai - #10909 Bootstrap shell commands via user shell snapshot @jif-oai - #10993 Fix flaky windows CI test @etraut-openai - #10987 Fixed a flaky Windows test that is consistently causing a CI failure @etraut-openai - #10958 feat(core): add network constraints schema to requirements.toml @viyatb-oai - #10983 app-server: treat null mode developer instructions as built-in defaults @charley-oai - #11039 feat: include state of [experimental_network] in /debug-config output @bolinfest - #11040 Simplify pre-connect @pakrym-oai - #10966 feat: enable premessage-deflate for websockets @apanasenko-oai - #9040 fix(tui): rehydrate drafts and restore image placeholders @Chriss4123 - #10824 Fallback to HTTP on UPGRADE_REQUIRED @pakrym-oai - #11028 Defer persistence of rollout file @etraut-openai - #10980 fix: remove config.schema.json from tag check @bolinfest - #11051 Gate view_image tool by model input_modalities @wiltzius-openai - #11109 [bazel] Upgrade some rulesets in preparation for enabling windows @zbarsky-openai - #11114 chore: refactor network-proxy so that ConfigReloader is injectable behavior @bolinfest - #10718 Upgrade rmcp to 0.14 @mzeng-openai - #11044 feat: include [experimental_network] in <environment_context> @bolinfest - #10994 [apps] Improve app loading. @mzeng-openai - #11121 chore: reverse the codex-network-proxy -> codex-core dependency @bolinfest - #11105 feat: include NetworkConfig through ExecParams @bolinfest - #11155 tui: avoid no-op status-line redraws @rakan-oai - #10799 feat: do not close unified exec processes across turns @jif-oai - #11172 chore: enable shell snapshot @jif-oai - #11175 fix: do not show closed agents in `/agent` @jif-oai - #11173 chore: enable sub agents @jif-oai - #11193 Deflake mixed parallel tools timing test @gt-oai - #10770 Load requirements on windows @gt-oai - #11132 core: account for all post-response items in auto-compact token checks @charley-oai - #11198 tools: remove get_memory tool and tests @jif-oai - #10937 Translate websocket errors @rasmusrygaard - #11217 Fixed bug in file watcher that results in spurious skills update events and large log files @etraut-openai - #11216 Move warmup to the task level @pakrym-oai - #11203 Try to stop small helper methods @pakrym-oai - #11197 [bazel] Upgrade some rulesets in preparation for enabling windows, part 2 @zbarsky-openai - #11158 chore: remove network-proxy-cli crate @viyatb-oai - #11230 Revert "chore: enable sub agents" @jif-oai - #11123 TUI: fix request_user_input wrapping for long option labels @charley-oai - #11133 core: add focused diagnostics for remote compaction overflows @charley-oai - #10657 feat: search_tool @apanasenko-oai - #11199 state: add memory consolidation lock primitives @jif-oai - #10835 feat: extend skills/list to support additional roots. @xl-openai - #10960 skill-creator: Remove invalid reference. @xl-openai - #11219 feat: use a notify instead of grace to close ue process @jif-oai - #11231 feat: tie shell snapshot to cwd @jif-oai - #10962 fix(tui): keep unified exec summary on working line @joshka-oai - #11232 Add originator to otel metadata tags @alexsong-oai - #11237 adding image support for gif and webp @natea-oai - #10924 [apps] Add gated instructions for Apps. @mzeng-openai - #11228 Use longest remote model prefix matching @aibrahim-oai - #11242 fix(feature) UnderDevelopment feature must be off @dylan-hurd-oai - #11185 fix: nix build by adding missing dependencies and fix outputHashes @Philipp-M - #10035 fix(tui): tab submits when no task running in steer mode @joshka-oai - #11238 Remove offline fallback for models @aibrahim-oai - #9739 Update models.json @github-actions - #11255 Revert "Update models.json" @aibrahim-oai - #11245 deflake linux-sandbox NoNewPrivs timeout @joshka-oai - #11256 Revert "Revert "Update models.json"" @aibrahim-oai - #11262 chore: change ConfigState so it no longer depends on a single config.toml file for reloading @bolinfest - #11263 test: deflake nextest child-process leak in MCP harnesses @joshka-oai - #11247 Adjust shell command timeouts for Windows @dylan-hurd-oai - #11240 fix(app-server): for external auth, replace id_token with chatgpt_acc… @owenlin0 - #11140 chore(deps): bump insta from 1.46.2 to 1.46.3 in /codex-rs @dependabot - #11139 chore(deps): bump anyhow from 1.0.100 to 1.0.101 in /codex-rs @dependabot - #11138 chore(deps): bump regex from 1.12.2 to 1.12.3 in /codex-rs @dependabot - #11239 Disable dynamic model refresh for custom model providers @etraut-openai - #11269 feat: reserve loopback ephemeral listeners for managed proxy @bolinfest - #11279 [apps] Add thread_id param to optionally load thread config for apps feature check. @mzeng-openai - #11244 feat: add SkillPolicy to skill metadata and support allow_implicit_invocation @alexsong-oai - #10215 chore(tui) cleanup /approvals @dylan-hurd-oai - #11113 feat(sandbox): enforce proxy-aware network routing in sandbox @viyatb-oai - #10940 feat: support configurable metric_exporter @alexsong-oai - #11294 chore: put crypto provider logic in a shared crate @bolinfest - #11207 feat: retain NetworkProxy, when appropriate @bolinfest - #11200 memories: add extraction and prompt module foundation @jif-oai - #11191 feat: add connector capabilities to sub-agents @jif-oai - #11304 Fix spawn_agent input type @jif-oai - #11300 feat: align memory phase 1 and make it stronger @jif-oai - #11311 Extract hooks into dedicated crate @jif-oai - #11306 feat: phase 2 consolidation @jif-oai - #11318 chore: split NPM packages @jif-oai - #11322 Fix pending input test waiting logic @jif-oai - #11265 Remove ApiPrompt @pakrym-oai - #11295 tui: keep history recall cursor at line end @joshka-oai - #11288 fix(protocol): approval policy never prompt @fouad-openai - #11323 Revert "Add app-server transport layer with websocket support (#10693)" @maxj-oai - #11162 Fix: update parallel tool call exec approval to approve on request id @shijie-oai - #11249 [apps] Improve app installation flow. @mzeng-openai - #11319 feat: opt-out of events in the app-server @jif-oai - #11241 Treat first rollout session_meta as canonical thread identity @guinness-oai - #11339 # Use `@openai/codex` dist-tags for platform binaries instead of separate package names @bolinfest - #11330 test(core): stabilize ARM bazel remote-model and parallelism tests @dylan-hurd-oai - #11345 core: remove stale apply_patch SandboxPolicy TODO in seatbelt @bolinfest - #11343 Compare full request for websockets incrementality @pakrym-oai - #11344 fix: reduce usage of `open_if_present` @jif-oai - #11336 Always expose view_image and return unsupported image-input error @aibrahim-oai - #11346 Sanitize MCP image output for text-only models @aibrahim-oai - #11337 Extract tool building @pakrym-oai - #10941 fix(core): canonicalize wrapper approvals and support heredoc prefix … @viyatb-oai - #10946 include sandbox (seatbelt, elevated, etc.) as in turn metadata header @iceweasel-oai - #11349 Strip unsupported images from prompt history to guard against model switch @aibrahim-oai - #11348 Use thin LTO for alpha Rust release builds @bolinfest - #11334 chore: unify memory job flow @jif-oai - #11364 feat: mem v2 - PR1 @jif-oai - #11365 feat: mem v2 - PR2 @jif-oai - #11366 feat: mem v2 - PR3 @jif-oai - #11274 Update models.json @github-actions - #11361 # Split command parsing/safety out of `codex-core` into new `codex-command` @bolinfest - #11369 feat: mem v2 - PR4 @jif-oai - #11362 Enable SOCKS defaults for common local network proxy use cases @viyatb-oai - #11359 ci: fall back to local Bazel on forks without BuildBuddy key @joshka-oai - #11372 feat: mem v2 - PR5 @jif-oai - #11374 feat: mem v2 - PR6 (consolidation) @jif-oai - #11377 feat: prevent double backfill @jif-oai - #11378 chore: rename codex-command to codex-shell-command @bolinfest - #11376 Update models.json @github-actions - #11394 Disable very flaky tests @pakrym-oai - #11386 Prefer websocket transport when model opts in @pakrym-oai - #11373 tui: queue non-pending rollback trims in app-event order @charley-oai - #11393 Remove `deterministic_process_ids` feature to avoid duplicate `codex-core` builds @bolinfest - #11397 fix(exec-policy) No empty command lists @dylan-hurd-oai

Gemini CLIv0.28.1

Release v0.28.1

11. Februar 2026

## Änderungen * fix(patch): cherry-pick 37f128a to release/v0.28.0-pr-18478 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18821 **Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.28.0...v0.28.1

OpenCodev1.1.59

v1.1.59

11. Februar 2026

## Desktop - Add translations support to the application - I don't have access to git commands. Could you provide the full commit message or the PR description for #13166 so I can create an appropriate changelog entry? (@neriousy) - Add more end-to-end tests for the application - Fixed sidebar remounting issue **Thank you to 1 community contributor:** - @neriousy: - Fix/reverception (#13166)

OpenCodev1.1.58

v1.1.58

11. Februar 2026

## TUI - Dismiss dialogs with Ctrl+C in TUI (@kitlangton) - Keep /share command available to copy existing share link (@kitlangton) - Add mode-specific input placeholders to improve context-aware prompts (@kitlangton) - Prevent home wordmark corruption in height-constrained terminals (@kitlangton) ## Desktop - Revert WSL backend mode feature for desktop - Notifications enabled on child sessions - Terminal PTY buffer now properly carries over between sessions - Terminal resize functionality fixed in app **Thank you to 1 community contributor:** - @kitlangton: - fix(tui): prevent home wordmark corruption in height-constrained terminals (#13069) - feat(prompt): mode-specific input placeholders (#12388) - fix(tui): keep /share available to copy existing link (#12532) - fix(tui): dismiss dialogs with ctrl+c (#12884)

OpenCodev1.1.57

v1.1.57

11. Februar 2026

## Core - Add additional context overflow cases and remove overcorrecting ones - Ensure compaction is more reliable by adding reserve token buffer to provide enough room for input window compaction - Improve /review prompt to detect behavior changes more explicitly - Improve Amazon Bedrock check to include container credentials (@OpeOginni) ## TUI - Default session sidebar to auto ## Desktop - Fixed terminal copy/paste functionality - Copy path button styles fixed in app - Use tracing for logging in desktop application (@Brendonovich) - Enhance Windows app resolution and UI loading states (@neriousy) - Improve desktop app server spawn resilience - Read Wayland preference from store in desktop application (@Brendonovich) - Add WSL backend mode for desktop (@Brendonovich) ## SDK - Encode non-ASCII directory paths in v1 SDK HTTP headers (@MrMushrooooom) **Thank you to 9 community contributors:** - @github-actions[bot]: - Update VOUCHED list - @OpeOginni: - fix(tui): improve amazon-bedrock check to include container credentials (#13037) - @ariane-emory: - docs: remove 'Migrating to 1.0' documentation section (#13076) - @Brendonovich: - feat(desktop): add WSL backend mode (#12914) - fix(desktop): read wayland preference from store (#13081) - desktop: use tracing for logging (#13135) - @webwww123: - fix(docs): avoid footer language selector truncation (#13124) - @Seungjun0906: - docs(ko): improve translations for intro, cli, and commands (#13094) - @neriousy: - feat(desktop): enhance Windows app resolution and UI loading states (#13084) - @MrMushrooooom: - fix: encode non-ASCII directory paths in v1 SDK HTTP headers (#13131) - @taroj1205: - fix(web): prevent language select label truncation (#13100)

Claude Codev2.1.39

v2.1.39

10. Februar 2026

## What's changed - Improved terminal rendering performance - Fixed fatal errors being swallowed instead of displayed - Fixed process hanging after session close - Fixed character loss at terminal screen boundary - Fixed blank lines in verbose transcript view

OpenCodev1.1.56

v1.1.56

10. Februar 2026

## Desktop - Task tool rendering fixed in app - Open apps with executables on Windows (@neriousy) - Don't close sidebar on session change **Thank you to 1 community contributor:** - @neriousy: - fix(desktop): open apps with executables on Windows (#13022)

Gemini CLIv0.28.0

Release v0.28.0

10. Februar 2026

## Was sich geändert hat * feat(commands): füge Slash-Befehl /prompt-suggest von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/17264 hinzu * feat(cli): Hooks-enable/disable im Einklang mit Skills bringen und Completion verbessern von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16822 * docs: CLI-Referenzdokumentation hinzufügen von @leochiu-a in https://github.com/google-gemini/gemini-cli/pull/17504 * chore(release): Version auf 0.28.0-nightly.20260128.adc8e11bb erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/17725 * feat(skills): finales stabiles Promotion-Cleanup von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17726 * test(core): Mock-Fetch in OAuth-Transport-Fallback-Tests von @jw409 in https://github.com/google-gemini/gemini-cli/pull/17059 * feat(cli): Auth-Methode in /bug einschließen von @erikus in https://github.com/google-gemini/gemini-cli/pull/17569 * Add a email privacy note to bug_report template von @nemyung in https://github.com/google-gemini/gemini-cli/pull/17474 * Rewind-Dokumentation von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/17446 * fix: MIME-Typen für Audio/Video mit Inhaltsprüfung verifizieren von @maru0804 in https://github.com/google-gemini/gemini-cli/pull/16907 * feat(core): Unterstützung für Positron IDE (#15045) hinzufügen von @kapsner in https://github.com/google-gemini/gemini-cli/pull/15047 * /oncall Dedup - Texte auf nächste Zeilen umbrechen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17782 * fix(admin): fortgeschrittene Features-Verwaltungs-Einstellung umbenennen von @skeshive in https://github.com/google-gemini/gemini-cli/pull/17786 * [extension config] Brechende optionale Werte non-optional machen von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/17785 * Fix docs-writer Skill-Probleme von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/17734 * fix(core): Duplikate von Hook-Fehler-Warnungen während Streaming unterdrücken von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17727 * test: weitere Tests für AskUser hinzufügen von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/17720 * feat(cli): Aktivierung der Aktivitätsprotokollierung im Nicht-Interaktiv-Modus und Eval-Modi von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/17703 * feat(core): Unterstützung für benutzerdefinierte Ablehnungsnachrichten in Richtlinienregeln von @allenhutchison in https://github.com/google-gemini/gemini-cli/pull/17427 * Fix unbeabsichtigte Credential-Exposition gegenüber MCP-Servern von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/17311 * feat(extensions): Unterstützung für benutzerdefinierte Themes in Erweiterungen von @spencer426 in https://github.com/google-gemini/gemini-cli/pull/17327 * fix: Arbeitsbereich-Verzeichnisse bei Sitzungs-Wiederaufnahme speichern und wiederherstellen von @korade-krushna in https://github.com/google-gemini/gemini-cli/pull/17454 * Aktualisiere Release-Notes-Seiten für 0.26.0 und 0.27.0-preview. von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/17744 * feat(ux): aktualisiere Zellengrenzfarbe und erstelle Testdatei für Tabellen-Rendering von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/17798 * Ändere Höhe für die ToolConfirmationQueue von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/17799 * feat(cli): Benutzeridentitätsinformationen in stats-Befehl aufnehmen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17612 * fix(ux): Off-by-Some-Wrap-Probleme durch festbreite Zeichen behoben von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/17816 * feat(cli): undo/redo Tastenkürzel auf Cmd+Z/Alt+Z und Shift+Cmd+Z/Shift+Alt+Z aktualisieren von @scidomino in https://github.com/google-gemini/gemini-cli/pull/17800 * fix(evals): absoluten Pfad für Aktivitäts-Protokollverzeichnis verwenden von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/17830 * test: Integrations-Test hinzufügen, um stdout/stderr-Routing zu überprüfen von @ved015 in https://github.com/google-gemini/gemini-cli/pull/17280 * fix(cli): gelistete Erweiterungen anzeigen, wenn Update-Ziel fehlt von @tt-a1i in https://github.com/google-gemini/gemini-cli/pull/17082 * fix(cli): PAT-Token und Anmeldeinformationen in Git-Remote-URL-Parsing behandeln von @afarber in https://github.com/google-gemini/gemini-cli/pull/14650 * fix(core): returnDisplay für Fehlerergebnisanzeige verwenden von @Nubebuster in https://github.com/google-gemini/gemini-cli/pull/14994 * Fix Erkennung von bun als Paket-Manager von @Randomblock1 in https://github.com/google-gemini/gemini-cli/pull/17462 * feat(cli): hooksConfig.enabled im Einstellungsdialog anzeigen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17810 * feat(cli): Anzeige der Benutzeridentität (auth, E-Mail, Stufe) beim Start von @yunaseoul in https://github.com/google-gemini/gemini-cli/pull/17591 * fix: ghost border für AskUserDialog verhindern von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/17788 * docs: A2A-Subagenten in subagents.md als experimentell markieren von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/17863 * Fehler bei sensiblen Werten für MCP-Serverbehandlung beheben von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/17826 * fix(admin): secureModeEnabled auf strictModeDisabled umbenennen von @skeshive in https://github.com/google-gemini/gemini-cli/pull/17789 * feat(ux): truncate-Dots in Tabellen kürzer machen von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/17825 * fix(core): DEP0040 Punnycode-Demigration via patch-package lösen von @ATHARVA262005 in https://github.com/google-gemini/gemini-cli/pull/17692 * feat(plan): generische Checklist-Komponente erstellen und Todo refaktorisieren von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/17741 * Bereinigung nach Entfernen des Delegates an Agent von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/17875 * fix(core): GIT_CONFIG_GLOBAL verwenden, um Schatten-Git-Repo-Konfiguration zu isolieren - Fixes #17877 von @cocosheng-g in https://github.com/google-gemini/gemini-cli/pull/17803 * Mouse-Tracking E2E deaktivieren von @alisa-alisa in https://github.com/google-gemini/gemini-cli/pull/17880 * fix(cli): korrekten Setting-Key für Cloud Shell-Auth verwenden von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17884 * chore: IDE-spezifisches ASCII-Logo zurücksetzen von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/17887 * Revert "fix(core): resolve DEP0040 punycode deprecation via patch-package" von

OpenCodev1.1.55

v1.1.55

10. Februar 2026

## Core - Increase test timeout to 30s to prevent failures during package installation - Memory leak fixed in platform fetch for events - Show helpful message when free usage limit is exceeded ## Desktop - Disable terminal transparency

Gemini CLIv0.27.4

Release v0.27.4

10. Februar 2026

## Was sich geändert hat * fix(patch): cherry-pick 37f128a to release/v0.27.3-pr-18478 [CONFLICTS] by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18743 **Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.27.3...v0.27.4

OpenCodev1.1.54

v1.1.54

10. Februar 2026

## Core - No changes in review pane - Fix terminal replay in application - Fix workspace reset issues in the app - Resolve Homebrew upgrade requiring multiple runs (@GeneCodeSavvy) - Publish session.error event for invalid model selection (@surma) - Create file if it doesn't exist when writing via ACP (@BryceRyan) - Adjust agent variant logic to check if variant is available for model instead of requiring exact match - Add new ContextOverflowError type - Remove obsolete Copilot model enablement instructions (@fgonzalezurriola) - Enable thinking for all reasoning models on Alibaba Cloud (DashScope) (@AdJIa) - Parse mid-stream OpenAI response errors to prevent infinite retries on unrecoverable errors - Revert web input focus shortcut feature (@gigamonster256) - Add web input focus shortcut (@ChangeHow) - Add models.dev schema reference for model autocomplete in opencode.json (@remorses) - Set variant in assistant messages (@shantur) - Add skill discovery from URLs via well-known RFC - Clean up orphaned worktree directories (@maharshi365) - Properly encode file URLs with special characters (@yudgnahk) - SessionPrompt.shell() now triggers loop if messages are queued (@goniz) - Use reasoning summary auto for GPT-5 models that are not chat (@scratchmex) - Add specific system prompt for Trinity model (@mariamjabara) - Correct prefix selection for amazon-bedrock provider in getSmallModel (@NachoFLizaur) - Don't rely on metadata.summary in task tool render - Handle step-start and step-finish parts in GitHub response text extraction (@elithrar) - Bump @gitlab/gitlab-ai-provider to 3.5.0 (@vglafirov) - Add directory parameter to plugin client for multi-project support (@cooooooooooode) - Correct /data API usage and data format for importing share URLs (@yuvrajvirk) - Parallelize skill downloads for faster loading in TUI - Add skill discovery from URLs via well-known RFC - Handle dollar sign character with file pattern in configuration (@hstove) ## TUI - Clean up dialog-model.tsx per code review - Revert addition of version to session header and status dialog - Revert showing connected providers in /connect dialog - Use sender color for queued messages in TUI (@mcostasilva) - Revert footer restoration in session view - Add Claude Code-style --fork flag to duplicate sessions before continuing (@ariane-emory) - Restore footer to session view (@ariane-emory) - Increase skill dialog width - Improve skills dialog readability (@kynnyhsap) ## Desktop - Back to platform fetch for now - Include basic authentication in app - Fixed global event default fetch behavior in app - Fixed memory leak in event fetch operation - Don't dispose instance after reset workspace - Based on the commit message "fix(app): regressions" without access to the actual changes, I can only provide a generic summary: Fix regressions in the app - Use agent configured variant in app - Persist defensiveness in app - Add redo and undo end-to-end test (@neriousy) - Use absolute paths for sidebar session navigation (@riftzen-bit) - New session in workspace now correctly uses the selected workspace - Add Cmd+[/] keybinds for session history navigation (@kitlangton) - Correct module name for linux_display in main.rs (@Hona) - Set maximum widths in app - Incorrect workspace on new session - Update tab file contents on change - Polish Open in icon treatment - Add fallback for when crypto.randomUUID is unavailable - Allow creating sessions on touch devices (@kitlangton) - Add native Wayland toggle on Linux desktop (@IsraelAraujo70) - Include sandboxes in project unseen message and error notifications (@Brendonovich) - Persist current sidecar URL in state when isSidecar prop is true (@OpeOginni) - Track current sidecar URL in desktop application (@Brendonovich) - Exclude devtools from production builds (@sneycampos) - Add isSidecar prop to AppInterface and persist sidecar server URLs (@OpeOginni) - Localize "close tab" in command palette (@alexyaroshuk) - Improve workspace header truncation and item interaction (@kitlangton) - Add default clipboard copy affordance to TextField component (@kitlangton) - Fix stale context in prompt input - Tighten slash command autocomplete matching (@kitlangton) - Display toast notification when session is missing on prompt submit (@DNGriffin) - Add native clipboard image paste and fix text paste in desktop app (@invarrow) - Respect terminal toggle keybind when terminal is focused (@ryanmiville) - Keep /share available to copy existing link (@kitlangton) - Disable 3 Safari prompt-input annoyances (@DNGriffin) - Add drag-and-drop support for @mentioning files in the app (@DNGriffin) - Handle Windows paths in frontend file URL encoding (@yudgnahk) - Move workspace New session button into header (@kitlangton) - Toggle file tree and review panel with improved UX (@ProdigyRahul) - Allow agent select to use full width on Windows (@abdiths) - Keep startup script field scrollable in edit project dialog (@itskritix) - Fix terminal replay issues in app - Added macOS support for displaying only installed editors and added Sublime Text editor (@OpeOginni) - Remove extra error page and use default error boundary - Add loading window and restructure Rust desktop code (@Brendonovich) - Hide 'open in app' button on narrow viewports - Update server removal logic to clear default server URL if removed (@OpeOginni) - Display session last updated time in command palette search (@alexyaroshuk) - Add Windows File Explorer icon for session header (@maharshi365) - Support desktop titlebar double-click maximize (@crob19) - Add keyboard accelerators to menu items (@Brendonovich) - Add more basic menu bar items to desktop application (@Brendonovich) - Always show project menu button on mobile for accessibility (@DNGriffin) - Maximize main window by default (@Brendonovich) **Thank you to 44 community contributors:** - @kynnyhsap: - fix(opencode): improve skills dialog readability (#12356) - @hstove: - fix(config): handle $ character with {file:} pattern (#12390) - @AksharP5: - feat(tui): highlight esc label on hover in dialog (#12383) - @Brendonovich: - desktop: maximize main window by default (#12433) - desktop: add more basic menu bar items - desktop: add key accelerators to menu itms - desktop: add loading window and restructure rust (#12176) - desktop: track currentSidecarUrl - app: include sandboxes in project unseen/error notifs - @DNGriffin: - fix(app): always show project menu button for mobile a11y (#11258) - feat(app): drag-n-drop to @mention file (#12569) - fix(app): disable 3 safari prompt-input annoyances (#12558) - fix(app): Toast when session is missing on prompt-submit (#12654) - @crob19: - fix(desktop): support desktop titlebar double-click maximize (#12459) - @maharshi365: - fix(ui): add Windows File Explorer icon for session header (#12386) - fix(opencode): cleanup orphaned worktree directories (#12399) - @alexyaroshuk: - feat(app): session last updated time display in command pallete's search (#12376) - fix(app): localize "close tab" in command pallete (#12756) - @OpeOginni: - fix(desktop): update server removal logic to clear default server URL if removed (#12372) - feat(desktop): added Macos support for displaying only installed editors & added sublime text editor (#12501) - fix(docs-windows-wsl): update caution note for server security (#12467) - feat(desktop): add isSidecar prop to AppInterface and logic to persist sidecar server urls (#12366) - feat(desktop): persist currentSidecarUrl in state when isSidecar prop is true (#12792) - @yuvrajvirk: - fix: correct /data API usage and data format for importing share URLs (#7381) - @cooooooooooode: - fix: add directory parameter to plugin client for multi-project support (#11344) - @vglafirov: - chore(deps): bump @gitlab/gitlab-ai-provider to 3.5.0 (#12496) - @elithrar: - fix(github): handle step-start/step-finish parts in extractResponseText (#12470) - @ariane-emory: - feat(tui): restore footer to session view (#12245) - feat(tui): add Claude Code-style --fork flag to duplicate sessions before continuing (resolves #11137) (#11340) - docs(cli): add documentation for --fork flag (#12561) - @NachoFLizaur: - fix(opencode): correct prefix selection for amazon-bedrock provider in getSmallModel (#12281) - @mariamjabara: - feat: add specific system prompt for Trinity model (#12144) - @scratchmex: - feat(opencode): use reasoning summary auto for gpt-5 models that are not chat (#12502) - @dbpolito: - feat(desktop): Session Review Images (#12360) - @goniz: - fix(opencode): SessionPrompt.shell() now triggers loop if messages are queued (#10987) - @itskritix: - fix(app): keep startup script field scrollable in edit project dialog (#12431) - @abdiths: - fix(desktop): allow agent select to use full width on windows (#12428) - @yudgnahk: - fix: properly encode file URLs with special characters (#12424) - fix(app): handle Windows paths in frontend file URL encoding (#12601) - @shantur: - feat(core): Set variant in assistant messages too (#12531) - @remorses: - feat: add models.dev schema ref for model autocomplete in opencode.json (#12528) - @ProdigyRahul: - fix(app): toggle file tree and review panel better ux (#12481) - @kitlangton: - fix(app): move workspace New session into header (#12624) - fix(web): keep /share available to copy existing link (#12533) - refine(app): tighten slash autocomplete matching (#12647) - ui: default TextField copy affordance to clipboard (#12714) - fix(layout): improve workspace header truncation and item interaction (#12655) - fix(app): allow creating sessions on touch devices (#12765) - feat(app): add Cmd+[/] keybinds for session history navigation (#12880) - @ChangeHow: - feat(app): add web input focus shortcut (#12493) - @gigamonster256: - fix: revert "feat(app): add web input focus shortcut (#12493)" (#12639) - feat(nix): disable build time models.dev fetching (#12644) - feat(nix): expose overlay for downstream use (#12643) - @ryanmiville: - fix(app): respect terminal toggle keybind when terminal is focused (#12635) - @invarrow: - fix(desktop): add native clipboard image paste and fix text paste (#12682) - @jerome-benoit: - refactor(nix): use native Bun APIs and propagate errors (#12694) - fix(nix): watch scripts in nix-hashes workflow (#12818) - fix(nix): restore install script in fileset for desktop build (#12842) - @AdJIa: - fix: enable thinking for all reasoning models on alibaba-cn (DashScope) (#12772) - @fgonzalezurriola: - fix(provider): remove obsolete copilot model enablement instructions (#12739) - @sneycampos: - feat: exclude devtools from production builds (#12290) - @IsraelAraujo70: - feat(desktop): add native Wayland toggle on Linux (#11971) - @jcampuza: - fix(app): make keyboard focus visible in settings (#12612) - @BryceRyan: - fix(opencode): ACP File write should create the file if it doesn't exist (#12854) - @Hona: - fix(desktop): correct module name for linux_display in main.rs (#12862) - @surma: - fix: publish session.error event for invalid model selection (#8451) - @mcostasilva: - fix(tui): use sender color for queued messages (#12832) - @github-actions[bot]: - Update VOUCHED list - @GeneCodeSavvy: - fix: resolve homebrew upgrade requiring multiple runs (#5375) (#10118) - @riftzen-bit: - fix: use absolute paths for sidebar session navigation (#12898) - @neriousy: - test(e2e): redo & undo test (#12974)

Claude Codev2.1.38

v2.1.38

10. Februar 2026

## What's changed - Fixed VS Code terminal scroll-to-top regression introduced in 2.1.37 - Fixed Tab key queueing slash commands instead of autocompleting - Fixed bash permission matching for commands using environment variable wrappers - Fixed text between tool uses disappearing when not using streaming - Fixed duplicate sessions when resuming in VS Code extension - Improved heredoc delimiter parsing to prevent command smuggling - Blocked writes to `.claude/skills` directory in sandbox mode

OpenClawv2026.2.9

openclaw 2026.2.9

9. Februar 2026

### Added - iOS: alpha node app + setup-code onboarding. (#11756) Thanks @mbelinky. - Channels: umfassende BlueBubbles- und Channel-Aufräumarbeiten. (#11093) Thanks @tyler6204. - Plugins: Geräte-Pairing + Telefonsteuerungs-Plugins (Telegram `/pair`, iOS/Android Node-Steuerung). (#11755) Thanks @mbelinky. - Tools: Grok (xAI) als `web_search`-Provider hinzufügen. (#12419) Thanks @tmchow. - Gateway: API-Methoden zur Agentenverwaltung für die Web UI hinzufügen (`agents.create`, `agents.update`, `agents.delete`). (#11045) Thanks @advaitpaliwal. - Web UI: in der Chat-Historie einen Compaction-Trenner anzeigen. (#11341) Thanks @Takhoffman. - Agents: Laufzeit-Shell in Agentenumschlägen einbeziehen. (#1835) Thanks @Takhoffman. - Paths: `OPENCLAW_HOME` hinzufügen, um das Home-Verzeichnis für die interne Pfadauflösung zu überschreiben. (#12091) Thanks @sebslight. ### Fixes - Telegram: Quote-Parsing abgesichert; Zitierkontext beibehalten; QUOTE_TEXT_INVALID vermeiden; falsche Kategorisierung verschachtelter Reply-Zitate vermeiden. (#12156) Thanks @rybnikov. - Telegram: proaktive Sends wiederherstellen, wenn veraltete Topic-Thread-IDs verwendet werden, durch erneutes Senden ohne `message_thread_id`. (#11620) - Telegram: Markdown-Spoiler mit `<tg-spoiler>` HTML-Tags rendern. (#11543) Thanks @ezhikkk. - Telegram: Registrierung von Befehlsregistrierungen auf 100 Einträge beschränken, um `BOT_COMMANDS_TOO_MUCH`-Fehler beim Start zu vermeiden. (#12356) Thanks @arosstale. - Telegram: übereinstimmende DM-`allowFrom` gegen Sender-Benutzer-ID (Fallback auf Chat-ID) und Logs zum Pairing klären. (#12779) Thanks @liuxiaopai-ai. - Onboarding: QuickStart installiert Shell-Completion automatisch (Prompt nur im Manual). - Auth: eingebettete Zeilenumbrüche aus eingefügten API-Schlüsseln und Tokens entfernen, bevor Anmelde-/Credentialen gespeichert/auflöst werden. - Web UI: Chat-Aktualisierung sanft zum neuesten Nachrichten-Stand scrollen und Flashen des New-Messages-Badges während manueller Aktualisierung unterdrücken. - Tools/web_search: plattformspezifische Einstellungen in den Web-Such-Cache-Schlüssel aufnehmen und `inlineCitations` für Grok übergeben. (#12419) Thanks @tmchow. - Tools/web_search: direkte Perplexity-Modell-IDs normalisieren, OpenRouter-Modell-IDs unverändert lassen. (#12795) Thanks @cdorsey. - Model Failover: HTTP-400-Fehler als failover-fähig behandeln, automatisches Modell-Fallback ermöglichen. (#1879) Thanks @orenyomtov. - Errors: falsche Positiv-Erkennung von Kontextüberlauf vermeiden, wenn Unterhaltung das Topic „context overflow“ erwähnt. (#2078) Thanks @sbking. - Gateway: kein Post-Compaction-Amnesie mehr; injizierte Transkript-Schreibvorgänge bewahren nun die Pi-Sitzungs-`parentId`-Kette, damit Agenten sich wieder erinnern können. (#12283) Thanks @Takhoffman. - Gateway: Multi-Agent-Sessions.Usage-Erkennung fixen. (#11523) Thanks @Takhoffman. - Agents: aus Kontextüberlauf durch zu große Tool-Ergebnisse wiederherstellen (vorausschauendes Begrenzen + Fallback-Trunkierung). (#11579) Thanks @tyler6204. - Subagents/compaction: Timing der Ankündigungen stabilisieren und Kompaktion-Metriken über Wiederholungen hinweg erhalten. (#11664) Thanks @tyler6204. - Cron: isolierte Ankündigungs-Flow teilen und Zuverlässigkeit von Scheduling/Lieferung erhöhen. (#11641) Thanks @tyler6204. - Cron-Tool: flache Parameter wiederherstellen, wenn LLM das `job`-Wrapper für Add-Anfragen auslässt. (#12124) Thanks @tyler6204. - Gateway/CLI: bei `gateway.bind=lan` eine LAN-IP für Proben-URLs und Control-UI-Links verwenden. (#11448) Thanks @AnonO6. - Hooks: gebündelte Hooks seit 2026.2.2 reparieren (tsdown-Migration). (#9295) Thanks @patrickshao. - Routing: Bindings pro Nachricht aktualisieren durch Laden der Config bei der Routenzuordnung, damit Änderungen ohne Neustart wirksam werden. (#11372) Thanks @juanpablodlc. - Exec-Bewertungen: weitergeleitete Befehle in Monospace rendern für sichereres Freigabe-Scanning. (#11937) Thanks @sebslight. - Config: `maxTokens` an `contextWindow` anpassen, um ungültige Modell-Konfigurationen zu verhindern. (#5516) Thanks @lailoo. - Thinking: xhigh für `github-copilot/gpt-5.2-codex` und `github-copilot/gpt-5.2` zulassen. (#11646) Thanks @LatencyTDH. - Discord: Starter-Nachrichten für Foren-/Medien-Threads unterstützen, `message thread create --message` verarbeiten und Routing härten. (#10062) Thanks @jarvis89757. - Paths: strukturiertes Auflösen von OPENCLAW_HOME-abgeleiteten Home-Pfaden und Windows-Laufwerksbuchstaben-Behandlung bei Tool-Meta-Verkürzungen korrigieren. (#12125) Thanks @mcaxtr. - Memory: Voyage-Embeddings `input_type` für verbesserte Abfrage. (#10818) Thanks @mcinteerj. - Memory/QMD: Standard-Modell-Cache über Agents hinweg wiederverwenden statt pro Agent erneut herunterladen. (#12114) Thanks @tyler6204. - Media Understanding: `.caf`-Audioanhänge für Transkription erkennen. (#10982) Thanks @succ985. - State dir: `OPENCLAW_STATE_DIR` für Standard-Geräte-Identität und Canvas-Speicherpfade berücksichtigen. (#4824) Thanks @kossoy.

Claude Codev2.1.37

v2.1.37

7. Februar 2026

## What's changed - Fixed an issue where /fast was not immediately available after enabling /extra-usage

Claude Codev2.1.36

v2.1.36

7. Februar 2026

## What's changed - Fast mode is now available for Opus 4.6. Learn more at https://code.claude.com/docs/en/fast-mode

OpenClawv2026.2.6

openclaw 2026.2.6

7. Februar 2026

### Changes - Models: unterstützt Anthropic Opus 4.6 und OpenAI Codex gpt-5.3-codex (Forward-compat-Fallbacks). (#9853, #10720, #9995) Danke an @TinyTb, @calvin-hpnet, @tyler6204. - Providers: xAI (Grok) Unterstützung hinzufügen. (#9885) Danke an @grp06. - Web UI: Token-Nutzungs-Dashboard hinzufügen. (#10072) Danke an @Takhoffman. - Memory: Native Voyage AI-Unterstützung. (#7078) Danke an @mcinteerj. - Sessions: cap payloads von sessions_history, um Kontextüberlauf zu reduzieren. (#10000) Danke an @gut-puncture. - CLI: Befehle alphabetisch in der Help-Ausgabe sortieren. (#8068) Danke an @deepsoumya617. - Agents: pi-mono auf 0.52.7 erhöhen; eingebettete Forward-Compat-Fallback für Opus 4.6 Modell-IDs hinzufügen. ### Fixes - Telegram: DM-Topic-threadId automatisch in Nachrichtentool + Subagent-Ankündigung einfügen. (#7235) Danke an @Lukavyi. - Security: Auth für Gateway-Canvas-Host und A2UI-Assets erforderlich machen. (#9518) Danke an @coygeek. - Cron: Scheduling- und Erinnerungslieferungs-Regressionen beheben; Next-Run-Neuberechnung + Timer-Nearmung + Legacy-Scheduling-Felder härten. (#9733, #9823, #9948, #9932) Danke an @tyler6204, @pycckuu, @j2h4u, @fujiwara-tofu-shop. - Update: Control UI-Asset-Handling im Update-Fluss härten. (#10146) Danke an @gumadeiras. - Security: Skill/Plugin-Code-Safety-Scanner hinzufügen; Anmeldeinformationen aus config.get Gateway-Antworten redigieren. (#9806, #9858) Danke an @abdelsfane. - Exec Approvals: Bare-String-Whitelist-Einträge in Objekte konvertieren. (#9903) Danke an @mcaxtr. - Slack: Erwähnungs-StripPatterns für /new und /reset hinzufügen. (#9971) Danke an @ironbyte-rgb. - Chrome-Erweiterung: Bündelpfad-Auflösung korrigieren. (#8914) Danke an @kelvinCB. - Kompaktion/Fehler: Mehrere Kompaktionsversuche bei Kontext-Overflow zulassen; klare Abrechnungsfehler anzeigen. (#8928, #8391) Danke an @Glucksberg.

Gemini CLIv0.27.3

Release v0.27.3

6. Februar 2026

## Was sich geändert hat * fix(patch): cherry-pick 1cae5ab to release/v0.27.2-pr-18376 to patch version v0.27.2 and create version 0.27.3 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18464 **Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.27.2...v0.27.3

Claude Codev2.1.34

v2.1.34

6. Februar 2026

## What's changed - Fixed a crash when agent teams setting changed between renders - Fixed a bug where commands excluded from sandboxing (via `sandbox.excludedCommands` or `dangerouslyDisableSandbox`) could bypass the Bash ask permission rule when `autoAllowBashIfSandboxed` was enabled

Claude Codev2.1.33

v2.1.33

6. Februar 2026

## What's changed - Fixed agent teammate sessions in tmux to send and receive messages - Fixed warnings about agent teams not being available on your current plan - Added `TeammateIdle` and `TaskCompleted` hook events for multi-agent workflows - Added support for restricting which sub-agents can be spawned via `Task(agent_type)` syntax in agent "tools" frontmatter - Added `memory` frontmatter field support for agents, enabling persistent memory with `user`, `project`, or `local` scope - Added plugin name to skill descriptions and `/skills` menu for better discoverability - Fixed an issue where submitting a new message while the model was in extended thinking would interrupt the thinking phase - Fixed an API error that could occur when aborting mid-stream, where whitespace text combined with a thinking block would bypass normalization and produce an invalid request - Fixed API proxy compatibility issue where 404 errors on streaming endpoints no longer triggered non-streaming fallback - Fixed an issue where proxy settings configured via `settings.json` environment variables were not applied to WebFetch and other HTTP requests on the Node.js build - Fixed `/resume` session picker showing raw XML markup instead of clean titles for sessions started with slash commands - Improved error messages for API connection failures — now shows specific cause (e.g., ECONNREFUSED, SSL errors) instead of generic "Connection error" - Errors from invalid managed settings are now surfaced - VSCode: Added support for remote sessions, allowing OAuth users to browse and resume sessions from claude.ai - VSCode: Added git branch and message count to the session picker, with support for searching by branch name - VSCode: Fixed scroll-to-bottom under-scrolling on initial session load and session switch

OpenCodev1.1.53

v1.1.53

5. Februar 2026

## Core - Load user plugins after built-in plugins - Fix unhandled errors when aborting with queued messages - User plugins override built-in plugins for the same provider (@rmk40) - Move Codex 5.3 model definition to plugin to avoid showing unsupported model to other users - Add session usage tracking to ACP (@SteffenDE) - Update transforms for GPT-5.3 ## TUI - Allow mouse escape via "esc" labels in dialogs (@AksharP5) ## Desktop - Make close comment button visible in prompt input (@alexyaroshuk) - Hide prompt input when there are permissions requests or questions - More terminal stability fixes - Modified file color contrast in app for better visibility - Add button to open files in external applications - Allow toggling file tree closed independently - Stop showing SessionSkeleton on new workspace (@dbpolito) - Set workspace name earlier to improve creation and deletion (@dbpolito) **Thank you to 7 community contributors:** - @SteffenDE: - feat(acp): add session usage (#12299) - @dbpolito: - feat(desktop): Set Workspace Name Earlier to Improve Creation / Deletion (#12213) - feat(desktop): Stop Showing SessionSkeleton on New Workspace (#12209) - @alexyaroshuk: - fix(app): make close comment button visible in prompt input (#12349) - @edoedac0: - docs: add Bosnian README translation (#12341) - @ariane-emory: - docs: websearch tool (#12359) - @rmk40: - fix(plugin): user plugins override built-in plugins for same provider (#12361) - @AksharP5: - fix(tui): allow mouse escape via "esc" labels in dialogs (#11421)

Gemini CLIv0.27.2

Release v0.27.2

5. Februar 2026

## Änderungen * fix(patch): cherry-pick 9ca7300 to release/v0.27.1-pr-18395 to patch version v0.27.1 and create version 0.27.2 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18399 **Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.27.1...v0.27.2

Gemini CLIv0.27.1

Release v0.27.1

5. Februar 2026

## Was sich geändert hat * fix(patch): cherry-pick a8b4c38 to release/v0.27.0-pr-18161 to patch version v0.27.0 and create version 0.27.1 by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/18393 **Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.27.0...v0.27.1

OpenCodev1.1.52

v1.1.52

5. Februar 2026

## Core - Enable Claude 3.5 Sonnet (new) model support - Silently ignore proxy command failures to prevent config initialization crashes - Ensure GitHub Copilot plugin properly sets headers when used in clients other than TUI - Bundle GitLab auth plugin directly instead of dynamic install - Fix plugin installation to use direct package.json manipulation instead of bun add - Fix image reading with OpenAI-compatible providers like Kimi K2.5 (@zhming0) - Downgrade xai ai-sdk package due to errors - Revert model autocomplete feature using models.dev schema reference - Add models.dev schema reference for model autocomplete in opencode.json (@remorses) - Adjust task tool description and input to reduce tool call failures with GPT models - Wait for dependencies before loading custom tools and plugins - Allow the function to hide or show thinking blocks to be bound to a key (@ariane-emory) - Skip dependency installation in read-only configuration directories (@shantur) - Ensure Kimi for Coding plan has thinking enabled by default for k2p5 (@monotykamary) - Fixed Cloudflare Workers AI provider - Prevent random hangs in plugin installs when using HTTP proxy by adding --no-cache flag - Session errors when attachment file not found are now handled gracefully - Support remote server connections in terminal and fix GLIBC compatibility (@lucas-jo) ## TUI - Add running spinner to bash tool in TUI (@goniz) - Add hover states to question tool tabs (@maharshi365) ## Desktop - File changes not always available in app - File tree kept in sync with filesystem changes - Add Bosnian locale (@edoedac0) - Fix terminal URL handling issues - Remove extra horizontal padding around prompt input on mobile (@Brendonovich) - Refresh workspace sessions when switching projects (@neriousy) - Fixed terminal URL handling in the application - Fix terminal end-of-line handling issues - Refresh file contents when changing workspaces to prevent stale data (@ParkerSm1th) - Derive terminal WebSocket URL from browser origin instead of hardcoded localhost (@0xdsqr) - Last turn changes rendered in review pane - Safety triangle for sidebar hover to prevent menu from closing - Clear comments on prompt submission - Fix e2e test action in app - Terminal hyperlink clicks now work correctly - Fix dated e2e tests in app - Don't show scroll-to-bottom button unnecessarily - File tree not staying in sync across projects/sessions - Move session options to the session page - Add session options to app - Opened tabs follow created session - Removed compression from RPM bundle to save 15 minutes in CI (@goniz) **Thank you to 13 community contributors:** - @lucas-jo: - fix(terminal): support remote server connections and fix GLIBC compatibility (#11906) - @goniz: - fix(desktop): removed compression from rpm bundle to save 15m in CI (#12097) - feat(tui): add running spinner to bash tool in TUI (#12317) - @monotykamary: - fix: ensure kimi-for-coding plan has thinking on by default for k2p5 (#12147) - @0xdsqr: - fix(app): derive terminal WebSocket URL from browser origin instead o… (#12178) - @ParkerSm1th: - fix(desktop): Refresh file contents when changing workspaces to not have stale contents (#11728) - @shantur: - fix(core): skip dependency install in read-only config dirs (#12128) - @maharshi365: - fix(tui): add hover states to question tool tabs (#12203) - @ariane-emory: - feat: Allow the function to hide or show thinking blocks to be bound to a key (resolves #12168) (#12171) - @neriousy: - fix(app): refresh workspace sessions on project switch (#12189) - @remorses: - feat: add models.dev schema ref for model autocomplete in opencode.json (#12112) - @zhming0: - fix(opencode): Fixes image reading with OpenAI-compatible providers like Kimi K2.5. (#11323) - @Brendonovich: - app: remove extra x padding around prompt input on mobile - @edoedac0: - feat(i18n): add Bosnian locale (#12283)

Claude Codev2.1.32

v2.1.32

5. Februar 2026

## What's changed - Claude Opus 4.6 is now available! - Added research preview agent teams feature for multi-agent collaboration (token-intensive feature, requires setting CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1) - Claude now automatically records and recalls memories as it works - Added "Summarize from here" to the message selector, allowing partial conversation summarization. - Skills defined in `.claude/skills/` within additional directories (`--add-dir`) are now loaded automatically. - Fixed `@` file completion showing incorrect relative paths when running from a subdirectory - Updated --resume to re-use --agent value specified in previous conversation by default. - Fixed: Bash tool no longer throws "Bad substitution" errors when heredocs contain JavaScript template literals like `${index + 1}`, which previously interrupted tool execution - Skill character budget now scales with context window (2% of context), so users with larger context windows can see more skill descriptions without truncation - Fixed Thai/Lao spacing vowels (สระ า, ำ) not rendering correctly in the input field - VSCode: Fixed slash commands incorrectly being executed when pressing Enter with preceding text in the input field - VSCode: Added spinner when loading past conversations list

Codex CLIvrust-v0.98.0

0.98.0

5. Februar 2026

## New Features - Introducing GPT-5.3-Codex. [Learn More](https://openai.com/index/introducing-gpt-5-3-codex/) - Steer mode is now stable and enabled by default, so `Enter` sends immediately during running tasks while `Tab` explicitly queues follow-up input. (#10690) ## Bug Fixes - Fixed `resumeThread()` argument ordering in the TypeScript SDK so resuming with local images no longer starts an unintended new session. (#10709) - Fixed model-instruction handling when changing models mid-conversation or resuming with a different model, ensuring the correct developer instructions are applied. (#10651, #10719) - Fixed a remote compaction mismatch where token pre-estimation and compact payload generation could use different base instructions, improving trim accuracy and avoiding context overflows. (#10692) - Cloud requirements now reload immediately after login instead of requiring a later refresh path to take effect. (#10725) ## Chores - Restored the default assistant personality to Pragmatic across config and related tests/UI snapshots. (#10705) - Unified collaboration mode naming and metadata across prompts, tools, protocol types, and TUI labels for more consistent mode behavior and messaging. (#10666) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.97.0...rust-v0.98.0 - #10709 fix: ensure resume args precede image args @cryptonerdcn - #10705 chore(config) Default Personality Pragmatic @dylan-hurd-oai - #10651 fix(core) switching model appends model instructions @dylan-hurd-oai - #10666 Sync collaboration mode naming across Default prompt, tools, and TUI @charley-oai - #10690 Make steer stable by default @aibrahim-oai - #10692 Fix remote compaction estimator/payload instruction small mismatch @charley-oai - #10725 Reload cloud requirements after user login @xl-openai - #10719 fix(core,app-server) resume with different model @dylan-hurd-oai

Codex CLIvrust-v0.97.0

0.97.0

5. Februar 2026

## New Features - Added a session-scoped “Allow and remember” option for MCP/App tool approvals, so repeated calls to the same tool can be auto-approved during the session. (#10584) - Added live skill update detection, so skill file changes are picked up without restarting. (#10478) - Added support for mixed text and image content in dynamic tool outputs for app-server integrations. (#10567) - Added a new `/debug-config` slash command in the TUI to inspect effective configuration. (#10642) - Introduced initial memory plumbing (API client + local persistence) to support thread memory summaries. (#10629, #10634) - Added configurable `log_dir` so logs can be redirected (including via `-c` overrides) more easily. (#10678) ## Bug Fixes - Fixed jitter in the TUI apps/connectors picker by stabilizing description-column rendering. (#10593) - Restored and stabilized the TUI “working” status indicator/shimmer during preamble and early exec flows. (#10700, #10701) - Improved cloud requirements reliability with higher timeouts, retries, and corrected precedence over MDM settings. (#10631, #10633, #10659) - Persisted pending-input user events more consistently for mid-turn injected input handling. (#10656) ## Documentation - Documented how to opt in to the experimental app-server API. (#10667) - Updated docs/schema coverage for new `log_dir` configuration behavior. (#10678) ## Chores - Added a gated Bubblewrap (`bwrap`) Linux sandbox path to improve filesystem isolation options. (#9938) - Refactored model client lifecycle to be session-scoped and reduced implicit client state. (#10595, #10664) - Added caching for MCP actions from apps to reduce repeated load latency for users with many installed apps. (#10662) - Added a `none` personality option in protocol/config surfaces. (#10688) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.96.0...rust-v0.97.0 - #10595 Stop client from being state carrier @pakrym-oai - #10584 Add option to approve and remember MCP/Apps tool usage @canvrno-oai - #10644 fix: flaky test @jif-oai - #10629 feat: add phase 1 mem client @jif-oai - #10633 Cloud Requirements: take precedence over MDM @gt-oai - #10659 Increase cloud req timeout @gt-oai - #9938 feat(linux-sandbox): add bwrap support @viyatb-oai - #10656 Persist pending input user events @aibrahim-oai - #10634 feat: add phase 1 mem db @jif-oai - #10593 Fix jitter in TUI apps/connectors picker @canvrno-oai - #10662 [apps] Cache MCP actions from apps. @mzeng-openai - #10649 Fix test_shell_command_interruption flake @gt-oai - #10642 Add /debug-config slash command @gt-oai - #10478 Added support for live updates to skills @etraut-openai - #10688 add none personality option @aibrahim-oai - #10567 feat(app-server, core): allow text + image content items for dynamic tool outputs @owenlin0 - #10667 chore(app-server): document experimental API opt-in @owenlin0 - #10664 Session-level model client @pakrym-oai - #10678 feat(core): add configurable log_dir @joshka-oai - #10631 Cloud Requirements: increase timeout and retries @gt-oai - #10650 chore(core) personality migration tests @dylan-hurd-oai - #10701 fix(tui): restore working shimmer after preamble output @joshka-oai - #10700 fix: ensure status indicator present earlier in exec path @sayan-oai

OpenClawv2026.2.3

openclaw 2026.2.3

5. Februar 2026

### Changes - Telegram: letzte `@ts-nocheck` aus `bot-handlers.ts` entfernen, Grammy-Typen direkt verwenden, `StickerMetadata` duplizierungsfrei machen. Keine `@ts-nocheck` mehr in `src/telegram/`. (#9206) - Telegram: `@ts-nocheck` aus `bot-message.ts` entfernen, Typ-Abhängigkeiten via `Omit<BuildTelegramMessageContextParams>`, `allMedia` auf `TelegramMediaRef[]` erweitern. (#9180) - Telegram: `@ts-nocheck` aus `bot.ts` entfernen, doppelten `bot.catch` Fehler-Handler fixen (Grammy-Overrides), Dead-Reaktion `message_thread_id`-Routing entfernen, Sticker-Cache-Guard härten. (#9077) - Onboarding: Cloudflare AI Gateway-Anbieter-Setup und Docs hinzufügen. (#7914) Danke @roerohan. - Onboarding: Moonshot (.cn) Auth-Auswahl hinzufügen und die China-Base-URL beim Beibehalten von Defaults behalten. (#7180) Danke @waynelwz. - Docs: tmux send-keys für TUI klären, Text und Enter trennen. (#7737) Danke @Wangnov. - Docs: die Überarbeitung der Landing-Page für zh-CN spiegeln (Features, Quickstart, docs-Verzeichnis, Netzwerkmodell, Credits). (#8994) Danke @joshp123. - Messages: pro Kanal und pro Konto Overrides für responsePrefix über alle Kanäle hinweg hinzufügen. (#9001) Danke @mudrii. - Cron: Delivery-Modus für Ankündigungs-Jobs (CLI + Control UI) und Delivery-Modus-Konfiguration hinzufügen. - Cron: isolierte Jobs standardmäßig auf Announce-Delivery setzen; ISO 8601 `schedule.at` in Tool-Eingaben akzeptieren. - Cron: isolierte Jobs migrationssicher auf announce/none Delivery migrieren; veraltete post-to-main/payload Delivery-Felder und `atMs`-Eingaben entfernen. - Cron: One-Shot-Jobs nach Erfolg standardmäßig löschen; CLI-Flag `--keep-after-run` hinzufügen. - Cron: Messaging-Tools während Announce-Delivery unterdrücken, damit Zusammenfassungen konsistent posten. - Cron: doppelte Deliveries vermeiden, wenn isolierte Runs Nachrichten direkt senden. ### Fixes - Heartbeat: explizite accountId-Routing für Multi-Account-Kanäle zulassen. (#8702) Danke @lsh411. - TUI/Gateway: Nicht-Streaming-Finals behandeln, History bei Nicht-Local-Chat-Läufen aktualisieren, Warnungen zu Event-Lücken für zielgerichtete Tool-Streams vermeiden. (#8432) Danke @gumadeiras. - Shell-Vervollständigung: langsame dynamische Muster automatisch erkennen und in gecachte Dateien migrieren für schnelleren Terminal-Start; Vervollständigungs-Gesundheitschecks zu Doctor/Update/Onboard hinzufügen. - Telegram: Sitzungs-Modell-Overrides bei Inline-Modell-Auswahl berücksichtigen. (#8193) Danke @gildo. - Web UI: Speichern der Modell-Auswahl für Standard-/Nicht-Standard-Agenten korrigieren und lange Arbeitsbereich-Pfade einwickeln. Danke @Takhoffman. - Web UI: Kopfzeilen-Logo-Pfad lösen, wenn `gateway.controlUi.basePath` gesetzt ist. (#7178) Danke @Yeom-JinHo. - Web UI: Button-Styles auf das neue Nachrichten-Indikator anwenden. - Onboarding: Auth-Auswahl aus Flags für API-Key (nicht interaktiv) ableiten. (#8484) Danke @f-trycua. - Sicherheit: unzuverlässige Kanal-Metadaten aus System-Prompts (Slack/Discord) fernhalten. Danke @KonstantinMirin. - Sicherheit: sandboxespezifische Medienpfade für Nachrichten-Tool-Anhänge erzwingen. (#9182) Danke @victormier. - Sicherheit: explizite Credentials für Gateway-URL-Overrides erforderlich machen, um Credential-Leckage zu verhindern. (#8113) Danke @victormier. - Sicherheit: `whatsapp_login`-Tool nur Besitzern senden und Kontexten von Nicht-Besitzern standardmäßig verweigern. (#8768) Danke @victormier. - Voice Call: Webhook-Überprüfung härten mit Host-Allowlists/Proxy-Vertrauen und Ngrok-Loopback-Bypass beibehalten. - Voice Call: Regression-Abdeckung für anonyme eingehende Anrufer-IDs mit Allowlist-Policy hinzufügen. (#8104) Danke @victormier. - Cron: Epoch-Timestamps und 0ms-Dauern in CLI `--at`-Parsing akzeptieren. - Cron: Store-Daten neu laden, wenn Store-Datei neu erstellt wird oder mtime sich ändert. - Cron: Announce-Läufe direkt liefern, Delivery-Modus beachten und WakeMode für Zusammenfassungen respektieren. (#8540) Danke @tyler6204. - Telegram: Forward-from-chat-Metadaten in weitergeleiteten Nachrichten einbeziehen und harte Cron-Lieferziel-Checks verbessern. (#8392) Danke @Glucksberg. - macOS: Cron-Payload-Zusammenfassung rendering beheben und ISO 8601-Formatter-Konkurrenzsicherheit.

Continue.devv1.2.15-vscode

v1.2.15-vscode

4. Februar 2026

## Änderungen * docs: aktualisiere die Beschreibung des /info-Befehls mit Nutzungsstatistiken von @continue[bot] in https://github.com/continuedev/continue/pull/9071 * chore(deps): bump undici von 7.16.0 auf 7.18.2 in /binary von @dependabot[bot] in https://github.com/continuedev/continue/pull/9534 * fix: füge GH_TOKEN zum continue-agents-Workflow hinzu von @ThanhNguyxn in https://github.com/continuedev/continue/pull/9497 * chore: prerelease-Versionen erhöhen von @RomneyDa in https://github.com/continuedev/continue/pull/9536 * Filter out .md files when loading agent configs von @RomneyDa in https://github.com/continuedev/continue/pull/9537 * feat: subagents von @uinstinct in https://github.com/continuedev/continue/pull/9128 * Update docs index page with new inbox card and copy changes von @BekahHW in https://github.com/continuedev/continue/pull/9533 * docs: Add documentation for running agents locally von @bdougie in https://github.com/continuedev/continue/pull/9507 * feat: agent skills von @uinstinct in https://github.com/continuedev/continue/pull/9353 * chore(deps): bump diff von 7.0.0 auf 8.0.3 in /core von @dependabot[bot] in https://github.com/continuedev/continue/pull/9552 * [Snyk] Upgrade @c15t/react von 1.8.1 auf 1.8.2 von @sestinj in https://github.com/continuedev/continue/pull/9540 * Refactor info command into separate functions and add usage statistics von @sestinj in https://github.com/continuedev/continue/pull/9070 * [Snyk] Upgrade posthog-js von 1.297.2 auf 1.310.1 von @sestinj in https://github.com/continuedev/continue/pull/9541 * Allow MCP/Bash tools by default in headless mode von @RomneyDa in https://github.com/continuedev/continue/pull/9538 * docs: fix title case in documentation headers von @sestinj in https://github.com/continuedev/continue/pull/9543 * fix: simplify config error von @uinstinct in https://github.com/continuedev/continue/pull/9556 * chore(deps): bump undici von 6.21.3 auf 6.23.0 in /core von @dependabot[bot] in https://github.com/continuedev/continue/pull/9565 * fix: duplicate tool messages added von @RomneyDa in https://github.com/continuedev/continue/pull/9596 * [Snyk] Upgrade react-hook-form von 7.62.0 auf 7.69.0 von @sestinj in https://github.com/continuedev/continue/pull/9588 * [Snyk] Upgrade @tiptap/extension-mention von 2.26.1 auf 2.27.1 von @sestinj in https://github.com/continuedev/continue/pull/9589 * fix: verhindern, dass auf das Laden der Sitzung aus dem Verlauf gewartet wird von @uinstinct in https://github.com/continuedev/continue/pull/9525 * fix: vollständige rechte Seiten-Verlaufsgradiensatz anzeigen von @uinstinct in https://github.com/continuedev/continue/pull/9672 * docs: IDE Extensions-Tab mit Customize- und Reference-Abschnitten erstellen von @sestinj in https://github.com/continuedev/continue/pull/9539 * Revert "docs: create IDE Extensions tab with Customize and Reference sections" von @sestinj in https://github.com/continuedev/continue/pull/9680 * Revert "Revert "docs: create IDE Extensions tab with Customize and Reference sections"" von @sestinj in https://github.com/continuedev/continue/pull/9681 * fix: Navigationsprobleme des Docs-Tabs lösen von @sestinj in https://github.com/continuedev/continue/pull/9682 * fix: use ide.runCommand, wenn Windows-Host mit WSL verbindet ist von @shanevcantwell in https://github.com/continuedev/continue/pull/9679 * fix: kompakte Unterhaltung mit schwebenden Tool-Aufrufen von @uinstinct in https://github.com/continuedev/continue/pull/9465 * feat: überladene Fehler erneut senden von @uinstinct in https://github.com/continuedev/continue/pull/9082 * feat(cli): agent skills von @uinstinct in https://github.com/continuedev/continue/pull/9696 * Add extension features and customization links to docs index von @sestinj in https://github.com/continuedev/continue/pull/9798 * fix: Befehl im contributing-file festlegen von @sarthakNITT in https://github.com/continuedev/continue/pull/9715 * fix: Autocompaction-Fehler durch überschrittene Kontextlänge von @RomneyDa in https://github.com/continuedev/continue/pull/9567 * chore(deps-dev): bump lodash-es von 4.17.21 auf 4.17.23 in /packages/config-yaml von @dependabot[bot] in https://github.com/continuedev/continue/pull/9773 * chore(deps-dev): bump lodash-es von 4.17.21 auf 4.17.23 in /packages/openai-adapters von @dependabot[bot] in https://github.com/continuedev/continue/pull/9775 * chore(deps-dev): bump lodash-es von 4.17.21 auf 4.17.23 in /extensions/cli von @dependabot[bot] in https://github.com/continuedev/continue/pull/9774 * chore(deps-dev): bump lodash von 4.17.21 auf 4.17.23 von @dependabot[bot] in https://github.com/continuedev/continue/pull/9781 * [Snyk] Upgrade yaml von 2.8.1 auf 2.8.2 von @sestinj in https://github.com/continuedev/continue/pull/9653 * feat: add remove rule button von @uinstinct in https://github.com/continuedev/continue/pull/9576 * Continue CLI support provider Ask Sage von @alex-mcgraw-askSage in https://github.com/continuedev/continue/pull/9719 * chore(deps-dev): bump lodash-es von 4.17.21 auf 4.17.23 in /packages/fetch von @dependabot[bot] in https://github.com/continuedev/continue/pull/9772 * Update asksage.mdx von @alex-mcgraw-askSage in https://github.com/continuedev/continue/pull/9757 * chore(deps): bump lodash von 4.17.21 auf 4.17.23 in /extensions/cli von @dependabot[bot] in https://github.com/continuedev/continue/pull/9802 * fix: Unterstützung von Objekt-Typ-Argumenten in der Tool-Aufrufanalyse von @silencezhoudev in https://github.com/continuedev/continue/pull/9743 * [Snyk] Upgrade @tiptap/extension-paragraph von 2.26.1 auf 2.27.1 von @sestinj in https://github.com/continuedev/continue/pull/9817 * chore(deps-dev): bump lodash von 4.17.21 auf 4.17.23 in /docs von @dependabot[bot] in https://github.com/continuedev/continue/pull/9803 * chore(deps): bump lodash von 4.17.21 auf 4.17.23 in /gui von @dependabot[bot] in https://github.com/continuedev/continue/pull/9830 * fix: Set isComplete-Metadaten, wenn Agents die Ausführung beenden von @sestinj in https://github.com/continuedev/continue/pull/9892 * Fix: Set isComplete=true nach Agenten-Runde endet ohne Tool-Aufrufe von @sestinj in https://github.com/continuedev/continue/pull/9899 * feat(cli): detect wsl und spawn entsprechende Shell von @uinstinct in https://github.com/continuedev/continue/pull/9810 * Refactor imports. von @alex-mcgraw-askSage in https://github.com/continuedev/continue/pull/9828 * chore(deps): bump lodash von 4.17.21 auf 4.17.23 in /core von @dependabot

Codex CLIvrust-v0.96.0

0.96.0

4. Februar 2026

## New Features - Added `thread/compact` to the v2 app-server API as an async trigger RPC, so clients can start compaction immediately and track completion separately. (#10445) - Added websocket-side rate limit signaling via a new `codex.rate_limits` event, with websocket parity for ETag/reasoning metadata handling. (#10324) - Enabled `unified_exec` on all non-Windows platforms. (#10641) - Constrained requirement values now include source provenance, enabling source-aware config debugging in UI flows like `/debug-config`. (#10568) ## Bug Fixes - Fixed `Esc` handling in the TUI `request_user_input` overlay: when notes are open, `Esc` now exits notes mode instead of interrupting the session. (#10569) - Thread listing now queries the state DB first (including archived threads) and falls back to filesystem traversal only when needed, improving listing correctness and resilience. (#10544) - Fixed thread path lookup to require that the resolved file actually exists, preventing invalid thread-id resolutions. (#10618) - Dynamic tool injection now runs in a single transaction to avoid partial state updates. (#10614) - Refined `request_rule` guidance used in approval-policy prompting to correct rule behavior. (#10379, #10598) ## Documentation - Updated app-server docs for `thread/compact` to clarify its asynchronous behavior and thread-busy lifecycle. (#10445) - Updated TUI docs to match the mode-specific `Esc` behavior in `request_user_input`. (#10569) ## Chores - Migrated state DB helpers to a versioned SQLite filename scheme and cleaned up legacy state files during runtime initialization. (#10623) - Expanded runtime telemetry with websocket timing metrics and simplified internal metadata flow in core client plumbing. (#10577, #10589) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.95.0...rust-v0.96.0 - #10569 tui: make Esc clear request_user_input notes while notes are shown @charley-oai - #10577 feat: log webscocket timing into runtime metrics @apanasenko-oai - #10445 Add thread/compact v2 @aibrahim-oai - #10589 Move metadata calculation out of client @pakrym-oai - #10379 fix(core) updated request_rule guidance @dylan-hurd-oai - #10598 fix(core) Request Rule guidance tweak @dylan-hurd-oai - #10544 Prefer state DB thread listings before filesystem @jif-oai - #10614 fix: single transaction for dyn tools injection @jif-oai - #10568 Requirements: add source to constrained requirement values @gt-oai - #10611 chore: simplify user message detection @jif-oai - #10618 fix: make sure file exist in `find_thread_path_by_id_str_in_subdir` @jif-oai - #10619 nit: cleaning @jif-oai - #10324 Add a codex.rate_limits event for websockets @rasmusrygaard - #10623 Migrate state DB path helpers to versioned filename @jif-oai - #10638 Update tests to stop using sse_completed fixture @pakrym-oai - #10641 feat: land unified_exec @jif-oai

OpenCodev1.1.51

v1.1.51

4. Februar 2026

## Core - Revert change that caused headers to be double merged if provider was authenticated in multiple places - Document the built-in agents - Prevent double-prefixing of Bedrock cross-region inference models (@sergical) - Prioritize OPENCODE_CONFIG_DIR for AGENTS.md (@lgladysz) ## TUI - Restore direct OSC52 support ## Desktop - Tighten up session padding-top for mobile (@DNGriffin) **Thank you to 4 community contributors:** - @lgladysz: - fix: prioritize OPENCODE_CONFIG_DIR for AGENTS.md (#11536) - @DNGriffin: - fix(app): tighten up session padding-top for mobile (#11247) - @sergical: - fix: prevent double-prefixing of Bedrock cross-region inference models (#12056) - @schaoss: - docs: add agent-compatible paths to skills documentation (#12067)

Codex CLIvrust-v0.95.0

0.95.0

4. Februar 2026

## New Features - Added `codex app <path>` on macOS to launch Codex Desktop from the CLI, with automatic DMG download if it is missing. (#10418) - Added personal skill loading from `~/.agents/skills` (with `~/.codex/skills` compatibility), plus app-server APIs/events to list and download public remote skills. (#10437, #10448) - `/plan` now accepts inline prompt arguments and pasted images, and slash-command editing/highlighting in the TUI is more polished. (#10269) - Shell-related tools can now run in parallel, improving multi-command execution throughput. (#10505) - Shell executions now receive `CODEX_THREAD_ID`, so scripts and skills can detect the active thread/session. (#10096) - Added vendored Bubblewrap + FFI wiring in the Linux sandbox as groundwork for upcoming runtime integration. (#10413) ## Bug Fixes - Hardened Git command safety so destructive or write-capable invocations no longer bypass approval checks. (#10258) - Improved resume/thread browsing reliability by correctly showing saved thread names and fixing thread listing behavior. (#10340, #10383) - Fixed first-run trust-mode handling so sandbox mode is reported consistently, and made `$PWD/.agents` read-only like `$PWD/.codex`. (#10415, #10524) - Fixed `codex exec` hanging after interrupt in websocket/streaming flows; interrupted turns now shut down cleanly. (#10519) - Fixed review-mode approval event wiring so `requestApproval` IDs align with the corresponding command execution items. (#10416) - Improved 401 error diagnostics by including server message/body details plus `cf-ray` and `requestId`. (#10508) ## Documentation - Expanded TUI chat composer docs to cover slash-command arguments and attachment handling in plan/review flows. (#10269) - Refreshed issue templates and labeler prompts to better separate CLI/app bug reporting and feature requests. (#10411, #10453, #10548, #10552) ## Chores - Completed migration off the deprecated `mcp-types` crate to `rmcp`-based protocol types/adapters, then removed the legacy crate. (#10356, #10349, #10357) - Updated the `bytes` dependency for a security advisory and cleaned up resolved advisory configuration. (#10525) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.94.0...rust-v0.95.0 - #10340 Session picker shows thread_name if set @pap-openai - #10381 chore: collab experimental @jif-oai - #10231 feat: experimental flags @jif-oai - #10382 nit: shell snapshot retention to 3 days @jif-oai - #10383 fix: thread listing @jif-oai - #10386 fix: Rfc3339 casting @jif-oai - #10356 feat: add MCP protocol types and rmcp adapters @bolinfest - #10269 Nicer highlighting of slash commands, /plan accepts prompt args and pasted images @charley-oai - #10274 Add credits tooltip @pakrym-oai - #10394 chore: ignore synthetic messages @jif-oai - #10398 feat: drop sqlx logging @jif-oai - #10281 Select experimental features with space @pakrym-oai - #10402 feat: add `--experimental` to `generate-ts` @jif-oai - #10258 fix: unsafe auto-approval of git commands @viyatb-oai - #10411 Updated labeler workflow prompt to include "app" label @etraut-openai - #10399 emit a separate metric when the user cancels UAT during elevated setup @iceweasel-oai - #10377 chore(tui) /personalities tip @dylan-hurd-oai - #10252 [feat] persist thread_dynamic_tools in db @celia-oai - #10437 feat: Read personal skills from .agents/skills @gverma-openai - #10145 make codex better at git @pash-openai - #10418 Add `codex app` macOS launcher @aibrahim-oai - #10447 Fix plan implementation prompt reappearing after /agent thread switch @charley-oai - #10064 TUI: Render request_user_input results in history and simplify interrupt handling @charley-oai - #10349 feat: replace custom mcp-types crate with equivalents from rmcp @bolinfest - #10415 Fixed sandbox mode inconsistency if untrusted is selected @etraut-openai - #10452 Hide short worked-for label in final separator @aibrahim-oai - #10357 chore: remove deprecated mcp-types crate @bolinfest - #10454 app tool tip @aibrahim-oai - #10455 chore: add phase to message responseitem @sayan-oai - #10414 Require models refresh on cli version mismatch @aibrahim-oai - #10271 [Codex][CLI] Gate image inputs by model modalities @ccy-oai - #10374 Trim compaction input @pakrym-oai - #10453 Updated bug and feature templates @etraut-openai - #10465 Restore status after preamble @pakrym-oai - #10406 fix: clarify deprecation message for features.web_search @sayan-oai - #10474 Ignore remote_compact_trims_function_call_history_to_fit_context_window on windows @pakrym-oai - #10413 feat(linux-sandbox): vendor bubblewrap and wire it with FFI @viyatb-oai - #10142 feat(secrets): add codex-secrets crate @viyatb-oai - #10157 chore: nuke chat/completions API @jif-oai - #10498 feat: drop wire_api from clients @jif-oai - #10501 feat: clean codex-api part 1 @jif-oai - #10508 Add more detail to 401 error @gt-oai - #10521 Avoid redundant transactional check before inserting dynamic tools @jif-oai - #10525 chore: update bytes crate in response to security advisory @bolinfest - #10408 fix WebSearchAction type clash between v1 and v2 @sayan-oai - #10404 Cleanup collaboration mode variants @charley-oai - #10505 Enable parallel shell tools @jif-oai - #10532 feat: `find_thread_path_by_id_str_in_subdir` from DB @jif-oai - #10524 fix: make $PWD/.agents read-only like $PWD/.codex @bolinfest - #10096 Inject CODEX_THREAD_ID into the terminal environment @maxj-oai - #10536 Revert "Load untrusted rules" @viyatb-oai - #10412 fix(app-server): fix TS annotations for optional fields on requests @owenlin0 - #10416 fix(app-server): fix approval events in review mode @owenlin0 - #10545 Improve Default mode prompt (less confusion with Plan mode) @charley-oai - #10289 [apps] Gateway MCP should be blocking. @mzeng-openai - #10189 implement per-workspace capability SIDs for workspace specific ACLs @iceweasel-oai - #10548 Updated bug templates and added a new one for app @etraut-openai - #10531 [codex] Default values from requirements if unset @gt-oai - #10552 Fixed icon for CLI bug template @etraut-openai - #10039 chore(arg0): advisory-lock janitor for codex tmp paths @viyatb-oai - #10448 feat: add APIs to list and download public remote skills @xl-openai - #10519 Handle exec shutdown on Interrupt (fixes immortal `codex exec` with websockets) @rasmusrygaard - #10556 Feat: add upgrade to app server modelList @shijie-oai - #10461 feat(tui): pace catch-up stream chunking with hysteresis @joshka-oai - #10367 chore: add `codex debug app-server` tooling @celia-oai

OpenCodev1.1.50

v1.1.50

4. Februar 2026

## Core - Prevent memory leaks from AbortController closures (@MaxLeiter) - Revert addition of Trinity model system prompt support - Add Trinity model system prompt support (@mariamjabara) - Add shell.env hook for manipulating environment in tools and shell (@tylergannon) - Use official ai-gateway-provider package for Cloudflare AI Gateway (@elithrar) - Allow theme colors in agent customization (@IdrisGit) - Add support for reading skills from .agents/skills directories - Provider headers from config not applied to fetch requests (@cloudyan) - Ensure MCP tools are sanitized - Add .slnx to C#/F# LSP root detection (@workedbeforepush) - Improve skills system with better prompting, fix permission requests after skill invocation, and ensure agents can locate scripts and resources - Exclude k2p5 from reasoning variants (@neavo) - Handle nested array items for Gemini schema validation (@mugnimaestra) - Plugins are always reinstalled (@neriousy) - Strip properties and required fields from non-object types in Gemini schema (@ChickenBreast-ky) - Make CLI run command non-interactive ## TUI - Add --thinking flag to show reasoning blocks in run command - Always fall back to native clipboard after OSC52 (@MartinWie) ## Desktop - Faster end-to-end tests (@neriousy) - Update command palette placeholder text - Model selector truncating too soon - Improve spacing in application UI - Allow empty prompt with review comments in desktop (@dbpolito) - Fixed terminal serialization bug in app - Don't force mount tooltips in the app - Restore previously opened session tabs on app restart (@ProdigyRahul) - Edit project dialog icon now shows on hover (@ProdigyRahul) - Move session search to command palette - Fix custom providers overflow in app (@DNGriffin) **Thank you to 18 community contributors:** - @ChickenBreast-ky: - fix(provider): strip properties/required from non-object types in Gemini schema (#11888) - @neriousy: - fix(core): plugins are always reinstalled (#9675) - refactor(e2e): faster tests (#12021) - @DNGriffin: - fix(app): custom providers overflow (#11252) - @ProdigyRahul: - fix(app): edit project dialog icon on hover (#11921) - fix(app): session tabs to open the previous opened (#11914) - @luiz290788: - docs: add --mdns-domain flag documentation (#11933) - @dbpolito: - feat(desktop): Allow empty prompt with review comments (#11953) - @mugnimaestra: - fix: handle nested array items for Gemini schema validation (#11952) - @neavo: - fix: exclude k2p5 from reasoning variants (#11918) - @workedbeforepush: - feat: Add .slnx to C#/F# LSP root detection (#11928) - @cloudyan: - fix: provider headers from config not applied to fetch requests (#11788) - @MartinWie: - fix: always fall back to native clipboard after OSC52 (#11994) - @IdrisGit: - docs: add missing environmental flags to the list (#11146) - feat(tui): allow theme colors in agent customization (#11444) - @elithrar: - fix(opencode): use official ai-gateway-provider package for Cloudflare AI Gateway (#12014) - @tylergannon: - feat(plugin): add shell.env hook for manipulating environment in tools and shell (#12012) - @tlinhart: - docs: fix logging example for plugin (#11989) - @Evren-os: - docs: fix grammar and formatting in README (#11985) - @mariamjabara: - feat: add Trinity model system prompt support (#12025) - @MaxLeiter: - fix: prevent memory leaks from AbortController closures (#12024)

OpenClawv2026.2.2

openclaw 2026.2.2

4. Februar 2026

### Changes - Feishu: add Feishu/Lark plugin support + docs. (#7313) Thanks @jiulingyun (openclaw-cn). - Web UI: add Agents dashboard for managing agent files, tools, skills, models, channels, and cron jobs. - Memory: implement the opt-in QMD backend for workspace memory. (#3160) Thanks @vignesh07. - Security: add healthcheck skill and bootstrap audit guidance. (#7641) Thanks @Takhoffman. - Config: allow setting a default subagent thinking level via `agents.defaults.subagents.thinking` (and per-agent `agents.list[].subagents.thinking`). (#7372) Thanks @tyler6204. - Docs: zh-CN translations seed + polish, pipeline guidance, nav/landing updates, and typo fixes. (#8202, #6995, #6619, #7242, #7303, #7415) Thanks @AaronWander, @taiyi747, @Explorer1092, @rendaoyuan, @joshp123, @lailoo. ### Fixes - Security: require operator.approvals for gateway /approve commands. (#1) Thanks @mitsuhiko, @yueyueL. - Security: Matrix allowlists now require full MXIDs; ambiguous name resolution no longer grants access. Thanks @MegaManSec. - Security: enforce access-group gating for Slack slash commands when channel type lookup fails. - Security: require validated shared-secret auth before skipping device identity on gateway connect. - Security: guard skill installer downloads with SSRF checks (block private/localhost URLs). - Security: harden Windows exec allowlist; block cmd.exe bypass via single &. Thanks @simecek. - fix(voice-call): harden inbound allowlist; reject anonymous callers; require Telnyx publicKey for allowlist; token-gate Twilio media streams; cap webhook body size (thanks @simecek) - Media understanding: apply SSRF guardrails to provider fetches; allow private baseUrl overrides explicitly. - fix(webchat): respect user scroll position during streaming and refresh (#7226) (thanks @marcomarandiz) - Telegram: recover from grammY long-poll timed out errors. (#7466) Thanks @macmimi23. - Agents: repair malformed tool calls and session transcripts. (#7473) Thanks @justinhuangcode. - fix(agents): validate AbortSignal instances before calling AbortSignal.any() (#7277) (thanks @Elarwei001) - Media understanding: skip binary media from file text extraction. (#7475) Thanks @AlexZhangji. - Onboarding: keep TUI flow exclusive (skip completion prompt + background Web UI seed); completion prompt now handled by install/update. - TUI: block onboarding output while TUI is active and restore terminal state on exit. - CLI/Zsh completion: cache scripts in state dir and escape option descriptions to avoid invalid option errors. - fix(ui): resolve Control UI asset path correctly. - fix(ui): refresh agent files after external edits. - Docs: finish renaming the QMD memory docs to reference the OpenClaw state dir. - Tests: stub SSRF DNS pinning in web auto-reply + Gemini video coverage. (#6619) Thanks @joshp123.

Claude Codev2.1.31

v2.1.31

4. Februar 2026

## What's changed - Added session resume hint on exit, showing how to continue your conversation later - Added support for full-width (zenkaku) space input from Japanese IME in checkbox selection - Fixed PDF too large errors permanently locking up sessions, requiring users to start a new conversation - Fixed bash commands incorrectly reporting failure with "Read-only file system" errors when sandbox mode was enabled - Fixed a crash that made sessions unusable after entering plan mode when project config in `~/.claude.json` was missing default fields - Fixed `temperatureOverride` being silently ignored in the streaming API path, causing all streaming requests to use the default temperature (1) regardless of the configured override - Fixed LSP shutdown/exit compatibility with strict language servers that reject null params - Improved system prompts to more clearly guide the model toward using dedicated tools (Read, Edit, Glob, Grep) instead of bash equivalents (`cat`, `sed`, `grep`, `find`), reducing unnecessary bash command usage - Improved PDF and request size error messages to show actual limits (100 pages, 20MB) - Reduced layout jitter in the terminal when the spinner appears and disappears during streaming - Removed misleading Anthropic API pricing from model selector for third-party provider (Bedrock, Vertex, Foundry) users

Gemini CLIv0.27.0

Release v0.27.0

4. Februar 2026

## Was sich geändert hat * Entferne fireAgent- und beforeAgent-Hooks von @ishaanxgupta in https://github.com/google-gemini/gemini-cli/pull/16919 * Entfernt ungenutzte modelHooks und toolHooks von @ved015 in https://github.com/google-gemini/gemini-cli/pull/17115 * feat(cli): ANSI-Escape-Sequenzen in nicht-interaktiver Ausgabe bereinigen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17172 * Aktualisiere Attempt-Text zu Retry, wenn der Versuch gezeigt wird, dass … von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17178 * chore(skills): aktualisiere pr-creator Skill-Workflow von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17180 * feat(cli): implementiere ereignisgesteuerten Scheduler für Tool-Ausführung von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17078 * chore(release): Versions-Hebe auf 0.27.0-nightly.20260121.97aac696f von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/17181 * Entferne andere Rewind-Referenz in der Dokumentation von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/17149 * feat(skills): füge Code-Reviewer-Skill hinzu von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/17187 * feat(plan): Erweitere `Shift+Tab`-Modus-Zyklus um `Plan`-Modus von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/17177 * feat(plan): refaktorisiere TestRig und Eval-Hilfsprogramm zur Unterstützung konfigurierbarer Freigabemodi von @jerop in https://github.com/google-gemini/gemini-cli/pull/17171 * feat(workflows): Unterstützung rekursiver Workstream-Beschriftung und neue IDs von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/17207 * Führe Eval-Tests für alle Modelle aus. von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/17123 * fix(github): verbessere Label-Workstream-Rollup-Effizienz mit GraphQL von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/17217 * Docs: Aktualisiere Changelogs für Releases v0.25.0 und v0.26.0-preview.0 von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/17215 * Migriere beforeTool- und afterTool-Hooks zu hookSystem von @ved015 in https://github.com/google-gemini/gemini-cli/pull/17204 * fix(github): verbessere Label-Workstream-Rollup-Effizienz und behebe Bugs von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/17219 * feat(cli): verbessere Fähigkeiten-Aktivierung/Deaktivierungstext von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/17192 * fix(admin): Sicherstellen, dass CLI-Befehle im Nicht-Interaktiv-Modus laufen von @skeshive in https://github.com/google-gemini/gemini-cli/pull/17218 * feat(core): Unterstützung dynamischer Variablenersetzung im Systemprompt-Override von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/17042 * fix(core,cli): ermögliche rekursiven Verzeichniszugriff durch @galz10 in https://github.com/google-gemini/gemini-cli/pull/17094 * Docs: Kennzeichnung experimenteller Features von @jkcinouye in https://github.com/google-gemini/gemini-cli/pull/16760 * Unterstütze Command/Ctrl/Alt Backspace korrekt von @scidomino in https://github.com/google-gemini/gemini-cli/pull/17175 * feat(plan): füge Anweisungen zum Freigabemodus in Systemprompt hinzu von @jerop in https://github.com/google-gemini/gemini-cli/pull/17151 * feat(core): standardisiere Enable/Disable von LLMCorrection standardmäßig von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/17223 * Remove unused slug from sidebar von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/17229 * leite stdin beim Beenden ab von @scidomino in https://github.com/google-gemini/gemini-cli/pull/17241 * refactor(cli): trenne UI von Live-Tool-Ausführung über ToolActionsContext von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17183 * fix(core): aktualisiere Token-Anzahl und Telemetrie bei /chat-Resume-History-Laden von @psinha40898 in https://github.com/google-gemini/gemini-cli/pull/16279 * fix: /policy zur Anzeige von Richtlinien gemäß Modus von @ishaanxgupta in https://github.com/google-gemini/gemini-cli/pull/16772 * fix(core): vereinfache Fehlermeldung bei Tool-Ersetzung von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/17246 * feat(cli): konsolidiere Inaktivität der Shell und Monitoring der Umleitung durch @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/17086 * fix(scheduler): Verhindere veraltete Tool-Neuveröffentlichung und behebe eingefrorenen UI-Zustand von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17227 * feat(config): Standardmäßig EnableEventDrivenScheduler auf true setzen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17211 * feat(hooks): standardmäßig Hook-System aktivieren von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/17247 * feat(core): AgentRegistry, um alle entdeckten Subagents nachverfolgen zu können, von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/17253 * feat(core): Subagents verwenden JSON-Schema-Typ für Input. von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/17152 * feat: ersetze großen Text-Einfügen mit Platzhalter [Pasted Text: X lines] von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/16422 * security(hooks): Kontext, der Hook-injektierte Kontext in eindeutige XML-Tags wickelt von @yunaseoul in https://github.com/google-gemini/gemini-cli/pull/17237 * Ermögliche das Queueing bestimmter nächtlicher Eval-Tests von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/17262 * docs(hooks): Umfassendes Update der Hook-Dokumentation und Specs von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/16816 * Refactor: verbessere Placeholder für Large-Text-Paste von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/17269 * feat: implementiere /rewind-Befehl von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/15720 * Feature/jetbrains IDE-Erkennung von @SoLoHiC in https://github.com/google-gemini/gemini-cli/pull/16243 * docs: Tippfehler in mcp-server.md-Datei von @schifferl in https://github.com/google-gemini/gemini-cli/pull/17099 * Sanitize Befehlsnamen und -beschreibungen von @ehedlund in https://github.com/google-gemini/gemini-cli/pull/17228 * fix(auth): verhindert Absturz, wenn anfängliche Auth fehlschlägt von @skeshive in https://github.com/google-gemini/gemini-cli/pull/17308 * Neue Bild-Paste-Fähigkeiten für Wayland und X11 auf Linux von @devr0306 in https://github.com/google-gemini/gemini-cli/pull/17144 * feat: füge AskUser

Claude Codev2.1.30

v2.1.30

3. Februar 2026

## What's changed - Added `pages` parameter to the Read tool for PDFs, allowing specific page ranges to be read (e.g., `pages: "1-5"`). Large PDFs (>10 pages) now return a lightweight reference when `@` mentioned instead of being inlined into context. - Added pre-configured OAuth client credentials for MCP servers that don't support Dynamic Client Registration (e.g., Slack). Use `--client-id` and `--client-secret` with `claude mcp add`. - Added `/debug` for Claude to help troubleshoot the current session - Added support for additional `git log` and `git show` flags in read-only mode (e.g., `--topo-order`, `--cherry-pick`, `--format`, `--raw`) - Added token count, tool uses, and duration metrics to Task tool results - Added reduced motion mode to the config - Fixed phantom "(no content)" text blocks appearing in API conversation history, reducing token waste and potential model confusion - Fixed prompt cache not correctly invalidating when tool descriptions or input schemas changed, only when tool names changed - Fixed 400 errors that could occur after running `/login` when the conversation contained thinking blocks - Fixed a hang when resuming sessions with corrupted transcript files containing `parentUuid` cycles - Fixed rate limit message showing incorrect "/upgrade" suggestion for Max 20x users when extra-usage is unavailable - Fixed permission dialogs stealing focus while actively typing - Fixed subagents not being able to access SDK-provided MCP tools because they were not synced to the shared application state - Fixed a regression where Windows users with a `.bashrc` file could not run bash commands - Improved memory usage for `--resume` (68% reduction for users with many sessions) by replacing the session index with lightweight stat-based loading and progressive enrichment - Improved `TaskStop` tool to display the stopped command/task description in the result line instead of a generic "Task stopped" message - Changed `/model` to execute immediately instead of being queued - [VSCode] Added multiline input support to the "Other" text input in question dialogs (use Shift+Enter for new lines) - [VSCode] Fixed duplicate sessions appearing in the session list when starting a new conversation

Continue.devv@continuedev/config-yaml@1.42.0

@continuedev/config-yaml@1.42.0

3. Februar 2026

# [1.42.0](https://github.com/continuedev/continue/compare/@continuedev/config-yaml@1.41.0...@continuedev/config-yaml@1.42.0) (2026-02-03) ### Bug Fixes * support für reasoning content field im Chat-Body hinzufügen ([ac93dd9](https://github.com/continuedev/continue/commit/ac93dd9381021989507b1c724205780074898786)) * support für reasoning content field im Chat-Body hinzufügen ([#10027](https://github.com/continuedev/continue/issues/10027)) ([6c56210](https://github.com/continuedev/continue/commit/6c56210f1c4cf52a15bf95325738d2dfb8637b51)) * **cli:** SSL-Verifizierungsoption für Client-Transports hinzufügen ([#9551](https://github.com/continuedev/continue/issues/9551)) ([0b6937a](https://github.com/continuedev/continue/commit/0b6937a3b8b0a6c56bbbe3b6ecaceb9869123508)) * Reasoning-Toggle aus dem lokalen Speicher abrufen ([#9688](https://github.com/continuedev/continue/issues/9688)) ([2c1a856](https://github.com/continuedev/continue/commit/2c1a85683c515b5a19c7d8a84ac1af8357d7d6dc)) * Fehler anzeigen bei leeren Eingaben bei der Verwendung von Hub-Blöcken in der lokalen Config ([20a069e](https://github.com/continuedev/continue/commit/20a069e87e7b62f94150240071193676d423f6ca)) ### Features * Nous Research Hermes-Modelle als Provider hinzufügen ([1ba9a66](https://github.com/continuedev/continue/commit/1ba9a669d6f9c432a3f17e2f6aa045442458df90)) * OpenRouter Provider-Unterstützung mit dynamischer Modell-Ladung hinzufügen ([#8738](https://github.com/continuedev/continue/issues/8738)) ([9ac154d](https://github.com/continuedev/continue/commit/9ac154d2a0690f7541b5b7727a472508e39a03f2))

Continue.devv@continuedev/config-yaml@1.41.0

@continuedev/config-yaml@1.41.0

3. Februar 2026

# [1.41.0](https://github.com/continuedev/continue/compare/@continuedev/config-yaml@1.40.0...@continuedev/config-yaml@1.41.0) (2026-02-03) ### Bug Fixes * detect WSL remote for shell PATH resolution ([939b44b](https://github.com/continuedev/continue/commit/939b44b39fbe1eeea0d4c501e820f23ea5079c87)), closes [#9737](https://github.com/continuedev/continue/issues/9737) * detect WSL remote for shell PATH resolution ([#9934](https://github.com/continuedev/continue/issues/9934)) ([895632c](https://github.com/continuedev/continue/commit/895632c0a5cea5a032b31df7e5fcc5a60e3136e4)), closes [#9737](https://github.com/continuedev/continue/issues/9737) * OpenAI Responses API parallel tool calls losing call_ids ([#9659](https://github.com/continuedev/continue/issues/9659)) ([dcac626](https://github.com/continuedev/continue/commit/dcac6260a7c33dc09b740507f3a93424b8afa047)) * **resource-monitoring:** prevent file descriptor count leak by adding caching ([a99288f](https://github.com/continuedev/continue/commit/a99288f25a1ff0939fa7b6581d8d22a9fa75e4fd)) * **ResourceMonitoringService:** initialize fd check time and count on start ([df83e66](https://github.com/continuedev/continue/commit/df83e66d5ce97cfcf28c1a815d567090cceda622)) * **ResourceMonitoringService:** prevent negative file descriptor counts and remove test cases ([676e31d](https://github.com/continuedev/continue/commit/676e31d28edd634307e289ee103ec6f42a40f416)) * simplify file descriptor count calculation ([056622f](https://github.com/continuedev/continue/commit/056622faa01779b2dd0c99fe5ea758951454822a)) * upgrade @openapitools/openapi-generator-cli from 2.25.1 to 2.27.0 ([1420644](https://github.com/continuedev/continue/commit/142064476870ed6b462dcbb8c651db7991942aec)) ### Features * add tool prompt override support in .continuerc.json ([#9314](https://github.com/continuedev/continue/issues/9314)) ([d80845c](https://github.com/continuedev/continue/commit/d80845c403db83686d140c72c04616348731fdbf))

OpenCodev1.1.49

v1.1.49

3. Februar 2026

## Core - Revert pull request that was mistakenly merged - Add --mdns-domain flag to customize mDNS hostname (@luiz290788) - Added and deleted file status now correctly calculated in app - Add User-Agent header for GitLab AI Gateway requests (@vglafirov) - Add Ormolu code formatter for Haskell (@mimi1vx) - Use OpenTUI OSC52 clipboard implementation - Convert system message content to string for Copilot provider (@jogi47) - Give OPENCODE_CONFIG_CONTENT proper priority for setting config based on docs (@OpeOginni) - Fixed session title generation with OpenAI models (@oomathias) - Simplify directory tree output for prompts - Fix task status to show current tool state from message store - Allow starting new sessions after errors by fixing stuck session status - Adjust resolve parts to properly order tool calls when messages contain multiple @ references - Hide badge for builtin slash commands - Add workspace tests to the app (@neriousy) - Send custom agent prompts as developer messages instead of user messages when using Codex subscriptions - Fixed variant logic for Anthropic models through OpenAI compatibility endpoint - Add prompt caching support for Claude Opus on AWS Bedrock (@rgodha24) - Scope agent variant to model (@neavo) - Prevent duplicate AGENTS.md injection when reading instruction files (@code-yeongyu) - OpenCode no longer hangs when using client.app.log() during initialization (@desmondsow) - Remove outer backtick wrapper in session transcript tool formatting (@zerone0x) - Allow user plugins to override built-in auth plugins (@JosXa) - Binary file handling in file view (@alexyaroshuk) - Ensure switching Anthropic models mid-conversation works without errors and fix reasoning opaque not being picked up for Gemini models - Fix issue where folders and files starting with "." could not be mentioned with @ - Show actual retry error message instead of generic error message - Use process.env directly for runtime environment mutations in provider (@jerome-benoit) - Add reasoning variants support for SAP AI Core (@jerome-benoit) - Add UTF-8 encoding defaults for Windows PTY (@01luyicheng) ## TUI - Respect terminal transparency in system theme (@pablopunk) - Truncate session title in exit banner to prevent display overflow - Show exit message banner in TUI - Add spinner animation for Task tool - Correct pluralization of match count in grep and glob tools (@adamjhf) - Remove extra padding between search and results in dialog-select - Add UI for skill tool in session view - Conditionally render bash tool output in TUI - Add skill dialog for selecting and inserting skills - Enable password authentication for remote session attachment - Fix documentation issues (@lailoo) - Revert skill slash commands feature - Add skill slash commands to the app ## Desktop - Fix prompt input overflow issue in app (@neriousy) - Sidebar losing projects on collapse has been fixed - Add workspace toggle command to command palette and prompt input (@alexyaroshuk) - Search through sessions - Standardize icon sizes in the application - Navigate to last project on open - User messages not rendering consistently - Add project context menu on right-click - Open project search in app (@neriousy) - Add tab close keybind to app (@ProdigyRahul) - Enhance responsive design with additional breakpoints for larger screen layout adjustments (@OpeOginni) - Add keyboard shortcuts for navigating between unread sessions (@Brendonovich) - Fix Rust build and bindings formatting in desktop application (@Brendonovich) - Remove unnecessary setTimeout in desktop app (@Brendonovich) - Throttle window state persistence in desktop application (@Brendonovich) - Integrate tauri-specta for desktop application (@Brendonovich) - Keep macOS titlebar stable under zoom (@Brendonovich) - Kill zombie server process on startup timeout (@heyitsmohdd) - Show skill and MCP badges for slash commands - Use static language names in Thai localization (@alexyaroshuk) - Add tests for general settings, shortcuts, providers and status popover (@neriousy) - Fixed session header 'share' button to hug content (@alexyaroshuk) **Thank you to 29 community contributors:** - @alexyaroshuk: - fix(app): session header 'share' button to hug content (#11371) - fix(app): use static language names in Thai localization (#11496) - fix(app): binary file handling in file view (#11312) - feat(app): add workspace toggle command to command palette and prompt input (#11810) - @01luyicheng: - fix(pty): Add UTF-8 encoding defaults for Windows PTY (#11459) - docs: improve zh-TW punctuation to match Taiwan usage (#11574) (#11589) - @jerome-benoit: - feat(opencode): add reasoning variants support for SAP AI Core (#8753) - fix(nix): restore native runners for darwin hash computation (#11495) - fix(provider): use process.env directly for runtime env mutations (#11482) - fix(ci): portable hash parsing in nix-hashes workflow (#11533) - @lailoo: - docs: fix documentation issues (#11435) - @neriousy: - test(app): general settings, shortcuts, providers and status popover (#11517) - fix(app): rendering question tool when the step are collapsed (#11539) - fix(app): show retry status only on active turn (#11543) - docs: prefer wsl over native windows stuff (#11637) - test(app): workspace tests (#11659) - fix(app): open project search (#11783) - fix(app): prompt input overflow issue (#11840) - @adamjhf: - fix: correct pluralization of match count in grep and glob tools (#11565) - @AlperKartkaya: - docs: add Turkish README translation (#11524) - @JosXa: - fix: allow user plugins to override built-in auth plugins (#11058) - @AxelMrak: - fix(ecosystem): fix link Daytona (#11621) - @zerone0x: - fix(tui): remove outer backtick wrapper in session transcript tool formatting (#11566) - @desmondsow: - fix: opencode hanging when using client.app.log() during initialization (#11642) - @code-yeongyu: - fix: prevent duplicate AGENTS.md injection when reading instruction files (#11581) - @neavo: - fix(opencode): scope agent variant to model (#11410) - @gigamonster256: - fix(desktop): nix - add missing dep (#11656) - @rgodha24: - fix: prompt caching for opus on bedrock (#11664) - @sum2it: - docs (web): Update incorrect Kimi model name in zen.mdx (#11677) - @aaroniker: - feat(ui): Select, dropdown, popover styles & transitions (#11675) - feat(ui): Smooth fading out on scroll, style fixes (#11683) - @oomathias: - fix: session title generation with OpenAI models. (#11678) - @OpeOginni: - fix(opencode): give OPENCODE_CONFIG_CONTENT proper priority for setting config based on docs (#11670) - fix(desktop): added inverted svg for steps expanded for nice UX (#10462) - feat(app): enhance responsive design with additional breakpoints for larger screen layout adjustments (#10459) - @jogi47: - fix: convert system message content to string for Copilot provider (#11600) - @heyitsmohdd: - fix(desktop): kill zombie server process on startup timeout (#11602) - @Brendonovich: - fix(desktop): keep mac titlebar stable under zoom (#11747) - chore(desktop): integrate tauri-specta (#11740) - fix(desktop): throttle window state persistence (#11746) - fix(desktop): remove unnecessary setTimeout - desktop: fix rust build + bindings formatting - feat(app): unread session navigation keybinds (#11750) - @ldelelis: - fix(ui): adjusts alignment of elements to prevent incomplete scroll (#11649) - @sam-huckaby: - Fix(app): the Vesper theme's light mode (#9892) - @mimi1vx: - feat(opencode): ormolu code formatter for haskell (#10274) - @ProdigyRahul: - feat(app): add tab close keybind (#11780) - @vglafirov: - feat(provider): add User-Agent header for GitLab AI Gateway requests (#11818) - @pablopunk: - fix(tui): respect terminal transparency in system theme (#8467) - @luiz290788: - feat(server): add --mdns-domain flag to customize mDNS hostname (#11796)

Continue.devv1.5.43

CLI Stable Release v1.5.43

2. Februar 2026

Direktes stabiles Release aus dem main-Branch. Version 1.5.43 wurde direkt aus dem neuesten main-Branch gebaut und veröffentlicht.

Codex CLIvrust-v0.94.0

0.94.0

2. Februar 2026

## New Features - Plan mode is now enabled by default with updated interaction guidance in the plan prompt. (#10313, #10308, #10329) - Personality configuration is now stable: default is friendly, the config key is `personality`, and existing settings migrate forward. (#10305, #10314, #10310, #10307) - Skills can be loaded from `.agents/skills`, with clearer relative-path instructions and nested-folder markers supported. (#10317, #10282, #10350) - Console output now includes runtime metrics for easier diagnostics. (#10278) ## Bug Fixes - Unarchiving a thread updates its timestamp so sidebar ordering refreshes. (#10280) - Conversation rules output is capped and prefix rules are deduped to avoid repeated rules. (#10351, #10309) - Override turn context no longer appends extra items. (#10354) ## Documentation - Fixed a broken image link in the npm README. (#10303) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.93.0...rust-v0.94.0 - #10278 feat: show runtime metrics in console @apanasenko-oai - #10285 display promo message in usage error @willwang-openai - #10302 fix(nix): update flake for newer Rust toolchain requirements @douglaz - #10296 chore(features) remove Experimental tag from UTF8 @dylan-hurd-oai - #10303 Fix npm README image link @fouad-openai - #10306 chore(app-server) add personality update test @dylan-hurd-oai - #10308 plan mode prompt @aibrahim-oai - #10305 chore(core) Default to friendly personality @dylan-hurd-oai - #10307 feat(core,tui,app-server) personality migration @dylan-hurd-oai - #10313 enable plan mode @aibrahim-oai - #10120 feat: fire tracking events for skill invocation @alexsong-oai - #10317 feat: Support loading skills from .agents/skills @gverma-openai - #10282 Make skills prompt explicit about relative-path lookup @xl-openai - #10316 Add websocket telemetry metrics and labels @apanasenko-oai - #10314 chore(config) Rename config setting to personality @dylan-hurd-oai - #10310 chore(features) Personality => Stable @dylan-hurd-oai - #10320 Sync system skills from public repo @gverma-openai - #10322 Sync system skills from public repo for openai yaml changes @gverma-openai - #10323 fix(config) config schema newline @dylan-hurd-oai - #10329 Improve plan mode interaction rules @charley-oai - #10280 Bump thread updated_at on unarchive to refresh sidebar ordering @charley-oai - #10350 fix: System skills marker includes nested folders recursively @gverma-openai - #10351 fix(rules) Limit rules listed in conversation @dylan-hurd-oai - #10354 Do not append items on override turn context @pakrym-oai - #10309 fix(core) Deduplicate prefix_rules before appending @dylan-hurd-oai - #10373 chore(core) gpt-5.2-codex personality template @dylan-hurd-oai - #10375 fix(personality) prompt patch @dylan-hurd-oai - #10371 feat: vendor app-server protocol schema fixtures @bolinfest

OpenClawv2026.2.1

openclaw 2026.2.1

2. Februar 2026

### Changes - Docs: onboarding/install/i18n/exec-approvals/Control UI/exe.dev/cacheRetention updates + misc nav/typos. (#3050, #3461, #4064, #4675, #4729, #4763, #5003, #5402, #5446, #5474, #5663, #5689, #5694, #5967, #6270, #6300, #6311, #6416, #6487, #6550, #6789) - Telegram: use shared pairing store. (#6127) Danke @obviyus. - Agents: add OpenRouter app attribution headers. Danke @alexanderatallah. - Agents: add system prompt safety guardrails. (#5445) Danke @joshp123. - Agents: update pi-ai to 0.50.9 and rename cacheControlTtl -> cacheRetention (mit Rückkompat-Mapping). - Agents: extend CreateAgentSessionOptions with systemPrompt/skills/contextFiles. - Agents: add tool policy conformance snapshot (no runtime behavior change). (#6011) - Auth: update MiniMax OAuth hint + portal auth note copy. - Discord: inherit thread parent bindings for routing. (#3892) Danke @aerolalit. - Gateway: timestamps in agent- und chat.send messages injizieren. (#3705) Danke @conroywhitney, @CashWilliams. - Gateway: TLS 1.3 minimum for TLS listeners erzwingen. (#5970) Danke @loganaden. - Web UI: Chat-Layout verfeinern + Sitzungs aktive Dauer verlängern. - CI: formale Conformance + Alias-Consistency-Checks hinzufügen. (#5723, #5807) ### Fixes - Plugins: validiere Plugin/Hook-Installationspfade und lehne Traversal-ähnliche Namen ab. - Telegram: Download-Timouts für File-Fetches hinzufügen. (#6914) Danke @hclsys. - Telegram: Thread-Specs für DM vs Forum-Sends erzwingen. (#6833) Danke @obviyus. - Streaming: Block-Streaming bei Absatzgrenzen flushen (Newline-Chunking). (#7014) - Streaming: partielle Streaming-Filter stabilisieren. - Auto-reply: Verweise auf Workspace-Dateien im /new Greeting Prompt vermeiden. (#5706) Danke @bravostation. - Tools: Align Tool Execute Adapters/Signatures (Legacy + Parameter-Reihenfolge + Arg-Normalisierung). - Tools: treat `"*"` Tool-Allowlist-Einträge als gültig, um spurious unknown-entry-Warnungen zu vermeiden. - Skills: Sitzung-Logs-Pfade von .clawdbot zu .openclaw aktualisieren. (#4502) - Slack: Medien-Abruffolgen härten und Slack-Datei-URL-Validierung verbessern. (#6639) Danke @davidiach. - Lint: Curly-Regel nach Import-Sortierung erfüllen. (#6310) - Process: Windows `spawn()`-Fehler bei npm-familien CLIs durch Anhängen von `.cmd` beheben. (#5815) Danke @thejhinvirtuoso. - Discord: PluralKit proxied Sender für Allowlists und Labels auflösen. (#5838) Danke @thewilloftheshadow. - Tlon: Timeout zu SSE-Client-Fetch-Aufrufen hinzufügen (CWE-400). (#5926) - Memory search: L2-Normalisierung lokaler Embedding-Vektoren für semantische Suche. (#5332) - Agents: Embedded Runner + Typings an pi-coding-agent API-Updates anpassen (pi 0.51.0). - Agents: sicherstellen, dass OpenRouter Attribution Headers im Embedded Runner angewendet werden. - Agents: Kontextfenster-Auflösung für Kompaktion begrenzen. (#6187) Danke @iamEvanYT. - System prompt: Overrides lösen und Hint mit session_status für aktuelles Datum/Zeit verwenden. (#1897, #1928, #2108, #3677) - Agents: Pi-Prompt-Template-Argument-Syntax beheben. (#6543) - Subagents: Announce-Failover-Race beheben (immer Lifecycle-End auslösen; timeout=0 bedeutet kein Timeout). (#6621) - Teams: Media-Auth-Retries begrenzen. - Telegram: Draft-Streaming-Partials wiederherstellen. (#5543) Danke @obviyus. - Onboarding: Freundlicher Windows-Onboarding-Message. (#6242) Danke @shanselman. - TUI: Absturz verhindern bei Suche mit Ziffern im Model-Selector. - Agents: vor_tool_call Plugin-Hook in Tool-Execution einbinden. (#6570, #6660) Danke @ryancnelson. - Browser: Chrome Extension Relay CDP-Sessions sichern. - Docker: Container-Port für Gateway-Befehl statt Host-Port verwenden. (#5110) Danke @mise42. - fix(lobster): blockiere beliebige exec via lobsterPath/cwd-Injektion (GHSA-4mhr-g7xj-cg8j). (#5335) Danke @vignesh07. - Security: WhatsApp accountId bereinigen, um Pfad-Traversal zu verhindern. (#4610) - Security: MEDIA Pfad-Extraktion einschränken, um LFI zu verhindern. (#4930) - Security: validate message-tool filePath/path gegen Sandbox-Root. (#6398) - Security: LD*/DYLD*-Umgebungs-Overrides für Host-Exec blockieren. (#4896) Danke @HassanFleyah. - Security: Web-Tool-Content-Wrapping + Dateiprüf-Schutzmaßnahmen härten. (#4058) Danke @VACInc. - Security: Twitch `allowFrom` Allowlist-Gating erzwingen (verweigere nicht-allowlisted Sender). Danke @MegaManSec.

Continue.devv1.5.42

CLI Stable Release v1.5.42

2. Februar 2026

Direktes stabiles Release aus dem Hauptzweig. Version 1.5.42 wurde direkt aus dem neuesten Hauptzweig gebaut und veröffentlicht.

Continue.devv1.5.41

CLI Stable Release v1.5.41

2. Februar 2026

Direktes stabiler Release aus dem Hauptzweig. Version 1.5.41 wurde direkt aus dem neuesten Hauptzweig gebaut und veröffentlicht.

Continue.devv1.5.40

CLI Stable Release v1.5.40

2. Februar 2026

Direktes stabiles Release vom Hauptbranch. Version 1.5.40 gebaut und direkt aus dem neuesten main branch veröffentlicht.

Claude Codev2.1.29

v2.1.29

31. Januar 2026

## What's changed - Fixed startup performance issues when resuming sessions that have `saved_hook_context`

Continue.devv@continuedev/config-yaml@1.40.0

@continuedev/config-yaml@1.40.0

31. Januar 2026

# [1.40.0](https://github.com/continuedev/continue/compare/@continuedev/config-yaml@1.39.0...@continuedev/config-yaml@1.40.0) (2026-01-31) ### Bug Fixes * Berücksichtigung verschiedener Tokenizer ([b4bc3ae](https://github.com/continuedev/continue/commit/b4bc3aedaa2e44149a1898c609405ebb80ca91b2)) * fehlender cancelStream-Aufruf und Rückgabe für nicht-retrybare Fehler hinzufügen ([c391353](https://github.com/continuedev/continue/commit/c391353709ad4367fb333680190e0ba73c868ef0)) * auch zugrunde liegenden Anbieter für offenes GitHub-Issue anzeigen ([0b7fe48](https://github.com/continuedev/continue/commit/0b7fe4813c996aa416843bbda7d1c784b49e65b4)) * **cli:** flaky Hub-Lade-Tests ([2fb8830](https://github.com/continuedev/continue/commit/2fb88307caeb6e72979396f70de17af087a27536)) * Befehl in Contribution-Datei korrigiert ([#9715](https://github.com/continuedev/continue/issues/9715)) ([a21eb15](https://github.com/continuedev/continue/commit/a21eb151d30969c292e290523941ebb520c61151)) * kompakte Unterhaltung mit schwebenden Tool-Aufrufen ([b33b8a4](https://github.com/continuedev/continue/commit/b33b8a4dadb821cab67d517223cb9ab19268d173)) * Updates zur Kompaktierung Teil 2 ([d18c6f0](https://github.com/continuedev/continue/commit/d18c6f04ad6ca518f849a78ab6cb8938e8ea1c24)) * core/package.json zur Reduzierung von Verwundbarkeiten ([8ca79b9](https://github.com/continuedev/continue/commit/8ca79b9a5adefa401b6bd7cc62e72e1597cb0012)) * Duplizierte Tool-Nachrichten ([77a2f12](https://github.com/continuedev/continue/commit/77a2f1274a6183bf43cb8df3d84e8e58762b1442)) * eslint ([50adea3](https://github.com/continuedev/continue/commit/50adea309b6635b125e55a13bb99b0a91827956f)) * eslint-Import ungelöst ([ccd0b74](https://github.com/continuedev/continue/commit/ccd0b74dc544687b3ec9439ee9a814efbacc5752)) * Linting-Problem Mising parallel count ([c0a9309](https://github.com/continuedev/continue/commit/c0a930919fe93ce7ad6a0eaf3546a956dee02383)) * Lokales Setup-Bildschirm-Input-Text ist schwarz auf dunklem Hintergrund ([363292b](https://github.com/continuedev/continue/commit/363292bdef47b14857a5a4dfa6b3f5a115510c31)) * GITHUB_TOKEN an vscode-E2E-Tests für Ripgrep-Download übergeben ([b8586d6](https://github.com/continuedev/continue/commit/b8586d6d90a6ee76d7186f53ea6f6521aee680f5)) * vorausgehenden Inhalt an RegistryClient für WSL-Kompatibilität übergeben ([#9739](https://github.com/continuedev/continue/issues/9739)) ([38aae62](https://github.com/continuedev/continue/commit/38aae62e56b1ffc48da626544bb512f03c19e8fe)), schließt [#6242](https://github.com/continuedev/continue/issues/6242) [#7810](https://github.com/continuedev/continue/issues/7810) * Warten auf das Laden der Sitzung aus dem Verlauf verhindern ([9fcc332](https://github.com/continuedev/continue/commit/9fcc332a99095a2b5635eb786ea02ca73311803a)) * Replace console.debug durch logger im Exit-Tool ([b65f440](https://github.com/continuedev/continue/commit/b65f44027a9c36d02e930ed2be41bac274ceb312)) * IDE-Erweiterungen-Tab-Navigation in der Dokumentation lösen ([7129415](https://github.com/continuedev/continue/commit/7129415db0a6609e126f464e3dfe0d92528fddf2)) * MDX-Parsing-Fehler im Run-agents-locally-Leitfaden beheben ([7074a50](https://github.com/continuedev/continue/commit/7074a50e4a39f5b29927c8c90a4fb52769b64dff)) * Setze isComplete-Metadaten, wenn Agenten die Ausführung beenden ([65a1e15](https://github.com/continuedev/continue/commit/65a1e1500b112de4e24364eff890bd9b4fcac895)) * vollständige rechte Seite-Verlauf-Linienschrift anzeigen ([7f46011](https://github.com/continuedev/continue/commit/7f46011e7c277d570907cfc733639f4a2f944d1d)) * Konfigurationsfehler vereinfacht ([e3a8711](https://github.com/continuedev/continue/commit/e3a8711ed90b1c3026726c80fecff92e449bb79b)) * cmd.exe-Wrapping für MCP-Server überspringen, wenn Windows-Host mit WSL verbindet ([1219985](https://github.com/continuedev/continue/commit/1219985ce13753b38747a0a258a39fbd0a5d1dd6)), schließt [#9151](https://github.com/continuedev/continue/issues/9151) [#9679](https://github.com/continuedev/continue/issues/9679) * Objekt-Typ-Argumente in Tool-Aufruf-Parsing unterstützen ([82f6f69](https://github.com/continuedev/continue/commit/82f6f6952202924952af6b9c84df42e5fd204bc3)) * Tool-Berechtigungen für MCP und Bash im Headless-Modus ([998cb76](https://github.com/continuedev/continue/commit/998cb76a7128e47fc94dad7a282a578459ee8015)) * Link zu ide-extensions auf Quick-Start-Seite aktualisieren ([b73fefe](https://github.com/continuedev/continue/commit/b73fefe7000cedd45431ca4cdade539bee1d0832)) * upgrade @c15t/react von 1.8.1 auf 1.8.2 ([9b192e4](https://github.com/continuedev/continue/commit/9b192e478420d9d21baee4da86e6fb95f91cfd36)) * upgrade @huggingface/jinja von 0.1.3 auf 0.5.3 ([e5afc41](https://github.com/continuedev/continue/commit/e5afc41492f1c837156a37e3427c3cc27db93ab5)) * upgrade @tiptap/extension-mention von 2.26.1 auf 2.27.1 ([8835b11](https://github.com/continuedev/continue/commit/8835b1107bdeb461302c8115b55c34f3405210c8)) * upgrade @tiptap/extension-paragraph von 2.26.1 auf 2.27.1 ([43a759d](https://github.com/continuedev/continue/commit/43a759d72ded5c67eca9f175a83cc6579aba0c1f)) * upgrade @tiptap/extension-text von 2.26.1 auf 2.27.1 ([aeee9b9](https://github.com/continuedev/continue/commit/aeee9b967016a9d9095e30c7026825ee28802f6f)) * upgrade posthog-js von 1.297.2 auf 1.310.1 ([14a5336](https://github.com/continuedev/continue/commit/14a533617f4e0811e4f75a2652b44453c640f82e)) * upgrade react-hook-form von 7.62.0 auf 7.69

OpenClawv2026.1.30

openclaw 2026.1.30

31. Januar 2026

### Änderungen - CLI: füge Befehl `completion` hinzu (Zsh/Bash/PowerShell/Fish) und automatische Einrichtung während Post-Install/Onboarding. - CLI: füge pro-Agent `models status` (`--agent` Filter) hinzu. (#4780) Danke @jlowin. - Agents: füge Kimi K2.5 zum synthetischen Modellkatalog hinzu. (#4407) Danke @manikv12. - Auth: Kimi Coding auf integrierten Anbieter umstellen; OAuth-Profil-E-Mail normalisieren. - Auth: füge MiniMax OAuth-Plugin + Onboarding-Option hinzu. (#4521) Danke @Maosghoul. - Agents: aktualisiere pi SDK/API-Nutzung und Abhängigkeiten. - Web UI: Sitzungen nach Chat-Befehlen neu laden und Anzeigenamen von Sitzungen verbessern. - Build: verschiebe TypeScript-Builds zu `tsdown` + `tsgo` (schnellere Builds, CI-Typprüfungen), aktualisiere Ziel in tsconfig und säubere Lint-Regeln. - Build: richte npm-Tar-Override und Bin-Metadaten aus, damit der `openclaw` CLI-Einstiegspunkt in npm-Veröffentlichungen erhalten bleibt. - Dokumentation: füge pi/pi-dev-Dokumentation hinzu und aktualisiere OpenClaw Branding + Installationslinks. ### Fehlerbehebungen - Sicherheit: beschränke lokale Pfad-Extraktion im Medienparser, um LFI zu verhindern. (#4880) - Gateway: verhindere, dass Token-Standards wörtlich „undefined“ werden. (#4873) Danke @Hisleren. - Control UI: behebe Auflösung von Assets bei npm-Global-Installationen. (#4909) Danke @YuriNachos. - macOS: vermeide Backpressure im STDERR-Pipe bei Gateway-Erkennung. (#3304) Danke @abhijeet117. - Telegram: normalisiere Token-Suche für nicht-normalisierte IDs. (#5055) Danke @jasonsschin. - Telegram: Behalte Deliver-Thread-Fallback bei und behebe threadId-Handling im Delivery-Kontext. - Telegram: behebe HTML-Verschachtelung bei sich überschneidenden Stilen/Links. (#4578) Danke @ThanhNguyxn. - Telegram: akzeptiere numerische messageId/chatId in React-Aktionen. (#4533) Danke @Ayush10. - Telegram: respektiere plattformweite Proxy-Dispatcher pro Konto über undici fetch. (#4456) Danke @spiceoogway. - Telegram: begrenze Skill-Befehle auf dem gebundenen Agenten pro Bot. (#4360) Danke @robhparker. - BlueBubbles: entprellen nach messageId, um Anhänge in Text+Bild-Nachrichten zu erhalten. (#4984) - Routing: bevorzuge requesterOrigin gegenüber veralteten Sitzungs-Einträgen für Sub-Agenten-Ankündigungszustellung. (#4957) - Erweiterungen: wiederherstelle eingebettete Erweiterungs-Erkennungs-Typen. - CLI: behebe Portauflösung bei `tui:dev`. - LINE: behebe TypeError im Status-Befehl. (#4651) - OAuth: überspringe Warnungen zu abgelaufenen Tokens, wenn Refresh Tokens noch gültig sind. (#4593) - Build: überspringe redundanten UI-Installationsschritt im Dockerfile. (#4584) Danke @obviyus.

OpenCodev1.1.48

v1.1.48

31. Januar 2026

## Core - Sync changes - Allow specifying custom models file path via OPENCODE_MODELS_PATH environment variable - Ensure models configuration is not empty before loading - Make skills invokable as slash commands in the TUI - Don't follow symbolic links by default in grep and ripgrep operations - Prevent parallel test runs from contaminating environment variables - Ensure Mistral ordering fixes also apply to Devstral - Add Copilot-specific provider to properly handle reasoning tokens (@SteffenDE) - Respect OPENCODE_MODELS_URL environment variable in build process (@bbartels) - Use snake_case for thinking parameter with OpenAI-compatible APIs (@Chesars) - Bump AI SDK packages - Ensure ask question tool isn't included when using acp - Handle redirected statement treesitter node in bash permissions (@pschiel) - Remove special case handling for Google Vertex Anthropic provider in response generation (@MichaelYochpaz) - Exclude chat models from textVerbosity setting ## Desktop - Revert transitions, spacing, scroll fade, and prompt area updates - Add session actions tests (@neriousy) - Refactored tests and added project tests (@neriousy) **Thank you to 7 community contributors:** - @neriousy: - refactor(app): refactored tests + added project tests (#11349) - test(app): session actions (#11386) - @MichaelYochpaz: - refactor(provider): remove google-vertex-anthropic special case in ge… (#10743) - @pschiel: - fix: handle redirected_statement treesitter node in bash permissions (#6737) - @IdrisGit: - docs: update agents options section to include color and top_p options (#11355) - @Chesars: - fix(provider): use snake_case for thinking param with OpenAI-compatible APIs (#10109) - @bbartels: - feat(build): respect OPENCODE_MODELS_URL env var (#11384) - @SteffenDE: - feat(opencode): add copilot specific provider to properly handle copilot reasoning tokens (#8900)

Codex CLIvrust-v0.93.0

0.93.0

31. Januar 2026

## New Features - Added an optional SOCKS5 proxy listener with policy enforcement and config gating. (#9803) - Plan mode now streams proposed plans into a dedicated TUI view, plus a feature-gated `/plan` shortcut for quick mode switching. (#9786, #10103) - Added `/apps` to browse connectors in TUI and `$` insertion for app prompts. (#9728) - App-server can now run in external auth mode, accepting ChatGPT auth tokens from a host app and requesting refreshes when needed. (#10012) - Smart approvals are now enabled by default, with explicit approval prompts for MCP tool calls. (#10286, #10200) - Introduced a SQLite-backed log database with an improved logs client, thread-id filtering, retention, and heuristic coloring. (#10086, #10087, #10150, #10151, #10229, #10228) ## Bug Fixes - MCP tool image outputs render reliably even when image blocks aren’t first or are partially malformed. (#9815) - Input history recall now restores local image attachments and rich text elements. (#9628) - File search now tracks session CWD changes and supports multi-root traversal with better performance. (#9279, #9939, #10240) - Resuming a thread no longer updates `updated_at` until the first turn actually starts. (#9950) - Shell snapshots no longer inherit stdin, avoiding hangs from startup scripts. (#9735) - Connections fall back to HTTP when WebSocket proxies fail. (#10139) ## Documentation - Documented app-server AuthMode usage and behavior. (#10191) ## Chores - Upgraded the Rust toolchain to 1.93. (#10080) - Updated pnpm versions used in the repo. (#9992, #10161) - Bazel build and runfiles handling improvements, including remote cache compression. (#10079, #10098, #10102, #10104) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.92.0...rust-v0.93.0 - #9988 nit: better tool description @jif-oai - #9991 nit: better unused prompt @jif-oai - #9994 chore: clean orchestrator prompt @jif-oai - #10001 backend-client: add get_config_requirements_file @gt-oai - #9992 update pnpm to 10.28.2 to address security issues @mjr-openai - #9993 description in role type @jif-oai - #9944 TUI footer: right-align context and degrade shortcut summary + mode cleanly @charley-oai - #9803 feat(network-proxy): add a SOCKS5 proxy with policy enforcement @viyatb-oai - #9950 fix(app-server, core): defer initial context write to rollout file until first turn @owenlin0 - #10003 feat: make it possible to specify --config flags in the SDK @bolinfest - #9797 remove sandbox globals. @iceweasel-oai - #10009 chore: introduce *Args types for new() methods @bolinfest - #10011 really fix pwd for windows codex zip @iceweasel-oai - #10007 Remove load from SKILL.toml fallback @alexsong-oai - #10008 enable live web search for DangerFullAccess sandbox policy @sayan-oai - #9815 Fix: Render MCP image outputs regardless of ordering @Kbediako - #9654 Show OAuth error descriptions in callback responses @blevy-oai - #10030 Clarify external editor env var message @joshka-oai - #9949 Ask user question UI footer improvements @charley-oai - #9359 tui: stabilize shortcut overlay snapshots on WSL @slkzgm - #10040 fix: enable per-turn updates to web search mode @sayan-oai - #10041 fix: allow unknown fields on Notice in schema @sayan-oai - #9628 Restore image attachments/text elements when recalling input history (Up/Down) @charley-oai - #9982 [skills] Auto install MCP dependencies when running skils with dependency specs. @mzeng-openai - #9986 fix(core) info cleanup @dylan-hurd-oai - #9941 error code/msg details for failed elevated setup @iceweasel-oai - #9489 feat(core) RequestRule @dylan-hurd-oai - #10026 Add exec policy TOML representation @gt-oai - #9821 feat: codex exec auto-subscribe to new threads @jif-oai - #10004 feat: sqlite 1 @jif-oai - #10083 feat: sort metadata by date @jif-oai - #10092 Update shell-tool-mcp.yml @dylan-hurd-oai - #10079 [bazel] Enable remote cache compression @zbarsky-openai - #10025 Refine request_user_input TUI interactions and option UX @charley-oai - #10080 Upgrade to rust 1.93 @zbarsky-openai - #10095 Update shell-tool-mcp.yml @dylan-hurd-oai - #9939 file-search: improve file query perf @nornagon-openai - #10097 chore: deprecate old web search feature flags @sayan-oai - #10034 compaction @aibrahim-oai - #10028 allow elevated sandbox to be enabled without base experimental flag @iceweasel-oai - #10101 fix(ci) fix shell-tool-mcp version v2 @dylan-hurd-oai - #10043 Added `tui.notifications_method` config option @etraut-openai - #10104 [bazel] Fix the build @zbarsky-openai - #10102 default enable compression, update test helpers @sayan-oai - #10111 fix(ci) more shell-tool-mcp issues @dylan-hurd-oai - #10115 update the ci pnpm workflow for shell-tool-mcp to use corepack for pnpm versioning @mjr-openai - #10098 [bazel] Improve runfiles handling @zbarsky-openai - #10129 Ensure auto-compaction starts after turn started @aibrahim-oai - #10116 chore(config) personality as a feature @dylan-hurd-oai - #10123 Add app-server compaction item notifications tests @aibrahim-oai - #10114 chore(config) Update personality instructions @dylan-hurd-oai - #10128 removing quit from dropdown menu, but not autocomplete [cli] @natea-oai - #9728 [connectors] Support connectors part 2 - slash command and tui @mzeng-openai - #10133 chore(core) personality under development @dylan-hurd-oai - #10125 emit a metric when we can't spawn powershell @iceweasel-oai - #10134 fix(tui) reorder personality command @dylan-hurd-oai - #10135 fix(ci) missing package.json for shell-mcp-tool @dylan-hurd-oai - #10131 fix: ignore key release events during onboarding @joshka-oai - #10138 fix: remove references to corepack @bolinfest - #10086 feat: add log db @jif-oai - #10089 feat: async backfill @jif-oai - #10087 feat: log db client @jif-oai - #10149 chore: improve client @jif-oai - #10161 nit: update npm @jif-oai - #10163 [experimental] nit: try to speed up apt-install @jif-oai - #10164 [experimental] nit: try to speed up apt-install 2 @jif-oai - #10150 feat: adding thread ID to logs + filter in the client @jif-oai - #10151 feat: add log retention and delete them after 90 days @jif-oai - #10152 chore: unify log queries @jif-oai - #10175 Add OpenAI docs MCP tooltip @joshka-oai - #10171 feat: reduce span exposition @jif-oai - #10139 Fall back to http when websockets fail @pakrym-oai - #10140 chore: ensure pnpm-workspace.yaml is up-to-date @bolinfest - #9017 Better handling skill depdenencies on ENV VAR. @xl-openai - #10182 fix: unify `npm publish` call across shell-tool-mcp.yml and rust-release.yml @bolinfest - #10180 Add features enable/disable subcommands @joshka-oai - #10184 fix: /approvals -> /permissions @bolinfest - #10179 Remove WebSocket wire format @pakrym-oai - #10186 updating comment to better indicate intent of skipping `quit` in the main slash command menu @natea-oai - #10118 [Codex][CLI] Show model-capacity guidance on 429 @ccy-oai - #10012 feat(app-server): support external auth mode @owenlin0 - #10103 tui: add feature-gated /plan slash command to switch to Plan mode @charley-oai - #10191 chore(app-server): document AuthMode @owenlin0 - #10130 [feat] persist dynamic tools in session rollout file @celia-oai - #10181 add error messages for the go plan type @willwang-openai - #10198 feat(tui): route employee feedback follow-ups to internal link @joshka-oai - #10194 load from yaml @alexsong-oai - #10147 chore(personality) new schema with fallbacks @dylan-hurd-oai - #10200 MCP tool call approval (simplified version) @mzeng-openai - #10154 feat: add output to `/ps` @jif-oai - #10217 nit: actually run tests @jif-oai - #10177 Add community links to startup tooltips @joshka-oai - #10210 Chore: plan mode do not include free form question and always include isOther @shijie-oai - #10218 feat: backfill timing metric @jif-oai - #10220 chore: unify metric @jif-oai - #8991 Conversation naming @pap-openai - #10167 Fetch Requirements from cloud @gt-oai - #10225 explorer prompt @jif-oai - #10222 fix: make sure the shell exists @jif-oai - #10232 chore: do not clean the DB anymore @jif-oai - #10229 feat: improve logs client @jif-oai - #10228 feat: heuristic coloring of logs @jif-oai - #10237 nit: fix db with multiple metadata lines @jif-oai - #10208 feat: refactor CodexAuth so invalid state cannot be represented @bolinfest - #10212 chore(feature) Experimental: Personality @dylan-hurd-oai - #10211 chore(feature) Experimental: Smart Approvals @dylan-hurd-oai - #10190 Load exec policy rules from requirements @gt-oai - #10195 plan mode: add TL;DR checkpoint and client behavior note @baumann-oai - #10239 chore: fix the build breakage that came from a merge race @bolinfest - #9786 Plan mode: stream proposed plans, emit plan items, and render in TUI @charley-oai - #10063 Tui: hide Code mode footer label @charley-oai - #10244 chore: rename ChatGpt -> Chatgpt in type names @bolinfest - #10238 Plan mode prompt @aibrahim-oai - #10251 Fix deploy @charley-oai - #10255 plan prompt @aibrahim-oai - #10253 Make plan highlight use popup grey background @charley-oai - #10207 Skip loading codex home as project layer @daniel-oai - #10256 Update copy @pakrym-oai - #9735 core: prevent shell_snapshot from inheriting stdin @swordfish444 - #10262 Fix main @pakrym-oai - #10265 Hide /approvals from the slash-command list @pakrym-oai - #10267 Update announcement_tip.toml @pakrym-oai - #10240 file-search: multi-root walk @nornagon-openai - #10266 fix: dont auto-enable web_search for azure @sayan-oai - #9279 fix: update file search directory when session CWD changes @yuvrajangadsingh - #10249 Validate CODEX_HOME before resolving @etraut-openai - #10272 chore: implement Mul for TruncationPolicy @bolinfest - #10241 Wire up cloud reqs in exec, app-server @gt-oai - #10263 Add enforce_residency to requirements @gt-oai - #10276 add missing fields to WebSearchAction and update app-server types @sayan-oai - #10283 Turn on cloud requirements for business too @gt-oai - #10287 Fix minor typos in comments and documentation @ruyut - #10286 feat(core) Smart approvals on @dylan-hurd-oai

OpenCodev1.1.47

v1.1.47

30. Januar 2026

## Core - I need to see the actual commit message to summarize it. Could you please provide the commit message you'd like me to summarize for the changelog entry?

Claude Codev2.1.27

v2.1.27

30. Januar 2026

## What's changed - Added tool call failures and denials to debug logs - Fixed context management validation error for gateway users, ensuring `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1` avoids the error - Added `--from-pr` flag to resume sessions linked to a specific GitHub PR number or URL - Sessions are now automatically linked to PRs when created via `gh pr create` - Fixed /context command not displaying colored output - Fixed status bar duplicating background task indicator when PR status was shown - VSCode: Enabled Claude in Chrome integration - Permissions now respect content-level `ask` over tool-level `allow`. Previously `allow: ["Bash"], ask: ["Bash(rm *)"]` allowed all bash commands, but will now permission prompt for `rm`. - Windows: Fixed bash command execution failing for users with `.bashrc` files - Windows: Fixed console windows flashing when spawning child processes - VSCode: Fixed OAuth token expiration causing 401 errors after extended sessions

OpenCodev1.1.46

v1.1.46

30. Januar 2026

## Core - Remove unused experimental keys from TUI (@IdrisGit) - Add continuous integration configuration - Remove AI SDK middleware that was preventing think blocks from being sent back as assistant message content ## Desktop - Change language test in app (@neriousy) - Add transitions, spacing improvements, scroll fade effects, and prompt area updates (@aaroniker) **Thank you to 3 community contributors:** - @aaroniker: - feat: Transitions, spacing, scroll fade, prompt area update (#11168) - @neriousy: - test(app): change language test (#11295) - @IdrisGit: - chore(tui): remove unused experimental keys (#11195)

OpenCodev1.1.45

v1.1.45

30. Januar 2026

## Desktop - Free model layout improvements in app

OpenClawv2026.1.29

openclaw 2026.1.29

30. Januar 2026

- Rebrand: npm-Paket/CLI umbenennen zu `openclaw`, eine `openclaw`-Kompatibilitätshim hinzufügen und Extensions in den `@openclaw/*`-Bereich verschieben. - Onboarding: Sicherheitshinweis-Text für Beta verstärken + Erwartungen an Zugriffskontrollen. - Onboarding: Venice API-Schlüssel in den nicht-interaktiven Flow aufnehmen. (#1893) Danke @jonisjongithub. - Config: Legacy-State/Config-Pfade automatisch migrieren und Config-Auflösung konsistent über legacy Dateinamen hinweg halten. - Gateway: Warnung bei Hook-Tokens über Abfrageparameter; Dokumentation zur bevorzugten Header-Auth. (#2200) Danke @YuriNachos. - Gateway: gefährlichen Control UI-Geräte-Auth-Bypass-Flag + Audit-Warnungen hinzufügen. (#2248) - Doctor: Warnung bei Gateway-Exposition ohne Auth. (#2016) Danke @Alex-Alaniz. - Web UI: Sub-Agent-Ankündigungs-Antworten in WebChat sichtbar halten. (#1977) Danke @andrescardonas7. - Browser: Browser-Control über Gateway/Node routen; eigenständigen Browser-Control-Befehl und Config für Kontroll-URL entfernen. - Browser: `browser.request` über Node-Proxy routen, wenn verfügbar; Proxy-Timeouts beachten; Browser-Ports aus `gateway.port` ableiten. - Browser: Falls nötig auf URL-Abgleich für Extension-Relay-Zielauflösung zurückgreifen. (#1999) Danke @jonit-dev. - Telegram: Caption-Parameter für Media-Sendungen zulassen. (#1888) Danke @mguellsegarra. - Telegram: Plugin sendPayload channelData (Media/Buttons) unterstützen und Plugin-Befehle validieren. (#1917) Danke @JoshuaLelon. - Telegram: Block-Antworten vermeiden, wenn Streaming deaktiviert ist. (#1885) Danke @ivancasco. - Telegram: optionale stillen Send-Flag (Benachrichtigungen deaktivieren) hinzufügen. (#2382) Danke @Suksham-sharma. - Telegram: Bearbeitung gesendeter Nachrichten via message(action="edit") unterstützen. (#2394) Danke @marcelomar21. - Telegram: QUIRRE Antworten für Message-Tool und eingehenden Kontext unterstützen. (#2900) Danke @aduk059. - Telegram: Sticker-Empfang/-Versand mit Vision-Caching hinzufügen. (#2629) Danke @longjos. - Telegram: Sticker-Pixel an Vision-Modelle senden. (#2650) - Telegram: Topic-IDs in Neustart-Sentinel-Benachrichtigungen beibehalten. (#1807) Danke @hsrvc. - Discord: konfigurierbare privilegierte Gateway-Intents für Präsenz/Members hinzufügen. (#2266) Danke @kentaro. - Slack: ACK-Reaktion nach gestreamten Antworten löschen. (#2044) Danke @fancyboi999. - Matrix: Plugin-SDK auf @vector-im/matrix-bot-sdk umstellen. - Tlon: Thread-Antwort-IDs als @ud formatieren. (#1837) Danke @wca4a. - Tools: pro-Sender-Gruppen-Tool-Richtlinien hinzufügen und Priorität klären. (#1757) Danke @adam91holt. - Agents: zusammenfassende Darstellung der gelöschten Nachrichten während der Kompaktions-Sicherheitsprüfungen. (#2509) Danke @jogi47. - Agents: Cron-Tool-Beschreibung mit vollständigen Schema-Dokumenten erweitern. (#1988) Danke @tomascupr. - Agents: Tools.exec.safeBins in Exec-Allowlist-Prüfungen berücksichtigen. (#2281) - Memory Search: zusätzliche Pfade für Memory-Indexierung zulassen (Ignore-Symlinks). (#3600) Danke @kira-ariaki. - Skills: Multi-Bild-Eingabeunterstützung für Nano Banana Pro Skill hinzufügen. (#1958) Danke @tyler6204. - Skills: fehlende Abhängigkeits-Metadaten für GitHub, Notion, Slack, Discord ergänzen. (#1995) Danke @jackheuberger. - Commands: Ausgabe von /help und /commands mit Telegram-Paging gruppieren. (#2504) Danke @hougangdev. - Routing: pro-Konto DM-Sitzungsumfang hinzufügen und Multi-Konto-Isolation dokumentieren. (#3095) Danke @jarvis-sam. - Routing: Sitzungs-Schlüssel-Regex vorcompilieren. (#1697) Danke @Ray0907. - CLI: Node-Modul-Compile-Cache für schnelleren Start verwenden. (#2808) Danke @pi0. - Auth: nach ASCII-Eingabeaufforderung URL für Google-Auth kopierbar anzeigen. (#1787) Danke @robbyczgw-cla. - TUI: Breitenüberschuss beim Rendern von Auswahllisten vermeiden. (#1686) Danke @mossein. - macOS: OpenClaw-App-Umbenennung für macOS-Quellen, Bundle-IDs und gemeinsame Kit-Pfade abschließen. (#2844) Danke @fal3. - Branding: Launchd-Labels, mobile Bundle-IDs und Logging-Subsysteme auf bot.molt aktualisieren (Legacy-Migrationen von com.clawdbot). Danke @thewilloftheshadow. - macOS: Projekt-lokales `node_modules/.bin` PATH-Preference auf Debug-Builds begrenzen (reduziert PATH-Hijacking-Risiko). - macOS: benutzerdefinierte SSH-Benutzernamen im Remote-Target behalten. (#2046) Danke @algal. - macOS: Absturz beim Rendern von Code-Blöcken vermeiden durch Erhöhung von Textual auf 0.3.1. (#2033) Danke @garricn. - Update: dist/control-ui bei Dirty-Checks ignorieren und nach UI-Builds wiederherstellen. (#1976) Danke @Glucksberg. - Build: A2UI-Assets während des Builds bündeln und Generierung von Bundles aufhören. (#2455) Danke @0oAstro. - CI: Node-Heap-Größe für macOS-Checks erhöhen. (#1890) Danke @realZachi. - Config: config.env vor ${VAR}-Substitution anwenden. (#1813) Danke @spanishflu-est1918. - Gateway: bevorzugt neueste Sitzungs-Metadaten bei der Zusammenführung von Stores. (#1823) Danke @emanuelst. - Docs: Fly-Private-Deployment-Schritte straffen. (#2289) Danke @dguido. - Docs: Migrationsleitfaden für den Umzug auf eine neue Maschine hinzufügen. (#2381) - Docs: One-Click-Deployment-Guide für Northflank hinzufügen. (#2167) Danke @AdeboyeDN. - Docs: Verweis auf Vercel AI Gateway in der Providern-Seitenleiste hinzufügen. (#1901) Danke @jerilynzheng. - Docs: Render-Deployment-Guide hinzufügen. (#1975) Danke @anurag. - Docs: Claude Max API Proxy-Anleitung hinzufügen. (#1875) Danke @atalovesyou. - Docs: DigitalOcean-Deployment-Guide hinzufügen. (#1870) Danke @0xJonHoldsCrypto. - Docs: Oracle Cloud (OCI) Plattformleitfaden + Cross-Links hinzufügen. (#2333) Danke @hirefrank. - Docs: Raspberry Pi-Installationsanleitung hinzufügen. (#1871) Danke @0xJonHoldsCrypto. - Docs: GCP Compute Engine-Deployment-Guide hinzufügen. (#1848) Danke @hougangdev. - Docs: LINE-Kanal-Anleitung hinzufügen. Danke @thewilloftheshadow. - Docs: Credits beider Mitwirkende für Control UI-Refresh. (#1852) Danke @EnzeD. - Docs: Kopfzeile der Docs fixieren, damit die Navbar beim Scrollen sichtbar bleibt. (#2445) Danke @chenyuan99. - Docs: exe.dev-Installationsanleitung aktualisieren. ([#3047](https://github.com/openclaw/openclaw/pull/3047)) Danke @zackerthescar. ### Breaking - **BREAKING:** Gateways-Auth-Modus "none" entfernt; Gateway erfordert nun Token/Passwort (Tailscale Serve-Identität weiterhin erlaubt). ### Fixes - Telegram: stille leere Antworten vermeiden durch Verfolgung von Normalisierungssprüngen vor dem Fallback. (#3796) - Mentions: Berücksichtige mentionPatterns auch wenn explizite Mentions vorhanden sind. (#3303) Danke @HirokiKobayashi-R. - Discord: Benutzernamensverzeichnis-Auflösung in Zielauflösung wiederherstellen. (#3131) Danke @bonald. - Agents: MiniMax-Base-URL-Tests mit Standardanbieter-Konfiguration abgleichen. (#3131) Danke @bonald. - Agents: Wiederholungen bei zu großen Bildfehlern verhindern und Größenbeschränkungen offenlegen. (#2871) Danke @Suksham-sharma. - Agents: Provider-BaseUrl/api für Inline-Modelle erben. (#2740) Danke @l

OpenCodev1.1.44

v1.1.44

30. Januar 2026

No notable changes

OpenCodev1.1.43

v1.1.43

30. Januar 2026

No notable changes

Claude Codev2.1.25

v2.1.25

29. Januar 2026

## What's changed - Fixed beta header validation error for gateway users on Bedrock and Vertex, ensuring `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1` avoids the error

OpenCodev1.1.42

v1.1.42

29. Januar 2026

## Core - Revert to Anthropic completions endpoint to avoid rate limiting issues with copilot - Fix frontmatter adding newlines causing invalid model IDs - Ensure that Kimi doesn't have fake variants available - Add sequential numbering for forked session titles (@ariane-emory) ## TUI - Add additional timeout race condition guards - Guard destroyed input field in timeout callback - Include cache tokens in CLI statistics (@bold84) **Thank you to 2 community contributors:** - @bold84: - feat(cli): include cache tokens in stats (#10582) - @ariane-emory: - feat: Sequential numbering for forked session titles (Issue #10105) (#10321)

OpenCodev1.1.41

v1.1.41

29. Januar 2026

## Core - Ensure variants for Copilot models work with maxTokens being set - Fixed maxOutputTokens being accidentally hardcoded to undefined which caused kimi-k2.5 issues on some inference providers (@ideallove) - Include provider ID in SDK cache key to prevent cache collisions (@saba-ch) - Allow media-src data: URL for small audio files in Content Security Policy (@tanapok) - Update experimental environment variables in CLI documentation (@bbabou) - Add `ctx.abort` to grep tool - Add AbortSignal support to Ripgrep.files() and GlobTool (@goniz) ## Desktop - Make settings more responsive for mobile and small web/desktop (@DNGriffin) - Fixed type errors in app package (@Brendonovich) - Auto scroll to file tabs on open and enable mouse wheel scrolling (@alexyaroshuk) - Remove duplicate keys causing TypeScript failures (@eXamadeus) - Normalize Chinese punctuation for Chinese UI (@MaxMiksa) - Fill missing Chinese translation keys to avoid English fallback (@MaxMiksa) - Fix alignment and padding in dialogs (@alexyaroshuk) **Thank you to 12 community contributors:** - @R44VC0RP: - fix(script): remove highlights template from release notes (#11052) - @zerone0x: - fix(ui): allow KaTeX inline math to be followed by punctuation (#11033) - @alexyaroshuk: - fix(app): alignment and padding in dialogs (#10866) - fix(app): file tabs - auto scroll on open & scroll via mouse wheel (#11070) - @MaxMiksa: - fix(app): fill missing zh keys to avoid English fallback (#10841) - fix(opencode): normalize zh punctuation for Chinese UI (#10842) - fix(ui): improve zh duration display formatting (#10844) - @goniz: - fix(opencode): add `AbortSignal` support to `Ripgrep.files()` and `GlobTool` (#10833) - @eXamadeus: - fix(typescript errors): remove duplicate keys causing typescript failures (#11071) - @Brendonovich: - fix(app): types - @bbabou: - docs: update experimental environment variables in CLI docs (#11030) - @DNGriffin: - fix(app): make settings more responsive for mobile and small web/desktop (#10775) - @tanapok: - fix(opencode): allow media-src data: URL for small audio files (#11082) - @saba-ch: - fix(provider): include providerID in SDK cache key (#11020) - @ideallove: - fix: maxOutputTokens was accidentally hardcoded to undefined (#10995)

Claude Codev2.1.23

v2.1.23

29. Januar 2026

## What's changed - Added customizable spinner verbs setting (`spinnerVerbs`) - Fixed mTLS and proxy connectivity for users behind corporate proxies or using client certificates - Fixed per-user temp directory isolation to prevent permission conflicts on shared systems - Fixed a race condition that could cause 400 errors when prompt caching scope was enabled - Fixed pending async hooks not being cancelled when headless streaming sessions ended - Fixed tab completion not updating the input field when accepting a suggestion - Fixed ripgrep search timeouts silently returning empty results instead of reporting errors - Improved terminal rendering performance with optimized screen data layout - Changed Bash commands to show timeout duration alongside elapsed time - Changed merged pull requests to show a purple status indicator in the prompt footer - [IDE] Fixed model options displaying incorrect region strings for Bedrock users in headless mode

OpenCodev1.1.40

v1.1.40

28. Januar 2026

## Core - Add SubtaskPart with metadata reference (@YangTianz) - Add global configuration support to app - Global configuration updates in app package - Bump plugins ## TUI - Add streaming prop to markdown element (@remorses) ## Desktop - Close review pane when pressing escape key - Enable ctrl+n and ctrl+p for popover navigation in desktop app (@sairajchouhan) - Add Thai locale support (@natt-v) - Add custom provider support to app **Thank you to 4 community contributors:** - @natt-v: - feat(i18n): add th locale support (#10809) - @sairajchouhan: - fix(desktop): enable ctrl+n and ctrl+p for popover navigation (#10777) - @YangTianz: - fix: add SubtaskPart with metadata reference (#10990) - @remorses: - fix(markdown): Add streaming prop to markdown element (#11025)

OpenCodev1.1.39

v1.1.39

28. Januar 2026

## Core - Ensure Copilot plugin properly sets headers for new messages API

OpenCodev1.1.38

v1.1.38

28. Januar 2026

## Core - Prevent AGENTS.md from being loaded multiple times during parallel tool execution - Bump plugin version - Ensure unsubscribe from PartUpdated is always called in TaskTool (@goniz) - Restore brand integrity of CLI wordmark (@mynameistito) - Add experimental OpenTUI markdown component to app (@remorses) **Thank you to 4 community contributors:** - @remorses: - feat(app): use opentui markdown component behind experimental flag (#10900) - @jamesmurdza: - docs: add Daytona OpenCode plugin to ecosystem (#10917) - @mynameistito: - fix(cli): restore brand integrity of CLI wordmark (#10912) - @goniz: - fix(opencode): ensure unsub(PartUpdated) is always called in TaskTool (#9992)

Claude Codev2.1.22

v2.1.22

28. Januar 2026

## What's changed - Fixed structured outputs for non-interactive (-p) mode

OpenCodev1.1.37

v1.1.37

28. Januar 2026

## Core - Support headless authentication for ChatGPT/Codex (@rgodha24) - Add recommended topP and temperature settings for Kimi K2.5 model - Adjust retry check to be more defensive - Handle Venice cache creation tokens (@dpuyosa) - Add worktree to plugin tool context - Expose Instance.directory to custom tools - Use Instance.directory instead of process.cwd() in read tool - Fix reactive file tree to properly update when files change - Attach Anthropic beta headers when using Messages API for Copilot - Add retry logic to handle certain provider problems - Ensure OpenAI 404 errors are retried (@tim-smart) - Use Anthropic-compatible messages API for Anthropic models through Copilot - Upgrade OpenTUI to v0.1.75 - Dynamically resolve AGENTS.md files from subdirectories as agent explores them - Add agent description to OpenCode (@SteffenDE) - Fix query selector handling of non-Latin characters - Await SessionRevert.cleanup for shell to prevent race conditions (@noamzbr) - Don't override source in custom provider loaders - Add provider settings to app - Run start command after reset in app - Remove log statement ## TUI - Handle 4-5 digit codes in copy logic - Make diff wrapping toggle always available in command list and fix type error (@ariane-emory) - Add visual feedback for diff wrap and conceal toggles (@IdrisGit) - Adjust TUI syncing logic to prevent agents from being undefined or missing - Remove broken app.tsx command option - Move animations toggle to global System category (@ariane-emory) ## Desktop - Reintroduce review tab in app - Auto-scroll to keep relevant content in view - Add 'connect provider' button to the manage models dialog (@alexyaroshuk) - Fix terminal corruption issue in app - File tree not always loading in app - Fix outdated e2e test in app package - Do not auto-navigate to last project on app load - Add Tauri localization support to desktop application - Add internationalization support for Tauri desktop application - Better memory management in app - Don't connect to localhost through VPN - Show 5 highlights in app - Highlight selected change in the app - Set filetree padding to 6px - Align file tree change styling - Delay navigation tooltips - Adjust titlebar left spacing - Shorten navigation tooltips - Add filetree tooltips with diff labels - Color file tree change dots by diff kind - Improved app layout - Fix session diffs not always loading in app - New end-to-end smoke tests for app - Add translations to the app - Update settings in general settings - Swallow file search errors in app - Fix tooltips causing getComputedStyle errors in model select - Clean up Tailwind CSS versus pure CSS usage in app - Navigate to tabs when opening file in app - Open markdown links in external browser - Shared terminal ghostty-web instance for better performance - Cleanup connect provider timers - Don't keep parts in memory - Don't show session skeleton after workspace reset - Add max-width to tabs with text truncation - Use smaller close icon on tabs to match comment cards - Only show left border on plus button when sticky - Update review empty states to 14px font size and align select file empty state - Center file tree empty state with 32px top margin - Implement non-fatal error handling in app - Fix agent fallback colors in app - Added forward and back navigation buttons - Add missing tooltips in app - Auto-scroll button sometimes sticks in the app - Deduplicate Tauri configuration files (@Brendonovich) - Only show files in select dialog when clicking + tab - Disable magnification gestures on macOS desktop (@ysm-dev) - Add Connect provider in model selector (@ProdigyRahul) - Add file tree specification tests - Enable file watcher in app - Improve file tree performance in app - Add highlights feature - Add new release modal - Fix Zen mode disconnect not working in app - Disconnect Zen provider - Fix e2e test in app package - Disable tooltips in file tree tabs - Move file tree toggle button to the right side - Fade filetree guide lines on hover - Dim non-deep filetree guide lines in the app interface - Reduce file tree folder indent - Refine file tree row spacing and indentation - Use medium font weight for filetree items - Use chevron icons for filetree folders - Tighten file tree row spacing - Adjust filetree panel padding - Scope filetree pill tabs styling - Add Vercel AI Gateway provider description - Add provider descriptions to settings - Add transition to command palette - Add transition to select provider dialog - Add optional transition animations to dialog - Style view all button with interactive color and margin - Add models icon and use in settings UI - Add providers icon and use in settings - Use default cursor for environment provider text - Add hover text for environment-connected providers - Set provider row height to 56px - Update button styles and disconnect button size in UI - Fix session synchronization issue in application - Add missing i18n strings in app - Add session load limit to prevent excessive memory usage - Add end-to-end test for sidebar navigation - Replace signals with createStore in app - Add file tree mode to app - Add file tree view to app - Use focus-visible instead of focus to prevent sticky hover effect on click (@ProdigyRahul) - Better sidebar hover behavior when collapsed - Make sidebar full-height in app - Add full-height hover effect to sidebar - Add model settings to application - Add model settings interface to app - Handle non-tool call permissions in app - Default servers on web - Performance improvements in the app - Add missing translations for status messages - Don't allow workspaces in non-VCS projects **Thank you to 17 community contributors:** - @MartinWie: - docs: fix permission event name (permission.asked not permission.updated) (#10588) - @ariane-emory: - fix(tui): Move animations toggle to global System category (resolves #10495) (#10497) - fix: Make diff wrapping toggle always available in command_list and correct a type error (resolves #10682) (#10683) - @R44VC0RP: - Add highlight tag parsing for changelog with video support - Add collapsible sections, sticky version header, and style refinements for changelog highlights - fix(web): update spacing on the changelog page - docs: add warning about Claude Pro/Max subscription support (#10595) - fix: move changelog footer outside content div to fix padding (#10712) - fix: add 44px top padding to sticky version header on changelog (#10715) - feat(release): add highlights template to draft releases (#10745) - feat: add /learn command to extract session learnings to scoped AGENTS.md files (#10717) - @IdrisGit: - fix(tui): add visual feedback for diff wrap and conceal toggles (#10655) - @noamzbr: - fix: await SessionRevert.cleanup for shell (#10669) - @SteffenDE: - feat(opencode): add agent description (#10680) - @sam-huckaby: - fix(app): Order themes alphabetically (#10698) - @alexyaroshuk: - fix(app): restore external link opening in system browser (#10697) - feat(app): add 'connect provider' button to the manage models dialog (#10887) - @ProdigyRahul: - fix(ui): use focus-visible instead of focus to prevent sticky hover effect on click (#10651) - fix(app): add connect provier in model selector (#10706) - @zerone0x: - fix(enterprise): add message navigation to share page desktop view (#10071) - @tim-smart: - fix: ensure openai 404 errors are retried (#10590) - @fabiomartino: - docs: add Italian README (#10732) - @ysm-dev: - fix(desktop): disable magnification gestures on macOS (#10605) - @Brendonovich: - desktop: deduplicate tauri configs - @dpuyosa: - feat(opencode): Handle Venice cache creation tokens (#10735) - @OpeOginni: - chore(docs): Better explanation on how to allow tools in external directories (#10862) - @rgodha24: - feat: support headless authentication for chatgpt/codex (#10890)

Claude Codev2.1.21

v2.1.21

28. Januar 2026

## What's changed - Added support for full-width (zenkaku) number input from Japanese IME in option selection prompts - Fixed shell completion cache files being truncated on exit - Fixed API errors when resuming sessions that were interrupted during tool execution - Fixed auto-compact triggering too early on models with large output token limits - Fixed task IDs potentially being reused after deletion - Fixed file search not working in VS Code extension on Windows - Improved read/search progress indicators to show "Reading…" while in progress and "Read" when complete - Improved Claude to prefer file operation tools (Read, Edit, Write) over bash equivalents (cat, sed, awk) - [VSCode] Added automatic Python virtual environment activation, ensuring `python` and `pip` commands use the correct interpreter (configurable via `claudeCode.usePythonEnvironment` setting) - [VSCode] Fixed message action buttons having incorrect background colors

Gemini CLIv0.26.0

Release v0.26.0

28. Januar 2026

## Was sich geändert hat * fix: PDF token estimation (#16494) von @korade-krushna in https://github.com/google-gemini/gemini-cli/pull/16527 * chore(release): Version auf 0.26.0-nightly.20260114.bb6c57414 erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/16604 * docs: F12 zum Öffnen der Debug-Konsole klären von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/16570 * docs: Entferne .md-Erweiterung aus internen Links in architecture.md von @medic-code in https://github.com/google-gemini/gemini-cli/pull/12899 * Add an experimental setting for extension config von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/16506 * feat: Rewind Confirmation-Dialog und Rewind Viewer-Komponente hinzufügen von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/15717 * fix(a2a): Werfe keine Fehler für GeminiEventType Retry und InvalidStream von @ehedlund in https://github.com/google-gemini/gemini-cli/pull/16541 * prefactor: rootCommands als Array hinzufügen, damit es für Policy-Parsing verwendet werden kann von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/16640 * remove unnecessary `\x7f` key bindings von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16646 * docs(skills): use body-file in pr-creator skill für bessere Zuverlässigkeit von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/16642 * chore(automation): rekursive Kennzeichnung für Nachfahren des Workstreams von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16609 * feat: Einführung des eingebauten Skill „skill-creator“ und CJS-Verwaltungstools von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16394 * chore(automation): Entferne automatisierte PR-Größen- und Komplexitätskennzeichnung von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16648 * refactor(skills): ersetze 'project' durch 'workspace'-Geltung von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16380 * Docs: Release notes für 13.01.2026 aktualisieren von @jkcinouye in https://github.com/google-gemini/gemini-cli/pull/16583 * Simplify paste handling von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16654 * chore(automation): verbesserte Planung zur Issue-Triage-Erkennung und Durchsatz von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16652 * fix(acp): Exit-Cleanup ausführen, wenn stdin schließt von @codefromthecrypt in https://github.com/google-gemini/gemini-cli/pull/14953 * feat(scheduler): Typen für ereignisgesteuerten Scheduler hinzufügen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/16641 * Remove unused rewind key binding von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16659 * Remove sequence binding von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16664 * feat(cli): deprecatiert das --prompt-Flag wieder aufheben von @alexaustin007 in https://github.com/google-gemini/gemini-cli/pull/13981 * chore: Dependabot-Konfiguration aktualisieren von @cosmopax in https://github.com/google-gemini/gemini-cli/pull/13507 * feat(config): Add 'auto' Alias für Standardmodell-Auswahl von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16661 * Enable & disable agents von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16225 * cleanup: Keybindings verbessern von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16672 * Add timeout for shell-utils to prevent hangs. von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/16667 * feat(plan): experimentelles Plan-Flag hinzufügen von @jerop in https://github.com/google-gemini/gemini-cli/pull/16650 * feat(cli): Sicherheitszustimmungsabfragen für Skill-Installation hinzufügen von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16549 * fix: Ersetze 3 aufeinanderfolgende Punkte durch Auslassungszeichen von @Vist233 in https://github.com/google-gemini/gemini-cli/pull/16587 * chore(automation): sicherstellen, dass status/need-triage angewendet wird und niemals automatisch gelöscht wird von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16657 * fix: Doppelpunkte in Skill-Beschreibung Frontmatter behandeln von @maru0804 in https://github.com/google-gemini/gemini-cli/pull/16345 * refactor(core): Robustheit der Skill-Frontmatter-Parse-Verarbeitung von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16705 * feat(skills): Konflikt-Erkennung und Warnungen für Skill-Overrides von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16709 * feat(scheduler): SchedulerStateManager für reaktiven Tool-Status von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/16651 * chore(automation): Berechtigung der 'help wanted'-Kennzeichnung erzwingen und Richtlinien aktualisieren von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16707 * fix(core): zirkuläre Abhängigkeit über tsconfig-Pfade lösen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16730 * chore/release: Version auf 0.26.0-nightly.20260115.6cb3ae4e0 erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/16738 * fix(automation): Status/need-issue-Kennzeichnungspass Wildcard korrigieren von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16727 * fix(automation): Verhindern von Loop beim Label-Enforcer durch Ignorieren aller Bots von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16746 * Add links to supported locations und kleine Fixes von @g-samroberts in https://github.com/google-gemini/gemini-cli/pull/16476 * feat(policy): Source Tracking zu Policy-Regeln hinzufügen von @allenhutchison in https://github.com/google-gemini/gemini-cli/pull/16670 * feat(automation): '🔒 maintainer only'-Rechte erzwingen und Bot-Loop beheben von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16751 * Make merged settings non-nullable und alle Lints diesbezüglich beheben von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/16647 * fix(core): ModelInfo-Ereignis emisiones bei abgebrochenem Signal verhindern von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16752 * Replace relative paths, um Website-Build zu reparieren von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/16755 * Restricting to localhost von @cocosheng-g in https://github.com/google-gemini/gemini-cli/pull/16548 * fix(cli): explizite Abhängigkeit auf color-convert hinzufügen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16757 * fix(automation): robuste Kennzeichnungsdurchsetzung mit Berechtigungsprüfungen von @bdmorgan in https://github.com/google-gemini/gemini-cli/pull/16762 * fix(cli): OOM-Absturz verhindern durch Begrenzung der Dateisuche und Timeout hinzufügen von @galz10 in https://github.com/google

Codex CLIvrust-v0.92.0

0.92.0

27. Januar 2026

## New Features - API v2 threads can now inject dynamic tools at startup and route their calls/responses end-to-end through the server and core tool pipeline. (#9539) - Added filtering on the thread list in the app server to make large thread sets easier to browse. (#9897) - Introduced a `thread/unarchive` RPC to restore archived rollouts back into active sessions. (#9843) - MCP servers can now define OAuth scopes in `config.toml`, reducing the need to pass `--scopes` on each login. (#9647) - Multi-agent collaboration is more capable and safer, with an explorer role, better collab event mapping, and max-depth guardrails. (#9817, #9818, #9918, #9899) - Cached `web_search` is now the default client behavior. (#9974) ## Bug Fixes - Fixed a TUI deadlock/freeze under high streaming throughput by avoiding blocking sends on the main Tokio thread. (#9951) - The `web_search` tool now handles and displays all action types, and shows in-progress activity instead of appearing stuck. (#9960) - Reduced high CPU usage in collaboration flows by eliminating busy-waiting on subagents. (#9776) - Fixed `codex resume --last --json` so prompts parse correctly without conflicting argument errors. (#9475) - Windows sandbox logging now handles UTF-8 safely, preventing failures when truncating multibyte content. (#8647) - `request_user_input` is now rejected outside Plan/Pair modes to prevent invalid tool calls. (#9955) ## Documentation - Updated the contribution guidelines for clearer onboarding and workflow expectations. (#9933) - Refreshed protocol/MCP docs to reflect `thread/unarchive` and the updated `request_user_input` question shape. (#9843, #9890) ## Chores - Self-update via Homebrew now uses an explicit cask upgrade command to avoid warnings and ambiguity. (#9823) - Release packaging now consistently writes the bundle zip to `dist/`. (#9934) - Updated key dependencies in the Rust workspace (including `axum`, `tracing`, `globset`, and `tokio-test`). (#9880, #9882, #9883, #9884) - Aligned feature stage naming with public maturity stages and added clearer warnings for underdevelopment features. (#9929, #9954) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.91.0...rust-v0.92.0 - #9850 chore: remove extra newline in println @SohailRaoufi - #9868 Adjust modes masks @aibrahim-oai - #9874 Prompt @aibrahim-oai - #9877 Plan prompt @aibrahim-oai - #9718 feat(tui) /personality @dylan-hurd-oai - #9871 chore(core) move model_instructions_template config @dylan-hurd-oai - #9539 feat: dynamic tools injection @jif-oai - #9818 feat: rebase multi-agent tui on `config_snapshot` @jif-oai - #9789 Fix flakey resume test @gt-oai - #9784 Fix flakey conversation flow test @gt-oai - #9918 feat: explorer collab @jif-oai - #9899 feat: disable collab at max depth @jif-oai - #9916 Fix up config disabled err msg @gt-oai - #9890 Feat: add isOther to question returned by request user input tool @shijie-oai - #9817 feat: codex exec mapping of collab tools @jif-oai - #9919 Fix flakey shell snapshot test @gt-oai - #9776 fix: attempt to reduce high cpu usage when using collab @eugeneoden - #9928 prompt @aibrahim-oai - #9925 chore: update interrupt message @jif-oai - #9843 Add thread/unarchive to restore archived rollouts @charley-oai - #9929 Aligned feature stage names with public feature maturity stages @etraut-openai - #9934 ensure codex bundle zip is created in dist/ @iceweasel-oai - #9897 [app-server] feat: add filtering on thread list @jif-oai - #9647 Add MCP server `scopes` config and use it as fallback for OAuth login @blevy-oai - #9943 plan prompt @aibrahim-oai - #8647 fix: handle utf-8 in windows sandbox logs @davidgilbertson - #9891 Add composer config and shared menu surface helpers @aibrahim-oai - #9880 chore(deps): bump tracing from 0.1.43 to 0.1.44 in /codex-rs @dependabot - #9882 chore(deps): bump tokio-test from 0.4.4 to 0.4.5 in /codex-rs @dependabot - #9883 chore(deps): bump axum from 0.8.4 to 0.8.8 in /codex-rs @dependabot - #9884 chore(deps): bump globset from 0.4.16 to 0.4.18 in /codex-rs @dependabot - #9901 fix: remove cli tooltip references to custom prompts @mattridley - #9823 fix: use `brew upgrade --cask codex` to avoid warnings and ambiguity @JBallin - #9951 fix: try to fix freezes 2 @jif-oai - #9955 Reject request_user_input outside Plan/Pair @charley-oai - #9933 Updated contribution guidelines @etraut-openai - #9892 Reuse ChatComposer in request_user_input overlay @aibrahim-oai - #9957 NIT larger buffer @jif-oai - #9953 feat: load interface metadata from SKILL.json @alexsong-oai - #9954 Warn users on enabling underdevelopment features @aibrahim-oai - #9961 Use test_codex more @pakrym-oai - #9960 fix: handle all web_search actions and in progress invocations @sayan-oai - #9966 plan prompt v7 @aibrahim-oai - #9968 Improve plan mode prompt @aibrahim-oai - #9969 prompt final @aibrahim-oai - #9475 Fix `resume --last` with `--json` option @etraut-openai - #9970 prompt @aibrahim-oai - #9975 plan prompt @aibrahim-oai - #9974 make cached web_search client-side default @sayan-oai - #9971 tui: wrapping user input questions @aibrahim-oai - #9977 make plan prompt less detailed @aibrahim-oai - #9980 Fixing main and make plan mode reasoning effort medium @aibrahim-oai - #9759 Fix: cap aggregated exec output consistently @Kbediako

Claude Codev2.1.20

v2.1.20

27. Januar 2026

## What's changed - Added arrow key history navigation in vim normal mode when cursor cannot move further - Added external editor shortcut (Ctrl+G) to the help menu for better discoverability - Added PR review status indicator to the prompt footer, showing the current branch's PR state (approved, changes requested, pending, or draft) as a colored dot with a clickable link - Added support for loading `CLAUDE.md` files from additional directories specified via `--add-dir` flag (requires setting `CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1`) - Added ability to delete tasks via the `TaskUpdate` tool - Fixed session compaction issues that could cause resume to load full history instead of the compact summary - Fixed agents sometimes ignoring user messages sent while actively working on a task - Fixed wide character (emoji, CJK) rendering artifacts where trailing columns were not cleared when replaced by narrower characters - Fixed JSON parsing errors when MCP tool responses contain special Unicode characters - Fixed up/down arrow keys in multi-line and wrapped text input to prioritize cursor movement over history navigation - Fixed draft prompt being lost when pressing UP arrow to navigate command history - Fixed ghost text flickering when typing slash commands mid-input - Fixed marketplace source removal not properly deleting settings - Fixed duplicate output in some commands like `/context` - Fixed task list sometimes showing outside the main conversation view - Fixed syntax highlighting for diffs occurring within multiline constructs like Python docstrings - Fixed crashes when cancelling tool use - Improved `/sandbox` command UI to show dependency status with installation instructions when dependencies are missing - Improved thinking status text with a subtle shimmer animation - Improved task list to dynamically adjust visible items based on terminal height - Improved fork conversation hint to show how to resume the original session - Changed collapsed read/search groups to show present tense ("Reading", "Searching for") while in progress, and past tense ("Read", "Searched for") when complete - Changed teammate messages to render with rich Markdown formatting (bold, code blocks, lists, etc.) instead of plain text - Changed `ToolSearch` results to appear as a brief notification instead of inline in the conversation - Changed the `/commit-push-pr` skill to automatically post PR URLs to Slack channels when configured via MCP tools - Changed the `/copy` command to be available to all users - Changed background agents to prompt for tool permissions before launching - Changed permission rules like `Bash(*)` to be accepted and treated as equivalent to `Bash` - Changed config backups to be timestamped and rotated (keeping 5 most recent) to prevent data loss

Continue.devv1.5.39

CLI Stable Release v1.5.39

26. Januar 2026

Direktes stabiles Release vom Main-Branch. Version 1.5.39 wurde direkt aus dem neuesten Main-Branch gebaut und veröffentlicht.

Continue.devv@continuedev/fetch@1.9.0

@continuedev/fetch@1.9.0

26. Januar 2026

# [1.9.0](https://github.com/continuedev/continue/compare/@continuedev/fetch@1.8.0...@continuedev/fetch@1.9.0) (2026-01-26) ### Bug Fixes * Replace console.debug with logger in exit tool ([b65f440](https://github.com/continuedev/continue/commit/b65f44027a9c36d02e930ed2be41bac274ceb312)) * Set isComplete metadata when agents finish execution ([65a1e15](https://github.com/continuedev/continue/commit/65a1e1500b112de4e24364eff890bd9b4fcac895)) * support object-type arguments in tool call parsing ([82f6f69](https://github.com/continuedev/continue/commit/82f6f6952202924952af6b9c84df42e5fd204bc3)) * upgrade @huggingface/jinja from 0.1.3 to 0.5.3 ([e5afc41](https://github.com/continuedev/continue/commit/e5afc41492f1c837156a37e3427c3cc27db93ab5)) * upgrade @tiptap/extension-paragraph from 2.26.1 to 2.27.1 ([43a759d](https://github.com/continuedev/continue/commit/43a759d72ded5c67eca9f175a83cc6579aba0c1f)) ### Features * **cli:** detect wsl and spawn appropriate shell ([a9ae420](https://github.com/continuedev/continue/commit/a9ae4205e12c7974d5b6c0feea964af9b97e1630))

Continue.devv1.5.38

CLI Stable Release v1.5.38

26. Januar 2026

Direktes stabiler Release aus dem Main-Branch. Version 1.5.38 wurde direkt aus dem neuesten Main-Branch gebaut und veröffentlicht.

Continue.devv1.5.37

CLI Stable Release v1.5.37

26. Januar 2026

Direktes stabiles Release vom Main-Branch. Version 1.5.37 gebaut und direkt aus dem neuesten Main-Branch veröffentlicht.

Windsurfv1.13.13

Fe

26. Januar 2026

Behebt die Generierung von Commit-Meldungen und Codemap-Vorschläge.

OpenCodev1.1.36

v1.1.36

25. Januar 2026

## Core - Enable thinking for Google Vertex Anthropic models (@MichaelYochpaz) - Make question validation more lax to avoid tool call failures ## TUI - Prevent crash when theme search returns no results (@ishaksebsib) ## Desktop - Visual cleanup in app - Scroll to comment on click in app - Fixed crash when no default model is set - Performance improvements in the app package - Clean up comment component usage in app **Thank you to 5 community contributors:** - @MichaelYochpaz: - fix(provider): enable thinking for google-vertex-anthropic models (#10442) - @ProdigyRahul: - fix(app): cursor on resize (#10293) - @DNGriffin: - fix(app): mobile horizontal scrolling due to session stat btn (#10487) - @ishaksebsib: - fix(tui): prevent crash when theme search returns no results (#10565) - @R44VC0RP: - fix(web): add & fix the download button (#10566)

Codex CLIvrust-v0.91.0

0.91.0

25. Januar 2026

## Chores - Reduced the maximum allowed number of sub-agents to 6 to tighten resource usage and guardrails in agent fan-out behavior (#9861) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.90.0...rust-v0.91.0 - #9861 chore: half max number of sub-agents @jif-oai

Codex CLIvrust-v0.90.0

0.90.0

25. Januar 2026

## New Features - Added a network sandbox proxy with policy enforcement to better control outbound network access. (#8442) - Introduced the first phase of connectors support via the app server and MCP integration, including new config/docs updates. (#9667) - Shipped collaboration mode as beta in the TUI, with a clearer plan → execute handoff and simplified mode selection (Coding vs Plan). (#9690, #9712, #9802, #9834) - Added ephemeral threads and improved collaboration tool provenance metadata for spawned threads. (#9765, #9769) - WebSocket connections now support proxy configuration. (#9719) - More strict limitation on multi-agents ## Bug Fixes - Fixed exec policy parsing for multiline quoted arguments. (#9565) - `--yolo` now skips the git repository check instead of failing outside a repo. (#9590) - Improved resume reliability by handling out-of-order events and prompting for the working directory when it differs. (#9512, #9731) - Backspace no longer deletes a text element when the cursor is at the element’s left edge. (#9630) - Config loading errors are clearer and more actionable across surfaces. (#9746) - Default model selection now respects filtered presets to avoid invalid defaults. (#9782) ## Documentation - Corrected a typo in the experimental collaboration prompt template. (#9716) - Added documentation for the new connectors configuration surface. (#9667) ## Chores - Refreshed the bundled model catalog/presets. (#9726) - Updated GitHub Actions for Node 24 compatibility. (#9722) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.89.0...rust-v0.90.0 - #9715 feat: fix formatting of `codex features list` @bolinfest - #9716 Fix typo in experimental_prompt.md @iudizm - #9719 feat: support proxy for ws connection @apanasenko-oai - #9712 TUI: prompt to implement plan and switch to Execute @charley-oai - #9713 use machine scope instead of user scope for dpapi. @iceweasel-oai - #9726 Update models.json @github-actions - #9667 [connectors] Support connectors part 1 - App server & MCP @mzeng-openai - #9674 feat(app-server) Expose `personality` @dylan-hurd-oai - #9733 Change the prompt for planning and reasoning effort @aibrahim-oai - #9730 Hide mode cycle hint while a task is running @charley-oai - #9720 feat: add session source as otel metadata tag @alexsong-oai - #9565 Fix execpolicy parsing for multiline quoted args @jdsalasca - #9745 chore: use some raw strings to reduce quoting @bolinfest - #9753 nit: exclude PWD for rc sourcing @jif-oai - #9690 feat: tui beta for collab @jif-oai - #9116 Persist text element ranges and attached images across history/resume @charley-oai - #9777 plan mode prompt change @aibrahim-oai - #9590 fix(exec): skip git repo check when --yolo flag is used @zerone0x - #9722 Upgrade GitHub Actions for Node 24 compatibility @salmanmkc - #9611 Print warning if we skip config loading @gt-oai - #9782 Select default model from filtered presets @aibrahim-oai - #9512 Fix resume picker when user event appears after head @jdsalasca - #9787 Remove stale TODO comment from defs.bzl @jcoens-openai - #9700 still load skills @gt-oai - #9791 Load untrusted rules @gt-oai - #9707 bundle sandbox helper binaries in main zip, for winget. @iceweasel-oai - #9792 Chore: remove mode from header @aibrahim-oai - #9793 change collaboration mode to struct @aibrahim-oai - #8442 feat: introducing a network sandbox proxy @viyatb-oai - #9802 Have a coding mode and only show coding and plan @aibrahim-oai - #9746 Another round of improvements for config error messages @etraut-openai - #9812 Remove batman reference from experimental prompt @charley-oai - #9769 feat: add thread spawn source for collab tools @jif-oai - #9765 feat: ephemeral threads @jif-oai - #9819 fix: libcc link @jif-oai - #9820 fix: musl build @jif-oai - #9316 fix(windows-sandbox): remove request files after read @MaxMiksa - #9630 Prevent backspace from removing a text element when the cursor is at the element’s left edge @charley-oai - #9840 Revert "fix: musl build" @aibrahim-oai - #9778 Raise welcome animation breakpoint to 37 rows @mzeng-openai - #9731 Ask for cwd choice when resuming session from different cwd @charley-oai - #9841 Revert "fix: libcc link" @aibrahim-oai - #9806 Use collaboration mode masks without mutating base settings @aibrahim-oai - #9834 Mark collab as beta @pakrym-oai - #9847 Revert "Revert "fix: musl build"" @jif-oai - #9855 feat: cap number of agents @jif-oai

OpenClawv2026.1.24

Clawdbot 2026.1.24

25. Januar 2026

## 2026.1.24 ### Highlights - Providers: Ollama discovery + docs; Venice guide upgrades + cross-links. (#1606) Thanks @abhaymundhara. https://docs.clawd.bot/providers/ollama https://docs.clawd.bot/providers/venice - Channels: LINE plugin (Messaging API) with rich replies + quick replies. (#1630) Thanks @plum-dawg. - TTS: Edge fallback (keyless) + `/tts` auto modes. (#1668, #1667) Thanks @steipete, @sebslight. https://docs.clawd.bot/tts - Exec approvals: approve in-chat via `/approve` across all channels (including plugins). (#1621) Thanks @czekaj. https://docs.clawd.bot/tools/exec-approvals https://docs.clawd.bot/tools/slash-commands - Telegram: DM topics as separate sessions + outbound link preview toggle. (#1597, #1700) Thanks @rohannagpal, @zerone0x. https://docs.clawd.bot/channels/telegram ### Changes - Channels: add LINE plugin (Messaging API) with rich replies, quick replies, and plugin HTTP registry. (#1630) Thanks @plum-dawg. - TTS: add Edge TTS provider fallback, defaulting to keyless Edge with MP3 retry on format failures. (#1668) Thanks @steipete. https://docs.clawd.bot/tts - TTS: add auto mode enum (off/always/inbound/tagged) with per-session `/tts` override. (#1667) Thanks @sebslight. https://docs.clawd.bot/tts - Telegram: treat DM topics as separate sessions and keep DM history limits stable with thread suffixes. (#1597) Thanks @rohannagpal. - Telegram: add `channels.telegram.linkPreview` to toggle outbound link previews. (#1700) Thanks @zerone0x. https://docs.clawd.bot/channels/telegram - Web search: add Brave freshness filter parameter for time-scoped results. (#1688) Thanks @JonUleis. https://docs.clawd.bot/tools/web - UI: refresh Control UI dashboard design system (typography, colors, spacing). (#1786) Thanks @mousberg. - Exec approvals: forward approval prompts to chat with `/approve` for all channels (including plugins). (#1621) Thanks @czekaj. https://docs.clawd.bot/tools/exec-approvals https://docs.clawd.bot/tools/slash-commands - Gateway: expose config.patch in the gateway tool with safe partial updates + restart sentinel. (#1653) Thanks @Glucksberg. - Diagnostics: add diagnostic flags for targeted debug logs (config + env override). https://docs.clawd.bot/diagnostics/flags - Docs: expand FAQ (migration, scheduling, concurrency, model recommendations, OpenAI subscription auth, Pi sizing, hackable install, docs SSL workaround). - Docs: add verbose installer troubleshooting guidance. - Docs: add macOS VM guide with local/hosted options + VPS/nodes guidance. (#1693) Thanks @f-trycua. - Docs: add Bedrock EC2 instance role setup + IAM steps. (#1625) Thanks @sergical. https://docs.clawd.bot/bedrock - Docs: update Fly.io guide notes. - Dev: add prek pre-commit hooks + dependabot config for weekly updates. (#1720) Thanks @dguido. ### Fixes - Web UI: fix config/debug layout overflow, scrolling, and code block sizing. (#1715) Thanks @saipreetham589. - Web UI: show Stop button during active runs, swap back to New session when idle. (#1664) Thanks @ndbroadbent. - Web UI: clear stale disconnect banners on reconnect; allow form saves with unsupported schema paths but block missing schema. (#1707) Thanks @Glucksberg. - Web UI: hide internal `message_id` hints in chat bubbles. - Gateway: allow Control UI token-only auth to skip device pairing even when device identity is present (`gateway.controlUi.allowInsecureAuth`). (#1679) Thanks @steipete. - Matrix: decrypt E2EE media attachments with preflight size guard. (#1744) Thanks @araa47. - BlueBubbles: route phone-number targets to DMs, avoid leaking routing IDs, and auto-create missing DMs (Private API required). (#1751) Thanks @tyler6204. https://docs.clawd.bot/channels/bluebubbles - BlueBubbles: keep part-index GUIDs in reply tags when short IDs are missing. - iMessage: normalize chat_id/chat_guid/chat_identifier prefixes case-insensitively and keep service-prefixed handles stable. (#1708) Thanks @aaronn. - Signal: repair reaction sends (group/UUID targets + CLI author flags). (#1651) Thanks @vilkasdev. - Signal: add configurable signal-cli startup timeout + external daemon mode docs. (#1677) https://docs.clawd.bot/channels/signal - Telegram: set fetch duplex="half" for uploads on Node 22 to avoid sendPhoto failures. (#1684) Thanks @commdata2338. - Telegram: use wrapped fetch for long-polling on Node to normalize AbortSignal handling. (#1639) - Telegram: honor per-account proxy for outbound API calls. (#1774) Thanks @radek-paclt. - Telegram: fall back to text when voice notes are blocked by privacy settings. (#1725) Thanks @foeken. - Voice Call: return stream TwiML for outbound conversation calls on initial Twilio webhook. (#1634) - Voice Call: serialize Twilio TTS playback and cancel on barge-in to prevent overlap. (#1713) Thanks @dguido. - Google Chat: tighten email allowlist matching, typing cleanup, media caps, and onboarding/docs/tests. (#1635) Thanks @iHildy. - Google Chat: normalize space targets without double `spaces/` prefix. - Agents: auto-compact on context overflow prompt errors before failing. (#1627) Thanks @rodrigouroz. - Agents: use the active auth profile for auto-compaction recovery. - Media understanding: skip image understanding when the primary model already supports vision. (#1747) Thanks @tyler6204. - Models: default missing custom provider fields so minimal configs are accepted. - Messaging: keep newline chunking safe for fenced markdown blocks across channels. - Messaging: treat newline chunking as paragraph-aware (blank-line splits) to keep lists and headings together. (#1726) Thanks @tyler6204. - TUI: reload history after gateway reconnect to restore session state. (#1663) - Heartbeat: normalize target identifiers for consistent routing. - Exec: keep approvals for elevated ask unless full mode. (#1616) Thanks @ivancasco. - Exec: treat Windows platform labels as Windows for node shell selection. (#1760) Thanks @ymat19. - Gateway: include inline config env vars in service install environments. (#1735) Thanks @Seredeep. - Gateway: skip Tailscale DNS probing when tailscale.mode is off. (#1671) - Gateway: reduce log noise for late invokes + remote node probes; debounce skills refresh. (#1607) Thanks @petter-b. - Gateway: clarify Control UI/WebChat auth error hints for missing tokens. (#1690) - Gateway: listen on IPv6 loopback when bound to 127.0.0.1 so localhost webhooks work. - Gateway: store lock files in the temp directory to avoid stale locks on persistent volumes. (#1676) - macOS: default direct-transport `ws://` URLs to port 18789; document `gateway.remote.transport`. (#1603) Thanks @ngutman. - Tests: cap Vitest workers on CI macOS to reduce timeouts. (#1597) Thanks @rohannagpal. - Tests: avoid fake-timer dependency in embedded runner stream mock to reduce CI flakes. (#1597) Thanks @rohannagpal. - Tests: increase embedded runner ordering test timeout to reduce CI flakes. (#1597) Thanks @rohannagpal.

OpenCodev1.1.35

v1.1.35

25. Januar 2026

## Core - Add tests - Sync changes from upstream - Fix models snapshot loading to prevent caching issues - Fix Nix configuration - Fix unicode filename handling in snapshot diff by disabling quote escaping - Deep merge providerOptions in applyCaching to preserve nested configuration (@zerone0x) - Construct apply_patch metadata before requesting permission in web interface (@brittlewis12) - Remove models.dev macro - Tell the model what model it is in the environment section of the prompt - Retry web fetch with simple user agent on 403 errors (@Arasple) - Allow GPT-5.1-Codex model in Codex auth plugin (@Danztee) - Bump gitlab-ai-provider version (@vglafirov) - Add model information as part of task metadata (@shantur) - Add xhigh reasoning effort for GitHub Copilot GPT-5 models (@Arthur742Ramos) - Normalize path separators for session diff filtering on Windows (@alexyaroshuk) - Faster workspace creation ## TUI - Fix question selection click when terminal is unfocused (@maharshi365) - Use min/maxHeight for question textarea - Permission prompt should ignore keyboard events while dialog stack length is greater than zero (@justfortheloveof) - Use selectedForeground for question prompt tab text visibility (@ariane-emory) - Always center selected item in selection dialogs (@ariane-emory) ## Desktop - Fix submit button state in app - Fix unnecessary suspense flash in app - Add active state to comment cards in prompt input - Fix type error - Terminal clone needs remount to function properly - Remove terminal connection error overlay - Remove space between ellipsis and truncated text in comment card tooltip - Update status popover empty state text color and centering - Prevent focus on share popover text field - Add link icon and use it for copy-to-clipboard buttons in UI - Adjust share popover position 64px to the left - Update share popover gutter to 6px and radius to match status dropdown - Add 8px spacing between share button and icon buttons in titlebar - Replace Popover with DropdownMenu for server options in app - Update options icon button styling with improved active state and hover effects - Change server item actions div padding from px-4 to pl-4 - Add onFilter callback to List component and hide add server row during search - Position status circle inside input wrapper and fix dialog padding - Remove hover background color from server list items - Allow add server row to grow for error messages in app - Update add server button and row styling in app - Update manage servers dialog styling and behavior - Add truncation tooltip to server items in status popover - Update server and MCP item styles in status popover - Update status popover styling and positioning - Use 4px border radius (rounded-sm) consistently in app components - Update titlebar spacing and status popover styling in app - Enable submit button when comment cards are present in app - Add back button in subagent sessions (@tomzhu1024) - Model select dropdown not closing when pressing Escape key - Improve model selector user experience - Fix inability to select text when creating a new session - Fix scrolling in unpaid model selector - Fix Adam integration issue - Add critical shadow for comment input validation and set editor popover radius to 14px - Improve comment popover with better error styling and click-outside detection - Change prompt placeholder based on comment count in app - Adjust prompt container padding to 16px on bottom and horizontal sides - Adjust prompt input positioning to 12px from bottom and right edges, remove session panel bottom padding - Add escape/click-away to close read-only comment popovers, 10px radius, and remove 'Click to view context' text - Don't use findLast in app package - Add state to pause existing audio for demo menus and add support for audio controls (@jcampuza) - Implement new server management for web and desktop (@OpeOginni) - New workspace appears expanded and at the top of the list - Missing translations in app - Better error screen when connecting to sidecar - Fix line selection issues in the app - Fixed reactive loops in the app - Better session stream rendering performance - Add Iosevka as a font choice in desktop app (@ariane-emory) - Always close hovercard when view-sessions is clicked (@DNGriffin) - Close comment input popover on Escape key or click away - Use shadow-lg-border-base on read-only comment popovers and align label spacing - Remove border from comment input popover - Position read-only comment popover below icon with 4px gutter - Style comment popovers with 14px radius, move label below comment, and adjust text colors and sizes - Move filename and line count below comment text in popovers - Add tooltip with path, 6px spacing before close icon, and reduce filename truncation to 14 characters in editor tabs - Add middle truncation for filename in comment card - Truncate filename from start to show end of path in the app - Remove gap between filename and comment in comment card - Add shadow-xs-border with hover state to comment card - Fix comment card styling with 48px height, 2px gap, and truncated filename while preserving line count visibility - Increase comment font size to 12px in the app - Add 8px gap before close icon and truncate long filenames in app - Keep close icon in top right of comment card - Add transition-all to comment card hover states - Add close-small icon and use it for comment card dismiss button - Truncate comment text and set card max-width to 200px in app - Indent comment text to align with filename in context card - Style submitted comment icons to match comment popup style - Update prompt input styling with refined borders, hover states, and spacing - Update context cards styling with 8px padding/gap and 6px border radius - Add button-primary-base variable and use primary variant for Comment button - Improve comment popup styling and add new comment icon - Center checkbox indicator in provider selection (@ProdigyRahul) - Auto-open OAuth links for Codex and Copilot (@edoedac0) - Close delete workspace dialog immediately - Don't remount directory layout in app for better performance - Reapply work in progress changes for line selection feature in app - Revert line selection work in progress - Fix view all sessions flakiness in app (@DNGriffin) **Thank you to 18 community contributors:** - @ProdigyRahul: - fix(app): enable dialog dismiss on model selector (dialog.tsx) (#10203) - fix(app): center checkbox indicator in provider selection (#10267) - fix(app): tooltip text in light mode to use inverted neutral scale (#9786) - fix(app): markdown rendering with morphdom for better dom functions (#10373) - @DNGriffin: - fix(app): View all sessions flakiness (#10149) - fix(app): Fix plan mode btn keyboard a11y issues (#10330) - fix(app): Always close hovercard when view-sessions clicked (#10326) - @shantur: - docs(ecosystem): Add CodeNomad entry to ecosystem documentation (#10222) - feat(tasks): Add model info as part of metadata (#10307) - @edoedac0: - feat(app): auto-open oauth links for codex and copilot (#10258) - @ariane-emory: - feat: always center selected item in selection dialogs (resolves #10209) (#10207) - feat(desktop): add Iosevka as a font choice (resolves #10103) (#10347) - fix(tui): Use selectedForeground for question prompt tab text visibility (resolves #10334) (#10337) - @alexyaroshuk: - fix(app): normalize path separators for session diff filtering on Windows (#10291) - @Arthur742Ramos: - fix: add xhigh reasoning effort for GitHub Copilot GPT-5 models (#10092) - @vglafirov: - fix: bump gitlab-ai-provider version (#10255) - @Danztee: - fix: allow gpt-5.1-codex model in codex auth pluginFixes (#10181) - @Arasple: - fix: retry webfetch with simple UA on 403 (#10328) - @neriousy: - feat(docs): add a desktop troubleshooting guide (#10397) - @justfortheloveof: - fix: permission prompt should ignore keyboard events while dialog stack len > 0 (#10338) - @brittlewis12: - fix(web): construct apply_patch metadata before requesting permission (#10422) - @OpeOginni: - feat(web): implement new server management for web and desktop (#8513) - @zerone0x: - fix(provider): deep merge providerOptions in applyCaching (#10323) - @jcampuza: - fix: add state to pause existing audio for demo menus, add support fo… (#10428) - @maharshi365: - fix(opentui): question selection click when terminal unfocused (#9731) - @tomzhu1024: - feat(app): back button in subagent sessions (#10439)

Windsurfv1.13.12

Enterprise-Funktionen

25. Januar 2026

Unternehmensadministratoren können organisationsweite Erlaubnis-/Verweigerungslisten für die automatische Ausführung von Befehlen festlegen.

OpenClawv2026.1.23

Clawdbot 2026.1.23

24. Januar 2026

## 2026.1.23 ### Highlights - TTS: verschiebe Telegram TTS in den Kern + aktiviere standardmäßig modellgesteuerte TTS-Tags für expressive Audio-Antworten. (#1559) Danke @Glucksberg. https://docs.clawd.bot/tts - Gateway: füge HTTP-Endpunkt `/tools/invoke` für direkte Tool-Aufrufe hinzu (Auth + Tool-Policy durchgesetzt). (#1575) Danke @vignesh07. https://docs.clawd.bot/gateway/tools-invoke-http-api - Heartbeat: kanalspezifische Sichtbarkeitssteuerungen (OK/Alerts/Indicator). (#1452) Danke @dlauer. https://docs.clawd.bot/gateway/heartbeat - Deploy: füge Fly.io Deployment-Unterstützung + Anleitung hinzu. (#1570) https://docs.clawd.bot/platforms/fly - Channels: füge Tlon/Urbit Channel-Plugin hinzu (DMs, Gruppen-Nennungen, Thread-Antworten). (#1544) Danke @wca4a. https://docs.clawd.bot/channels/tlon ### Changes - Channels: erlaube pro-Gruppe Tool-Allow/Deny-Policies über integrierte + Plugin-Channels hinweg. (#1546) Danke @adam91holt. https://docs.clawd.bot/multi-agent-sandbox-tools - Agents: füge Bedrock Auto-Discovery-Defaults + Config-Overrides hinzu. (#1553) Danke @fal3. https://docs.clawd.bot/bedrock - CLI: füge `clawdbot system` für System-Ereignisse + Heartbeat-Steuerungen hinzu; entferne eigenständiges `wake`. (Commit 71203829d) https://docs.clawd.bot/cli/system - CLI: füge Live-Auth-Probes zu `clawdbot models status` für pro-Profil-Verifizierung hinzu. (Commit 40181afde) https://docs.clawd.bot/cli/models - CLI: starte Gateway standardmäßig neu nach `clawdbot update`; füge `--no-restart` hinzu, um das zu überspringen. (Commit 2c85b1b40) - Browser: füge node-host Proxy-Auto-Routing für Remote Gateways hinzu (je Gateway/Knoten konfigurierbar). (Commit c3cb26f7c) - Plugins: füge optionalen `llm-task` JSON-nur Tool für Workflows hinzu. (#1498) Danke @vignesh07. https://docs.clawd.bot/tools/llm-task - Markdown: füge pro-Kanal Tabellenkonvertierung hinzu (Aufzählungen für Signal/WhatsApp, Code-Blöcke sonst). (#1495) Danke @odysseus0. - Agents: behalte System-Prompt Zeitzone nur und verschiebe aktuelle Zeit zu `session_status` für bessere Cache-Treffer. (Commit 66eec295b) - Agents: entferne redundante Bash-Tool-Benennung aus Tool-Registrierung/Anzeige. (#1571) Danke @Takhoffman. - Docs: füge Cron- vs. Heartbeat-Entscheidungsleitfaden hinzu (mit Lobster-Workflow-Hinweisen). (#1533) Danke @JustYannicc. https://docs.clawd.bot/automation/cron-vs-heartbeat - Docs: kläre, dass HEARTBEAT.md leere Dateien Herzschläge überspringt, fehlende Datei läuft weiterhin. (#1535) Danke @JustYannicc. https://docs.clawd.bot/gateway/heartbeat ### Fixes - Sessions: akzeptiere non-UUID SessionIds für History/Send/Status bei Beibehaltung der Agenten-Scope. (#1518) - Messaging/Sessions: spiegel outbound Sends in Ziel-Session-Schlüssel wider (Threads + dmScope), erstelle Session-Einträge beim Senden und normalisiere Groß-/Kleinschreibung des Session-Schlüssels. (#1520, Commit 4b6cdd1d3) - Sessions: lehne array-basierte Session-Stores ab, um stille Löschungen zu verhindern. (#1469) - Gateway: vergleiche Startzeit des Linux-Prozesses, um PID-Wiederverwendung-Lock-Loops zu vermeiden; halte Locks, sofern sie nicht veraltet sind. (#1572) Danke @steipete. - Gateway: akzeptiere `null` optionale Felder in Ausführungs-Autorisierungsanfragen. (#1511) Danke @pvoo. - Exec-Berechtigungen: speichere Allowlist-Eintrags-IDs, um macOS-Allowlist-Zeilen stabil zu halten. (#1521) Danke @ngutman. - Exec: berücksichtige `tools.exec`-Ask/Security-Standards für erhöhte Freigaben (vermeide unerwünschte Abfragen). (Commit 5662a9cdf) - Daemon: benutze Plattform PATH-Trennzeichen beim Aufbau minimaler Service-Pfade. (Commit a4e57d3ac) - Linux: schließe env-konfigurierte User-Bin-Verzeichnisse in systemd PATH ein und angle PATH-Audits aus. (#1512) Danke @robbyczgw-cla. - Tailscale: erneute Bereitstellung von Serve/Funnel nur bei Berechtigungsfehlern mit sudo und behalte ursprüngliche Fehlermeldungen. (#1551) Danke @sweepies. - Docker: aktualisiere Gateway-Befehl in docker-compose und Hetzner-Anleitung. (#1514) - Agents: zeige Tool-Fehler-Fallback an, wenn der letzte Assistant-Turn nur Tools aufgerufen hat (verhindert stille Stopps). (Commit 8ea8801d0) - Agents: ignoriere IDENTITY.md Template-Platzhalter beim Parsen der Identität. (#1556) - Agents: entferne verwaiste OpenAI-Responses-Begründungsblöcke bei Modellwechseln. (#1562) Danke @roshanasingh4. - Agents: füge CLI-Log-Hinweis zu „Agent fehlgeschlagen vor Antwort“ hinzu. (#1550) Danke @sweepies. - Agents: warnen und ignorieren Tool-Allowlists, die nur unbekannte oder entladene Plugin-Tools referenzieren. (#1566) - Agents: behandeln Plugin-nur Tool-Allowlists als Opt-ins; Kern-Tools bleiben aktiviert. (#1467) - Agents: berücksichtige Queue-Overrides für eingebettete Runs, um Deadlocks in Tests zu vermeiden. (Commit 084002998) - Slack: berücksichtige offene GroupPolicy für nicht aufgeführte Kanäle bei Nachrichten + Slash-Gating. (#1563) Danke @itsjaydesu. - Discord: Beschränkung der AutoThread-Nennung-Bypass auf bot-eigenen Threads; ack-Reaktionen mention-gated belassen. (#1511) Danke @pvoo. - Discord: erneute Versuche bei rate-limitiertem Allowlist-Auflösung + Befehl-Deployment, um Gateway-Abstürze zu vermeiden. (Commit f70ac0c7c) - Mentions: ignoriere MentionPattern-Matches, wenn in Gruppen-Chats eine andere explizite Erwähnung vorhanden ist (Slack/Discord/Telegram/WhatsApp). (Commit d905ca0e0) - Telegram: Markdown in Media-Untertitel rendern. (#1478) - MS Teams: entferne `.default`-Suffix aus Graph-Scope und Bot Framework-Probe-Scope. (#1507, #1574) Danke @Evizero. - Browser: halte Extensions-Relay-Tabs steuerbar, wenn die Extension eine Session-ID nach Tab-Wechsel wiederverwendet. (#1160) - Voice Wake: Wake-Words beim Blur/Submit auf iOS/Android automatisch speichern und Limits an macOS anpassen. (Commit 69f645c66) - UI: halte die Control-UI-Seitenleiste sichtbar beim Scrollen langer Seiten. (#1515) Danke @pookNast. - UI: cached die Control-UI-Markdown-Darstellung + memoisiere Chat-Text-Extraktion, um Safari-Tippel-Jank zu reduzieren. (Commit d57cb2e1a) - TUI: weiterleite unbekannte Slash-Befehle, inkludiere Gateway-Befehle in Auto-Vervollständigung und renderSlash-Antworten als System-Ausgabe. (Commit 1af227b61, Commit 8195497ce, Commit 6fba598ea) - CLI: Auth-Probe-Ausgabe verfeinern (Tabellen-Ausgabe, Inline-Fehler, reduziert Noise, Wrap-Fixes in `clawdbot models status`). (Commit da3f2b489, Commit 00ae21bed, Commit 31e59cd58, Commit f7dc27f2d, Commit 438e782f8, Commit 886752217, Commit aabe0bed3, Commit 81535d512, Commit c63144ab1) - Media: parse nur `MEDIA:`-Tags, wenn sie am Anfang der Zeile stehen, um Prosa-Nennungen nicht zu entfernen. (#1206) - Media: bewahre PNG-Alpha, wenn möglich; fallback auf JPEG

Gemini CLIv0.25.2

Release v0.25.2

23. Januar 2026

## Was sich geändert hat * fix(patch): cherry-pick 9866eb0 to release/v0.25.1-pr-17166 to patch version v0.25.1 and create version 0.25.2 by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/17419 **Vollständiges Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.25.1...v0.25.2

Claude Codev2.1.19

v2.1.19

23. Januar 2026

## What's changed - Added env var `CLAUDE_CODE_ENABLE_TASKS`, set to `false` to keep the old system temporarily - Added shorthand `$0`, `$1`, etc. for accessing individual arguments in custom commands - Fixed crashes on processors without AVX instruction support - Fixed dangling Claude Code processes when terminal is closed by catching EIO errors from `process.exit()` and using SIGKILL as fallback - Fixed `/rename` and `/tag` not updating the correct session when resuming from a different directory (e.g., git worktrees) - Fixed resuming sessions by custom title not working when run from a different directory - Fixed pasted text content being lost when using prompt stash (Ctrl+S) and restore - Fixed agent list displaying "Sonnet (default)" instead of "Inherit (default)" for agents without an explicit model setting - Fixed backgrounded hook commands not returning early, potentially causing the session to wait on a process that was intentionally backgrounded - Fixed file write preview omitting empty lines - Changed skills without additional permissions or hooks to be allowed without requiring approval - Changed indexed argument syntax from `$ARGUMENTS.0` to `$ARGUMENTS[0]` (bracket syntax) - [SDK] Added replay of `queued_command` attachment messages as `SDKUserMessageReplay` events when `replayUserMessages` is enabled - [VSCode] Enabled session forking and rewind functionality for all users

Continue.devv1.5.36

CLI Stable Release v1.5.36

23. Januar 2026

Direktes Stable-Release vom Main-Branch. Version 1.5.36 wurde direkt aus dem neuesten Main-Branch gebaut und veröffentlicht.

OpenClawv2026.1.22

clawdbot 2026.1.22

23. Januar 2026

### Änderungen - Highlight: Compaction-Schutz verwendet jetzt adaptives Chunking, progressiven Fallback und UI-Status + Retries. (#1466) Danke @dlauer. - Providers: Antigravity-Nutzungstracking zum Statusausgabe hinzufügen. (#1490) Danke @patelhiren. - Slack: Chat-Typ Antwort-Threading-Overrides via `replyToModeByChatType` hinzufügen. (#1442) Danke @stefangalescu. - BlueBubbles: `asVoice`-Support für MP3/CAF-Sprachnotizen in sendAttachment hinzufügen. (#1477, #1482) Danke @Nicell. - Onboarding: Hatch-Auswahl (TUI/Web/Later), Token-Erklärung, Hintergrund-Dashboard-Seed auf macOS und Showcase-Link hinzufügen. ### Fehlerbehebungen - BlueBubbles: Tippindikator bei Idle/Kein-Antwort stoppen. (#1439) Danke @Nicell. - Message-Tool: Pfad/Dateipfad unverändert belassen beim Senden; Buffers nur für sendAttachment hydratisieren. (#1444) Danke @hopyky. - Auto-Antwort: Modellwechsel nur melden, wenn Sitzungsstatus verfügbar ist. (#1465) Danke @robbyczgw-cla. - Control UI: Lokale Avatar-URLs mit basePath über Injection + Identity RPC auflösen. (#1457) Danke @dlauer. - Agents: Assistenten-Verlaufstext säubern, Marker für Tool-Aufrufe entfernen. (#1456) Danke @zerone0x. - Discord: Klarstellung des Onboardings-Hinweises zur Message Content Intent. (#1487) Danke @kyleok. - Gateway: Dienst vor Deinstallation stoppen und scheitern, wenn er geladen bleibt. - Agents: Konkrete API-Fehlerdetails statt generischer AI-Service-Fehler anzeigen. - Exec: Bei PTY-Spawn-Ausfall (EBADF) auf Nicht-PTY zurückfallen. (#1484) - Exec-Zulassungen: Pro-Segment erlaubte Ausschlüsse für verkettete Shell-Befehle auf Gateway- + Node-Hosts zulassen. (#1458) Danke @czekaj. - Agents: OpenAI-Sitzungen bildschirmreinigungsfähig gestalten; Reinigung von Tool-ID/Repair durch Anbieter steuern. - Doctor: CLAWDBOT_GATEWAY_TOKEN für Auth-Checks und Wiederverwendung von Sicherheits-Audit-Tokens beachten. (#1448) Danke @azade-c. - Agents: Tool-Summen übersichtlicher gestalten und optionale Parameter nur anzeigen, wenn gesetzt. - Agents: SOUL.md-Richtlinien auch berücksichtigen, wenn Datei verschachtelt oder pfadqualifiziert ist. (#1434) Danke @neooriginal. - Matrix (Plugin): m.direct für gelöste DMs persistieren und Fallback für Räume härten. (#1436, #1486) Danke @sibbl. - CLI: Bevorzugt `~`-Pfad-Angaben für Home-Verzeichnisse in Ausgaben verwenden. - Mattermost (Plugin): Pairing/Allowlist-Gating erzwingen, Targeting von @Benutzernamen beibehalten und plugin-only-Dokumentation klären. (#1428) Danke @damoahdominic. - Agents: Transcript-Sanitation im Runner zentralisieren; `<final>`-Tags und Fehler-Turns unverändert belassen. - Auth: Auth-Profile während der initialen Auswahl und Rotation im Cooldown überspringen. (#1316) Danke @odrobnik. - Agents/TUI: Benutzers-pinned Auth-Profile während Cooldown berücksichtigen und Ranking der Suchauswahl beibehalten. (#1432) Danke @tobiasbischoff. - Docs: Gog-Auth-Services-Beispiel korrigieren, um Docs-Scopes einzuschließen. (#1454) Danke @zerone0x. - Slack: WebClient-Wiederholungen reduzieren, um doppelte Sendungen zu vermeiden. (#1481) - Slack: Thread-Antworten beim Lesen von Nachrichten berücksichtigen, falls threadId angegeben (Replies-only). (#1450) Danke @rodrigouroz. - macOS: Verknüpfte Kanäle in Gateway-Zusammenfassung bevorzugen, um falschen „nicht verlinkt“-Status zu vermeiden. - macOS/tests: Gateway-Summen-Lookup nach Guard-Unwrap korrigieren; verhindern, dass Browser während Tests geöffnet werden. (ECID-1483)

OpenCodev1.1.34

v1.1.34

23. Januar 2026

## Desktop - Fixed a reactive feedback loop in the global project cache sync (@Eric-Guo) - Add line selection functionality to the app **Thank you to 1 community contributor:** - @Eric-Guo: - fix(desktop): Fixed a reactive feedback loop in the global project cache sync (#10139)

OpenCodev1.1.33

v1.1.33

23. Januar 2026

## Core - Add thinking presets for Google Vertex Anthropic provider (@zerone0x) - Added support for OpenAI-based GitLab Duo models (@vglafirov) - Show file path in apply_patch request permission screen (@iltenahmet) ## TUI - Add weight to fuzzy search to maintain title priority in TUI (@IdrisGit) ## Desktop - Session shouldn't be keyed - Flash of fallback icon for projects fixed - Fix project icon color flash on load - Fixed folder suggestions missing last part in path completion - Allow adding projects from any directory depth - Allow adding projects from root directory in the app - Non-git projects can now be renamed - Add app version display to settings (@alexyaroshuk) **Thank you to 5 community contributors:** - @iltenahmet: - fix(ui): show file path in apply_patch request permission screen (#10079) - @alexyaroshuk: - feat(app): add app version display to settings (#10095) - @IdrisGit: - fix(tui): add weight to fuzzy search to maintain title priority (#10106) - @vglafirov: - feat(gitlab): Added support for OpenAI based GitLab Duo models (#10108) - @zerone0x: - fix(provider): add thinking presets for Google Vertex Anthropic (#9953)

Continue.devv@continuedev/fetch@1.8.0

@continuedev/fetch@1.8.0

22. Januar 2026

# [1.8.0](https://github.com/continuedev/continue/compare/@continuedev/fetch@1.7.0...@continuedev/fetch@1.8.0) (2026-01-22) ### Bug Fixes * berücksichtige unterschiedliche Tokenizer ([b4bc3ae](https://github.com/continuedev/continue/commit/b4bc3aedaa2e44149a1898c609405ebb80ca91b2)) * GH_TOKEN zum continue-agents Workflow hinzufügen ([dd93919](https://github.com/continuedev/continue/commit/dd939190701be55917d4d224f012017e8ea17be9)), schließt [#9493](https://github.com/continuedev/continue/issues/9493) * fehlenden cancelStream-Aufruf und Rückgabe für nicht-retrybare Fehler hinzufügen ([c391353](https://github.com/continuedev/continue/commit/c391353709ad4367fb333680190e0ba73c868ef0)) * fehlende Hub-Mock-Verweigerung im Test hinzufügen ([53a9fa4](https://github.com/continuedev/continue/commit/53a9fa4cb34d97a9f3e2af1f26a02b10bbd0a9a1)) * auch Umgebungsvariablen im Write-Job-Zusammenfassungs-Schritt verwenden ([8969f42](https://github.com/continuedev/continue/commit/8969f422619f63becf2f778c4465cf192ee83e18)) * **anthropic:** Azure-gehostete Anthropic-Endpunkte unterstützen ([9e8bfcd](https://github.com/continuedev/continue/commit/9e8bfcd9396b9abf7643c79bb53f7263da8138c6)) * Zirkularabhängigkeiten aufbrechen, indem Tool-Namen extrahiert werden ([6a5f80b](https://github.com/continuedev/continue/commit/6a5f80b0fc739ed36448b62bd036c01476b91d49)) * 'Agent:' in Intro-Nachricht zu 'Config:' ändern und Tests aktualisieren ([c1e4668](https://github.com/continuedev/continue/commit/c1e466836acd0b2a5e81f2925c5d2e14691543d9)) * klären, dass parallel mit sich selbst nicht bearbeitet werden kann ([46b31af](https://github.com/continuedev/continue/commit/46b31af330aa669d93c777d5fa577b2ca7c72225)) * **cli:** Kompaktierung für fehlende Tool-Ergebnisse ([0792b94](https://github.com/continuedev/continue/commit/0792b949bdf629fd14dac45478b2027962b77454)) * **cli:** Modellfähigkeits-Erkennung auf Llama, Nemotron und Mistral erweitern ([#8845](https://github.com/continuedev/continue/issues/8845)) ([528a8ab](https://github.com/continuedev/continue/commit/528a8abceb69a6afb12e0a32ea5eb8acbc1e8241)), schließt [#1](https://github.com/continuedev/continue/issues/1) * **cli:** zirkuläre Abhängigkeit im uploadArtifact-Tool auflösen ([b5da44d](https://github.com/continuedev/continue/commit/b5da44ddd665d571b9669a9832e2d5b34a9f743c)) * **cli:** CLI für manuelles Update neu starten ([595986a](https://github.com/continuedev/continue/commit/595986a6fb46ccd4e023078e9e3a8aeb911629c7)) * **cli:** logger.info statt console.log für Konsistenz verwenden ([00b665f](https://github.com/continuedev/continue/commit/00b665f089e1b1c1166bf8c7d275465b6fb8f5e2)) * Befehl in CONTRIBUTING-Datei behoben ([#9715](https://github.com/continuedev/continue/issues/9715)) ([a21eb15](https://github.com/continuedev/continue/commit/a21eb151d30969c292e290523941ebb520c61151)) * kompakte Unterhaltung mit schwebenden Tool-Aufrufen ([b33b8a4](https://github.com/continuedev/continue/commit/b33b8a4dadb821cab67d517223cb9ab19268d173)) * Updates zur Kompaktion Teil 2 ([d18c6f0](https://github.com/continuedev/continue/commit/d18c6f04ad6ca518f849a78ab6cb8938e8ea1c24)) * Beschränke Root eslint auf v8, um Core-Abhängigkeit anzupassen ([ad6ab80](https://github.com/continuedev/continue/commit/ad6ab8024b888816dcf0c8eb39921a98777ec5ab)) * Kontextlängen-Fehlererkennung ([1bfb72e](https://github.com/continuedev/continue/commit/1bfb72edf3f6c0eccb66454eb9e3ad423f58bee5)) * Kontextlängen-Fixes, Kürzungen u. ä. ([613ffb8](https://github.com/continuedev/continue/commit/613ffb8ca9847fcb3ff99beec6f191b25cd85f9b)) * korrekte eingebundene Tool-Namen-Liste ([26bef77](https://github.com/continuedev/continue/commit/26bef77a8859788b870e42036e0958bdbaa9d93a)) * Korrekte Import-Reihenfolge für ESLint-Konformität ([657c5fe](https://github.com/continuedev/continue/commit/657c5fe023cf25473d4365112f0162af38922d50)) * korrekte Import-Reihenfolge im ToolPermissionService ([2003128](https://github.com/continuedev/continue/commit/2003128f084517bdf5343d6e2205edc85db8ff64)) * URI-enkodete Pfade für WSL-Arbeitsbereichspfade decodieren ([074a6c7](https://github.com/continuedev/continue/commit/074a6c74baf35b98e0092f0e6b37027bb6561eba)) * fallback nicht auf relativen Pfad, wenn Markdown-Datei nicht gegeben ([d93b432](https://github.com/continuedev/continue/commit/d93b4321a550acf7151949d1fee1c436945e2ee0)) * doppelte Tool-Nachrichten ([77a2f12](https://github.com/continuedev/continue/commit/77a2f1274a6183bf43cb8df3d84e8e58762b1442)) * sicherstellen, dass cross-target LanceDB-Binärdateien korrekt kopiert werden ([#9100](https://github.com/continuedev/continue/issues/9100)) ([291f8f5](https://github.com/continuedev/continue/commit/291f8f5dd2efec7a53e49cc7e2a1b7e6a0833466)) * Env-Stubs für CLI-Plattform-Tests ([6b3c0be](https://github.com/continuedev/continue/commit/6b3c0be2f1443e47db3e985560cb859b8e42f6a9)) * ESLint ([50adea3](https://github.com/continuedev/continue/commit/50adea309b6635b125e55a13bb99b0a91827956f)) * ESLint Import unresolved ([ccd0b74](https://github.com/continuedev/continue/commit/ccd0b74dc544687b3ec9439ee9a814efbacc5752)) * Lint-Fehler behoben und Tests für Markdown-only-Fallback aktualisiert ([4c0c532](https://github.com/continuedev/continue/commit/4c0c53257baf7edb943c76155fa6b0bde79b554f)) * Usage-Feld in createSession und startNewSession initialisiert ([5c50e91](https://github.com/continuedev/continue/commit/5c50e911d203fc0a7ae6407f84bdfafd8c1f19f4)) * Injektions-Blocks-Tests für CLI-Geheimresolution ([219f566](https://github.com/continuedev/continue/commit/219f5661c9999daa5e185740dd2fa1d24f0187b

Claude Codev2.1.17

v2.1.17

22. Januar 2026

## What's changed - Fixed crashes on processors without AVX instruction support

Codex CLIvrust-v0.89.0

0.89.0

22. Januar 2026

## New Features - Added a `/permissions` command with a shorter approval set while keeping `/approvals` for compatibility. (#9561) - Added a `/skill` UI to enable or disable individual skills. (#9627) - Improved slash-command selection by prioritizing exact and prefix matches over fuzzy matches. (#9629) - App server now supports `thread/read` and can filter archived threads in `thread/list`. (#9569, #9571) - App server clients now support layered `config.toml` resolution and `config/read` can compute effective config from a given cwd. (#9510) - Release artifacts now include a stable URL for the published config schema. (#9572) ## Bug Fixes - Prevented tilde expansion from escaping HOME on paths like `~//...`. (#9621) - TUI turn timing now resets between assistant messages so elapsed time reflects the latest response. (#9599) ## Documentation - Updated MCP subcommand docs to match current CLI behavior. (#9622) - Refreshed the `skills/list` protocol README example to match the latest response shape. (#9623) ## Chores - Removed the TUI2 experiment and its related config/docs, keeping Codex on the terminal-native UI. (#9640) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.88.0...rust-v0.89.0 - #9576 [bazel] Upgrade to bazel9 @zbarsky-openai - #9606 nit: ui on interruption @jif-oai - #9609 chore: defensive shell snapshot @jif-oai - #9621 fix: Fix tilde expansion to avoid absolute-path escape @tiffanycitra - #9573 define/emit some metrics for windows sandbox setup @iceweasel-oai - #9622 docs: fix outdated MCP subcommands documentation @htiennv - #9623 Update skills/list protocol readme @gverma-openai - #9616 [bazel] Upgrade llvm toolchain and enable remote repo cache @zbarsky-openai - #9624 forgot to add some windows sandbox nux events. @iceweasel-oai - #9633 Add websockets logging @pakrym-oai - #9592 Chore: update plan mode output in prompt @shijie-oai - #9583 Add collaboration_mode to TurnContextItem @charley-oai - #9510 Add layered config.toml support to app server @etraut-openai - #9629 feat: better sorting of shell commands @jif-oai - #9599 fix(tui) turn timing incremental @dylan-hurd-oai - #9572 feat: publish config schema on release @sayan-oai - #9549 Reduce burst testing flake @charley-oai - #9640 feat(tui): retire the tui2 experiment @joshka-oai - #9597 feat(core) ModelInfo.model_instructions_template @dylan-hurd-oai - #9627 Add UI for skill enable/disable. @xl-openai - #9650 chore: tweak AGENTS.md @dylan-hurd-oai - #9656 Add tui.experimental_mode setting @pakrym-oai - #9561 feat(tui) /permissions flow @dylan-hurd-oai - #9653 Fix: Lower log level for closed-channel send @Kbediako - #9659 Chore: add cmd related info to exec approval request @shijie-oai - #9693 Revert "feat: support proxy for ws connection" @pakrym-oai - #9698 Support end_turn flag @pakrym-oai - #9645 Modes label below textarea @charley-oai - #9644 feat(core) update Personality on turn @dylan-hurd-oai - #9569 feat(app-server): thread/read API @owenlin0 - #9571 feat(app-server): support archived threads in thread/list @owenlin0

Claude Codev2.1.16

v2.1.16

22. Januar 2026

## What's changed - Added new task management system, including new capabilities like dependency tracking - [VSCode] Added native plugin management support - [VSCode] Added ability for OAuth users to browse and resume remote Claude sessions from the Sessions dialog - Fixed out-of-memory crashes when resuming sessions with heavy subagent usage - Fixed an issue where the "context remaining" warning was not hidden after running `/compact` - Fixed session titles on the resume screen not respecting the user's language setting - [IDE] Fixed a race condition on Windows where the Claude Code sidebar view container would not appear on start

OpenCodev1.1.32

v1.1.32

22. Januar 2026

<highlight source="Desktop"> <h2>Multi-Language Support</h2> <p short="OpenCode Desktop now available in 14 languages."> OpenCode Desktop now available in 14 languages. </p> https://github.com/user-attachments/assets/e5d612cf-20fc-4d55-824d-7f66b9733da2 </highlight> ## Core - Revert handling of special characters in paths and git snapshot reading logic - Revert snapshot regression fix - Ensure cat command triggers external directory permission check in Bash tool - Print MCP stderr to OpenCode log file (@karta0807913) - Write refresh tokens to OpenAI auth in Codex (@CasualDeveloper) - Add workspace startup script to projects - Ensure images are properly returned as tool results - Rename `setSessionModel` to `unstable_setSessionModel` in ACP (@maxious) - Correct spelling of 'supercedes' to 'supersedes' (@luojiyin1987) - Add prompt cache key support for Venice provider (@dpuyosa) - Include noop tool in activeTools for LiteLLM proxy compatibility (@danielfrg) ## TUI - Allow remote directory specification in attach command (@gricha) ## Desktop - Add Brazilian Portuguese locale support (@ygorsimoes) - Make 'Learn More' link functional in theme settings (@LAG-4) - Render images in session review - Fix Windows path handling in app tests - Fix new session button in app - Add translations support to app - Delete sessions - Close projects from hover card - Don't auto-navigate to workspace after reset - Expand font stacks to include macOS Nerd Font default names (@bewareoftheleopard) - Clear session hover state on navigation (@kyashrathore) - Correct NO_PROXY environment variable syntax in desktop application (@Brendonovich) - Improve desktop connection reliability (@Brendonovich) - Add terminal smoke test to app test suite - Add file viewer smoke test - Add model picker smoke test - Add smoke test for @ attachment functionality - Add slash command smoke tests to app - Settings smoke tests - Fix type error in codebase - Fix TypeScript satisfies operator usage - Fix hover card scrolling behavior - Auto-scroll to bottom when new messages arrive - Add scrollbar styling to session page (@alexyaroshuk) - Add support for Norwegian translations (@Finityno) - Snap to bottom when user submits a prompt - Add accessibility translations to the app - Change project path tooltip position to bottom in desktop app (@ProdigyRahul) - Session screen accessibility improvements (@ndarilek) - Move markdown rendering to Rust in desktop app (@nexxeln) - Add 8px left margin to sidebar toggle on desktop - Match terminal toggle button size with sidebar and review toggles - Update jump to latest button with circular design and animation - Add Arabic language support (@NourEldin-Osama) - Always show Toggle-Review button in app (@DNGriffin) - Open terminal pane when creating new terminal (@kearns-cu) **Thank you to 24 community contributors:** - @R44VC0RP: - fix: scope PR recap to only PRs from today (#9905) - @danielfrg: - fix: include _noop tool in activeTools for LiteLLM proxy compatibility (#9912) - @dpuyosa: - feat: Add promptCacheKey for Venice provider (#9915) - @kearns-cu: - fix(app): open terminal pane when creating new terminal (#9926) - @luojiyin1987: - fix: correct spelling 'supercedes' to 'supersedes' (#9935) - @maxious: - fix(acp): rename setSessionModel to unstable_setSessionModel (#9940) - @IdrisGit: - docs: add API server section in CONTRIBUTING.md (#9888) - @DNGriffin: - feat(app): always show Toggle-Review button (#9944) - @NourEldin-Osama: - feat(i18n): add Arabic language support (#9947) - @nexxeln: - refactor(desktop): move markdown rendering to rust (#10000) - @ndarilek: - fix(app): session screen accessibility improvements (#9907) - @ProdigyRahul: - fix(desktop): change project path tooltip position to bottom (#9497) - @Finityno: - feat: Add support for Norwegian translations (#10018) - @alexyaroshuk: - feat(app): add scrollbar styling to session page (#10020) - fix(ui): prevent copy buttons from stealing focus from prompt input (#10084) - @Brendonovich: - fix(desktop): attempt to improve connection reliability - fix(desktop): correct NO_PROXY syntax - @kyashrathore: - fix(app): clear session hover state on navigation (#10031) - @ryanmiville: - fix(app): support ctrl-n/p in lists (#10036) - @bewareoftheleopard: - fix(desktop): Expand font stacks to include macOS Nerd Font default names (#10045) - @CasualDeveloper: - fix(codex): write refresh tokens to openai auth (#10010) (#10011) - @karta0807913: - feat(mcp log): print mcp stderr to opencode log file (#9982) - @LAG-4: - fix: make 'Learn More' link functional in theme settings (#10078) - @shubh73: - feat(ui): add aura theme (#10056) - @ygorsimoes: - feat(i18n): add br locale support (#10086) - @gricha: - fix(attach): allow remote --dir (#8969)

Gemini CLIv0.25.1

Release v0.25.1

22. Januar 2026

## Geändert * fix(patch): cherry-pick 87a0db2 to release/v0.25.0-pr-17308 [CONFLICTS] by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/17316 **Vollständiger Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.25.0...v0.25.1

OpenClawv2026.1.21

clawdbot 2026.1.21

22. Januar 2026

### Highlights - Lobster optionales Plugin-Tool für typisierte Arbeitsabläufe + Approvals-Gates. [docs](https://docs.clawd.bot/tools/lobster) - Custom Assistant Identity + Avatare in der Control UI. [docs](https://docs.clawd.bot/cli/agents) [docs](https://docs.clawd.bot/web/control-ui) - Cache-Optimierungen: cache-ttl Pruning + Defaults reduzieren Token-Verbrauch bei Cold Requests. [docs](https://docs.clawd.bot/concepts/session-pruning) - Exec Approvals + Elevated Ask/Full-Modi. [docs](https://docs.clawd.bot/tools/elevated) [docs](https://docs.clawd.bot/tools/exec-approvals) - Signal Typing/Read Receipts + MSTeams-Anhänge. [docs](https://docs.clawd.bot/channels/signal) [docs](https://docs.clawd.bot/channels/msteams) - `/models` UX-Refresh + `clawdbot update wizard`. [docs](https://docs.clawd.bot/cli/models) [docs](https://docs.clawd.bot/cli/update) ### Changes - Highlight: Lobster optionales Plugin-Tool für typisierte Workflows + Genehmigungsgates. [docs](https://docs.clawd.bot/tools/lobster) (#1152) Thanks @vignesh07. - Agents/UI: Identity-Avatar-Konfig-Support hinzufügen und Control-UI-Avatar-Rendering. (#1329, #1424) Thanks @dlauer. [docs](https://docs.clawd.bot/gateway/configuration) [docs](https://docs.clawd.bot/cli/agents) - Control UI: benutzerdefinierte Assistant-Identity-Unterstützung und per-Session Identity-Anzeige hinzufügen. (#1420) Thanks @robbyczgw-cla. [docs](https://docs.clawd.bot/web/control-ui) - CLI: `clawdbot update wizard` mit interaktiver Channel-Auswahl + Neustart-Prompts, plus Preflight-Checks vor dem Rebasing. [docs](https://docs.clawd.bot/cli/update) - Models/Commands: `/models` hinzufügen, UX von `/model` verbessern und Paging von `clawdbot models` erweitern. (#1398) Thanks @vignesh07. [docs](https://docs.clawd.bot/cli/models) - CLI: Gateways-Service-Befehle unter `clawdbot gateway` verschieben, Node-Service-Befehle unter `clawdbot node` flattenen, und `gateway probe` für Reichweite hinzufügen. [docs](https://docs.clawd.bot/cli/gateway) [docs](https://docs.clawd.bot/cli/node) - Exec: Elevated Ask/Full-Modi hinzufügen, Allowlist-Gating straffen, und Genehmigungs-Tabellen beim Schreiben rendern. [docs](https://docs.clawd.bot/tools/elevated) [docs](https://docs.clawd.bot/tools/exec-approvals) - Exec Approvals: standardmäßig auf localhost setzen, Gateway/Node-Targeting + Ziel-Details hinzufügen, Wildcard-Agent-Allowlists unterstützen und Allowlist-Parsing/Safe-Bins straffen. [docs](https://docs.clawd.bot/cli/approvals) [docs](https://docs.clawd.bot/tools/exec-approvals) - Heartbeat: explizite Session-Keys und aktive Stunden zulassen. (#1256) Thanks @zknicker. [docs](https://docs.clawd.bot/gateway/heartbeat) - Sessions: pro-Kanal inaktive Dauer via `sessions.channelIdleMinutes` hinzufügen. (#1353) Thanks @cash-echo-bot. - Nodes: Execution im Stil von Exec ausführen, PATH-Umgebung im Status/Beschreibung sichtbar machen, und PATH für Node-Host-Execution bootstrappern. [docs](https://docs.clawd.bot/cli/node) - Cache: `cache.ttlPrune`-Modus hinzufügen und auth-aware Defaults für Cache-TTL-Verhalten. - Queue: pro-Kanal Debounce-Overrides für Auto-Reply hinzufügen. [docs](https://docs.clawd.bot/concepts/queue) - Discord: Wildcard-Channel-Konfig-Unterstützung hinzufügen. (#1334) Thanks @pvoo. [docs](https://docs.clawd.bot/channels/discord) - Signal: Tipp-/Typing-Indikatoren und DM-Read-Receipts über signal-cli hinzufügen. [docs](https://docs.clawd.bot/channels/signal) - MSTeams: Dateiuploads, Adaptive Cards und Verbesserungen bei Attachment-Handling. (#1410) Thanks @Evizero. [docs](https://docs.clawd.bot/channels/msteams) - Onboarding: Run-Setup-token Auth-Option entfernen (stattdessen Setup-token einfügen oder CLI-Creds verwenden). - macOS: Settings aktualisieren (Standortzugriff in Permissions, Verbindungsmodus im Menü, CLI-Installations-UI entfernen). - Diagnostics: Cache-Trace-Konfiguration zum Debugging hinzufügen. (#1370) Thanks @parubets. - Docs: Lobster-Guides + Organisations-URL-Updates, /model-Allowlist-Fehlerbehebung, Gmail-Nachrichten-Suchbeispiele, gateway.mode-Fehlerbehebung, Prompt-Injection-Anleitung, npm-Prefix/Node-CLI-Hinweise, Control-UI Dev-GatewayUrl-Hinweis, Tool_use FAQ, Showcase-Video, und sharp/node-gyp-Workaround. (#1427, #1220, #1405) Thanks @vignesh07, @mbelinky. ### Breaking - **BREAKING:** Control UI lehnt insecure HTTP ohne Geräte-Identity standardmäßig ab. Verwenden Sie HTTPS (Tailscale Serve) oder setzen Sie `gateway.controlUi.allowInsecureAuth: true`, um tokenbasierte Auth zu ermöglichen. [docs](https://docs.clawd.bot/web/control-ui#insecure-http) - **BREAKING:** Umschlag- und Systemereignis-Timestamps standardmäßig auf host-lokale Zeit (früher UTC), damit Agents nicht ständig konvertieren müssen. ### Fixes - Streaming/Typing/Media: Antwort-Tags über gestreamte Chunks beibehalten, Typing-Indikatoren zu Beginn des Laufes starten, und MEDIA-Pfade mit Leerzeichen/tilde akzeptieren, während das Nachrichten-Tool-Hinweis für Bildantworten bevorzugt wird. - Agents/Providers: unsigned Thinking-Blocks für Claude-Modelle (Google Antigravity) entfernen und alphanumerische Tool-Call-IDs bei strengen Providers (Mistral/OpenRouter) erzwingen. (#1372) Thanks @zerone0x. - Exec Approvals: Main als Standard-Agent behandeln, Node/Gateway-Allowlist-Prechecks angleichen, aufgelöste Pfade validieren, Rennen bei Allowlist-Auflösung vermeiden und optionale Parameter null vermeiden. (#1417, #1414, #1425) Thanks @czekaj. - Exec/Windows: Windows-Exec-Pfade mit Erweiterungen auflösen und sichere-Bin-Exe-Namen behandeln. - Nodes/macOS: Hinweis bei Allowlist-Miss für Node-Exec-Approvals, Allowlist-Entscheidungen persistieren, und Node-Invoke-Fehler flattenen. (#1394) Thanks @ngutman. - Gateway: Verhindern, dass mehrere Gateways dieselbe Config/State teilen (Singleton-Lock), Auto-Bind-Schleife beibehalten mit expliziter Tailnet-Bindung, und SSH-Auth-Handling verbessern. (#1380) - Control UI: Chat-Stopp-Button entfernen, Composer am unteren Rand ausrichten, Session-Previews stabilisieren, Debug-Panel bei route-gesteuerten Tab-Wechseln aktualisieren. (#1373) Thanks @yazinsai. - UI/config: SECTION_META für Config-Form-Module exportieren. (#1418) Thanks @MaudeBot. - macOS: Chat während Streaming-Antworten anheften, Textual-Ressourcen einschließen, Wildcard-Exec-Approvals respektieren, SSH-Agent-Auth erlauben, Standard-Distributions-Builds auf Universal-Binaries setzen. (#1279, #1362, #1384, #1396) Thanks @ameno-, @JustYannicc. - BlueBubbles: Kurze Message-IDs sicher auflösen, volle IDs in Templates sichtbar machen, Short-ID-Fetch-Wrappers absichern. (#1369, #1387) Thanks @tyler6204. - Models/Configure: Session-Model-Overrides in Threads/Topics vererben, OpenCode Zen-Modelle den korrekten APIs zuordnen, Anthropic OAuth-Allowlist-Handhabung einschränken, Allowlist-Fallbacks seed, vollständigen Katalog listen, wenn keine Allowlist gesetzt ist, und `/model`-Listenausgabe begrenzen. (#1376, #1416) - Memory: CLI-Hänger verhindern durch Verzögerung von Vektor-Probes, sqlite-vec/Embedding-Timeouts hinzufügen, und Session-Memory-Indexierung asynchron gestalten. - Cron: Erinnerungs-Kontextverlauf auf 10 Nachrichten begrenzen und `contextMessages` beachten. (#1103) Thanks @mkbehr. - Cache: die 1h Cache-T

OpenCodev1.1.31

v1.1.31

22. Januar 2026

## Core - Mark subagent sessions as agent-initiated to exclude them from quota limits - Add chat.headers hook and update codex and copilot plugins to use it (thanks to @ananas-viber for catching) ## TUI - Add Gruvbox theme to Web and App (@swolg) ## Desktop - Add Traditional Chinese language support and rename Chinese to Chinese (Simplified) (@alexyaroshuk) - Add Russian language support (@alexyaroshuk) - Polish translations in app (@neriousy) **Thank you to 4 community contributors:** - @neriousy: - feat(app): polish translations (#9884) - @alexyaroshuk: - feat(i18n): add Russian language support (#9882) - feat(i18n): add Traditional Chinese language support & rename 'Chinese' to 'Chinese (Simplified)' (#9887) - @swolg: - feat(ui): add gruvbox theme (Web/App) (#9855) - @R44VC0RP: - feat: add daily Discord recaps for issues and PRs (#9904)

OpenCodev1.1.30

v1.1.30

21. Januar 2026

## Core - Add session list and session fork support to ACP (@remorses) - Fix metadata matching to use model.id instead of model.api.id to support legacy model identifiers - Adjust textVerbosity and reasoningEffort defaults to better align with Codex CLI behavior - Implement home directory expansion for permission patterns using `~` and `$HOME` prefixes (@Danztee) - Add back todo list tools for OpenAI models - Persist loaded model and mode when loading ACP sessions (@noamzbr) - Update GitLab packages for better self-hosted instance support (@vglafirov) - Make the models.dev domain configurable for offline environments (@bartbroere) ## Desktop - Rename favicon on web again - Improve provider OAuth error handling in app - Fix renaming non-git projects affecting other projects - Tab should select suggestion in the app - Fix type error - Fix terminal issues and regressions - Remove terminal button border to align with close button (@allanhvam) - Fixed navigation issues when working with large sessions in desktop app (@dbpolito) **Thank you to 8 community contributors:** - @bartbroere: - feat: Make the models.dev domain configurable for offline environments (#9258) - @vglafirov: - chore(deps): update GitLab packages for better self-hosted instance support (#9856) - @noamzbr: - fix: Persist loaded model and mode on ACP session load (#9829) - @Danztee: - feat: implement home directory expansion for permission patterns using `~` and `$HOME` prefixes. (#9813) - @dbpolito: - fix(desktop): Navigation with Big Sessions (#9529) - @allanhvam: - fix(app): remove terminal button border to align with close button (#9874) - @remorses: - feat(acp): add session/list and session/fork support (#7976) - @gigamonster256: - fix: update pre-push hook to allow caret version differences (#9876)

Claude Codev2.1.15

v2.1.15

21. Januar 2026

## What's changed - Added deprecation notification for npm installations - run `claude install` or see https://docs.anthropic.com/en/docs/claude-code/getting-started for more options - Improved UI rendering performance with React Compiler - Fixed the "Context left until auto-compact" warning not disappearing after running `/compact` - Fixed MCP stdio server timeout not killing child process, which could cause UI freezes

Codex CLIvrust-v0.88.0

0.88.0

21. Januar 2026

## New Features - Added device-code auth as a standalone fallback in headless environments. (#9333) ## Bug Fixes - Load configs from trusted folders only and fix symlinked `config.toml` resolution. (#9533, #9445) - Fixed Azure endpoint invalid input errors. (#9387) - Resolved a memory leak in core runtime. (#9543) - Prevented interrupted turns from repeating. (#9043) - Fixed WSL TUI image paste regression. (#9473) ## Documentation - Updated MCP documentation link to the current destination. (#9490) - Corrected a “Multi-agents” naming typo in docs. (#9542) - Added developer instructions for collaboration modes. (#9424) ## Chores - Upgraded to Rust 1.92 and refreshed core Rust dependencies. (#8860, #9465, #9466, #9467, #9468, #9469) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.87.0...rust-v0.88.0 - #9373 fix: flaky tests @jif-oai - #9333 [device-auth] Add device code auth as a standalone option when headless environment is detected. @mzeng-openai - #9352 Made `codex exec resume --last` consistent with `codex resume --last` @etraut-openai - #9324 add codex cloud list @nornagon-openai - #9332 Turn-state sticky routing per turn @aibrahim-oai - #9364 feat: tool call duration metric @jif-oai - #8860 chore: upgrade to Rust 1.92.0 @viyatb-oai - #9385 feat: /fork the current session instead of opening session picker @apanasenko-oai - #9247 feat(app-server, core): return threads by created_at or updated_at @owenlin0 - #9330 feat: show forked from session id in /status @apanasenko-oai - #9340 Introduce collaboration modes @aibrahim-oai - #9328 Support enable/disable skill via config/api. @xl-openai - #9408 Add collaboration_mode override to turns @aibrahim-oai - #9400 fix(codex-api): treat invalid_prompt as non-retryable @fouad-openai - #9401 Defer backtrack trim until rollback confirms @aibrahim-oai - #9414 fix unified_exec::tests::unified_exec_timeouts to use a more unique match value @ahornby - #9421 Expose collaboration presets @aibrahim-oai - #9422 chore(core) Create instructions module @dylan-hurd-oai - #9423 chore(instructions) Remove unread SessionMeta.instructions field @dylan-hurd-oai - #9424 Add collaboration developer instructions @aibrahim-oai - #9425 Preserve slash command order in search @aibrahim-oai - #9059 tui: allow forward navigation in backtrack preview @slkzgm - #9443 Add collaboration modes test prompts @aibrahim-oai - #9457 fix(tui2): running /mcp was not printing any output until another event triggered a flush @bolinfest - #9445 Fixed symlink support for config.toml @etraut-openai - #9466 chore(deps): bump log from 0.4.28 to 0.4.29 in /codex-rs @dependabot - #9467 chore(deps): bump tokio from 1.48.0 to 1.49.0 in /codex-rs @dependabot - #9468 chore(deps): bump arc-swap from 1.7.1 to 1.8.0 in /codex-rs @dependabot - #9469 chore(deps): bump ctor from 0.5.0 to 0.6.3 in /codex-rs @dependabot - #9465 chore(deps): bump chrono from 0.4.42 to 0.4.43 in /codex-rs @dependabot - #9473 Fixed TUI regression related to image paste in WSL @etraut-openai - #9382 feat: timer total turn metrics @jif-oai - #9478 feat: close all threads in `/new` @jif-oai - #9477 feat: detach non-tty childs @jif-oai - #9479 prompt 3 @jif-oai - #9387 Fix invalid input error on Azure endpoint @etraut-openai - #9463 Remove unused protocol collaboration mode prompts @aibrahim-oai - #9487 chore: warning metric @jif-oai - #9490 Fixed stale link to MCP documentation @etraut-openai - #9461 TUI: collaboration mode UX + always submit `UserTurn` when enabled @aibrahim-oai - #9472 Feat: request user input tool @shijie-oai - #9402 Act on reasoning-included per turn @aibrahim-oai - #9496 chore: fix beta VS experimental @jif-oai - #9495 Feat: plan mode prompt update @shijie-oai - #9451 tui: avoid Esc interrupt when skill popup active @prateek-oai - #9497 Migrate tui to use UserTurn @aibrahim-oai - #9427 fix(core) Preserve base_instructions in SessionMeta @dylan-hurd-oai - #9393 Persist text elements through TUI input and history @charley-oai - #9407 fix(tui) fix user message light mode background @dylan-hurd-oai - #9525 chore: collab in experimental @jif-oai - #9374 nit: do not render terminal interactions if no task running @jif-oai - #9529 feat: record timer with additional tags @jif-oai - #9528 feat: metrics on remote models @jif-oai - #9527 feat: metrics on shell snapshot @jif-oai - #9533 Only load config from trusted folders @gt-oai - #9409 feat: support proxy for ws connection @apanasenko-oai - #9507 Tui: use collaboration mode instead of model and effort @aibrahim-oai - #9193 fix: `writable_roots` doesn't recognize home directory symbol in non-windows OS @tiffanycitra - #9542 Fix typo in feature name from 'Mult-agents' to 'Multi-agents' @simonw - #9459 feat(personality) introduce model_personality config @dylan-hurd-oai - #9543 fix: memory leak issue @jif-oai - #9509 Fixed config merging issue with profiles @etraut-openai - #9043 fix: prevent repeating interrupted turns @swordfish444 - #9553 fix(core): don't update the file's mtime on resume @owenlin0 - #9552 lookup system SIDs instead of hardcoding English strings. @iceweasel-oai - #9314 fix(windows-sandbox): deny .git file entries under writable roots @MaxMiksa - #9319 fix(windows-sandbox): parse PATH list entries for audit roots @MaxMiksa - #9547 merge remote models @aibrahim-oai - #9545 Add total (non-partial) TextElement placeholder accessors @charley-oai - #9532 fix(cli): add execute permission to bin/codex.js @zerone0x - #9162 Improve UI spacing for queued messages @charley-oai - #9554 Enable remote models @aibrahim-oai - #9558 queue only when task is working @aibrahim-oai - #8590 fix(core): require approval for force delete on Windows @hdcodedev - #9293 [codex-tui] exit when terminal is dumb @jmickey-oai - #9562 feat(tui2): add /experimental menu @joshka-oai - #9563 fix: bminor/bash is no longer on GitHub so use bolinfest/bash instead @bolinfest - #9568 Show session header before configuration @aibrahim-oai - #9555 feat: rename experimental_instructions_file to model_instructions_file @bolinfest - #9518 Prompt Expansion: Preserve Text Elements @charley-oai - #9560 Reject ask user question tool in Execute and Custom @charley-oai - #9575 feat: add skill injected counter metric @alexsong-oai - #9578 Feature to auto-enable websockets transport @pakrym-oai - #9587 fix CI by running pnpm @aibrahim-oai - #9586 don't ask for approval for `just fix` @aibrahim-oai - #9585 Add request-user-input overlay @aibrahim-oai - #9596 fix going up and down on questions after writing notes @aibrahim-oai - #9483 feat: max threads config @jif-oai - #9598 feat: display raw command on user shell @jif-oai - #9594 Added "codex." prefix to "conversation.turn.count" metric name @etraut-openai - #9600 feat: async shell snapshot @jif-oai - #9602 fix: nit tui on terminal interactions @jif-oai - #9551 nit: better collab tui @jif-oai

OpenCodev1.1.29

v1.1.29

21. Januar 2026

## Core - Adjust skill tool description to clarify available skills - Add caching support for Claude with Bedrock custom inference profiles (@stevoland) - Add Laravel Pint as a PHP formatter (@nunomaduro) - Change session title prompt to respond in user's language (@spoons-and-mirrors) - Terminal improvements including focus handling, rename functionality, error state management, and CSP compliance (@unitythemaker) - Fix snapshot regression in core module - Handle special characters in paths and git snapshot reading logic (@shirukai) - Update session summary after revert to show file changes - Add OPENCODE_DISABLE_PROJECT_CONFIG environment variable to disable project configuration (@kdcokenny) - Add OPENCODE_DISABLE_FILETIME_CHECK flag to skip file modification time checks (@luojiyin1987) - Ensure apply patch tool emits edited events - Ensure synthetic user message following subtasks is only added when user manually invoked subtask - Don't update session timestamp for metadata-only changes (@ariane-emory) - Fix issue when switching models between providers where past reasoning/metadata would cause 400 errors ## TUI - Support responsive layout for narrow terminal screens (@vmotta8) ## Desktop - New sound effects, downmixed to mono - Fix keybind search dialog height in app - Add new sounds to app - Add missing internationalization keys in app package - Add search functionality to settings shortcuts (@ProdigyRahul) - Add new workspace button supporting all available languages (@ProdigyRahul) - Add aria-labels to titlebar and sidebar buttons for improved accessibility (@ndarilek) - Fix e2e tests in app package - Fix type error - Breaking out of auto-scroll in the app - Add model tooltip metadata in chooser per Figma request (@kearns-cu) - Fix workspaces padding inconsistency (@DNGriffin) - Properly integrate window controls on Windows desktop (@Brendonovich) - Add Japanese language support to the app - Add workaround for nushell in desktop application (@Brendonovich) - Add Danish translations to the application - Add Japanese translations to the app - Add French translations to the app - Add Spanish translations to the app - Add German translations to the application - Add Korean language translations to the application - Don't show notification on session if active - Add manage models icon to selector (@kearns-cu) - Allow users to select directory text on new session (@DNGriffin) - Fix desktop app not connecting to sidecar when proxy is configured - Standardize desktop layout icons (@jacobabahn) - Fix numbered list rendering in web markdown (@R44VC0RP) - Remove portal spacer and fix terminal toggle padding in UI (@alexyaroshuk) - Add internationalization support to the app - Add internationalization support for prompt input component - Initial internationalization stubbing for the app - Don't interfere with scroll when using message navigation - Add keyboard shortcut (Mod+,) to open settings dialog - Improve auto-scroll user experience in app **Thank you to 20 community contributors:** - @R44VC0RP: - docs: add Anthropic subscription warning and update feature list to highlight GitHub Copilot (#9721) - fix(app): fix numbered list rendering in web markdown (#9723) - @alexyaroshuk: - fix(ui): remove portal spacer and fix terminal toggle padding (#9728) - @jacobabahn: - feat(desktop): standardize desktop layout icons (#9685) - @yxshee: - fix: correct dot prefix display in directory names for RTL text rendering issue #9579 (#9591) - @ariane-emory: - fix: don't update session timestamp for metadata-only changes (resolves #9494) (#9495) - @vmotta8: - fix(tui): responsive layout for narrow screens (#9703) - @luojiyin1987: - feat: add OPENCODE_DISABLE_FILETIME_CHECK flag (#6581) - @kdcokenny: - feat: add OPENCODE_DISABLE_PROJECT_CONFIG env var (#8093) - @RiskyMH: - chore(opencode): Use Bun.semver instead of node-semver (#9773) - @shirukai: - fix: handle special characters in paths and git snapshot reading logic(#9804) (#9807) - @DNGriffin: - feat(app): allow users to select directory text on new session (#9760) - fix(app): workspaces padding wonkiness (#9772) - @zerone0x: - fix(ui): preserve filename casing in edit/write tool titles (#9752) - @kearns-cu: - feat(app): add manage models icon to selector (per Figma request) (#9722) - feat(app): model tooltip metadata in chooser (per Figma request) (#9707) - @Brendonovich: - fix(desktop): add workaround for nushell - feat(desktop): properly integrate window controls on windows (#9835) - @unitythemaker: - fix(app): terminal improvements - focus, rename, error state, CSP (#9700) - @ndarilek: - fix(app): add aria-labels to titlebar and sidebar buttons (#9843) - @ProdigyRahul: - fix(app): new workspace button with all languages (#9848) - feat(app): search on settings shortcuts (#9850) - @spoons-and-mirrors: - tweak(session title): change prompt to have the response with user language (#9847) - @nunomaduro: - feat(formatters): add laravel pint as a `.php` formatter (#7312) - @stevoland: - fix: Claude w/bedrock custom inference profile - caching support (#9838)

OpenClawv2026.1.20

clawdbot 2026.1.20

21. Januar 2026

### Änderungen - Control UI: Kopieren als Markdown mit Fehler-Rückmeldung hinzufügen. (#1345) https://docs.clawd.bot/web/control-ui - Control UI: die Legacy-Listenansicht entfernen. (#1345) https://docs.clawd.bot/web/control-ui - TUI: Syntaxhervorhebung für Code-Blöcke hinzufügen. (#1200) https://docs.clawd.bot/tui - TUI: Sitzungs-Auswahl zeigt abgeleitete Titel, fuzzy Suche, relative Zeiten und Vorschau der letzten Nachricht. (#1271) https://docs.clawd.bot/tui - TUI: eine durchsuchbare Modell-Auswahl hinzufügen für schnellere Modell-Auswahl. (#1198) https://docs.clawd.bot/tui - TUI: Eingabeliste (Auf-/Ab) für eingereichte Nachrichten hinzufügen. (#1348) https://docs.clawd.bot/tui - ACP: `clawdbot acp` für IDE-Integrationen hinzufügen. https://docs.clawd.bot/cli/acp - ACP: `clawdbot acp client` interaktives Harness zum Debugging hinzufügen. https://docs.clawd.bot/cli/acp - Skills: Download-Installationen mit OS-Filter-Optionen hinzufügen. https://docs.clawd.bot/tools/skills - Skills: die lokale sherpa-onnx-tts Skill hinzufügen. https://docs.clawd.bot/tools/skills - Memory: hybride BM25 + Vektor-Suche (FTS5) mit gewichteter Zusammenführung und Fallback hinzufügen. https://docs.clawd.bot/concepts/memory - Memory: SQLite-Embedding-Cache zum Beschleunigen von Reindexing und häufigen Updates hinzufügen. https://docs.clawd.bot/concepts/memory - Memory: OpenAI Batch Indexing für Embeddings bei Konfiguration hinzufügen. https://docs.clawd.bot/concepts/memory - Memory: OpenAI Batch Indexing standardmäßig für OpenAI-Embeddings aktivieren. https://docs.clawd.bot/concepts/memory - Memory: parallele OpenAI Batch Indexing-Jobs zulassen (Standard-Konkurrenz: 2). https://docs.clawd.bot/concepts/memory - Memory: Fortschritt sofort rendern, Batch-Status in verbose Logs farblich kennzeichnen und OpenAI-Batch-Status alle 2s standardmäßig abfragen. https://docs.clawd.bot/concepts/memory - Memory: `--verbose` Logging für Memory-Status + Batch-Indexing-Details hinzufügen. https://docs.clawd.bot/concepts/memory - Memory: native Gemini-Embeddings-Anbieter für Memory-Suche hinzufügen. (#1151) https://docs.clawd.bot/concepts/memory - Browser: Standardwerte für effiziente Snapshots im Tool/CLI zulassen. (#1336) https://docs.clawd.bot/tools/browser - Nostr: Nostr-Kanalkanal-Plugin mit Profilverwaltung + Onboarding-Standards hinzufügen. (#1323) https://docs.clawd.bot/channels/nostr - Matrix: Migration zu matrix-bot-sdk mit E2EE-Unterstützung, Standort-Handhabung und Upgrades des Gruppen-Allowlists. (#1298) https://docs.clawd.bot/channels/matrix - Slack: HTTP-Webhook-Modus über Bolt HTTP-Empfänger hinzufügen. (#1143) https://docs.clawd.bot/channels/slack - Telegram: Kontext von weitergeleiteten Nachrichten mit normalisierten Origin-Details + Legacy-Fallback anreichern. (#1090) https://docs.clawd.bot/channels/telegram - Discord: fallback auf `/skill`, wenn native Befehlslimits überschritten sind. (#1287) - Discord: globalen Zugriff auf `/skill` freischalten. (#1287) - Zalouser: Channel-Dock-Metadaten, Konfigurationsschema, Aufbau-Verkabelung, Probe und Statusprobleme hinzufügen. (#1219) https://docs.clawd.bot/plugins/zalouser - Plugins: Manifest-integrierte Config-Schemas mit Preflight-Validierungswarnungen required. (#1272) https://docs.clawd.bot/plugins/manifest - Plugins: Channel-Katalog-Metadaten in Plugin-Manifeste verschieben. (#1290) https://docs.clawd.bot/plugins/manifest - Plugins: Nextcloud Talk-Policy-Helfer mit Kernmustern ausrichten. (#1290) https://docs.clawd.bot/plugins/manifest - Plugins/UI: Channel-Plugin-Metadaten steuern UI-Bezeichnungen/Icons und Cron-Channel-Optionen. (#1306) https://docs.clawd.bot/web/control-ui - Plugins: Plugin-Slots mit dediziertem Memory-Slot-Auswähler hinzufügen. https://docs.clawd.bot/plugins/agent-tools - Plugins: das gebundene BlueBubbles Channel-Plugin ausliefern (standardmäßig deaktiviert). https://docs.clawd.bot/channels/bluebubbles - Plugins: gebundene Messaging-Erweiterungen auf das Plugin-SDK migrieren und Plugin-SDK-Importe im Loader lösen. - Plugins: Zalo-Plugin auf das shared Plugin-SDK-Laufzeitumgebung migrieren. https://docs.clawd.bot/channels/zalo - Plugins: Zalo Personal-Plugin auf das shared Plugin-SDK-Laufzeitumgebung migrieren. https://docs.clawd.bot/plugins/zalouser - Plugins: optionale Agent-Tools mit expliziten Allow-Listen erlauben und Guide zum Plugin-Tool-Erstellen hinzufügen. https://docs.clawd.bot/plugins/agent-tools - Plugins: gebundene Channel/Provider-Plugins automatisch aktivieren, wenn Konfiguration vorhanden ist. - Plugins: Plugin-Quellen bei Channel-Wechsel synchronisieren und Plugins, die via npm installiert sind, während `clawdbot update` aktualisieren. - Plugins: npm-Plugin-Aktualisierungslogik zwischen `clawdbot update` und `clawdbot plugins update` teilen. - Gateway/API: `/v1/responses` (OpenResponses) mit item-basiertem Input + semantischen Streaming-Ereignissen hinzufügen. (#1229) - Gateway/API: `/v1/responses` erweitern um Datei-/Bild-Eingaben, tool_choice, Nutzung und Output-Limits. (#1229) - Usage: `/usage cost` Übersichten und macOS-Menü-Kosten-Diagramme hinzufügen. https://docs.clawd.bot/reference/api-usage-costs - Security: Warnung, wenn Modelle <= 300B ohne Sandbox laufen, während Web-Tools aktiviert sind. https://docs.clawd.bot/cli/security - Exec: Host/Security/Ask Routing für Gateway + Node-Exec hinzufügen. https://docs.clawd.bot/tools/exec - Exec: `/exec` Direktiv für per-Session-Defaults (host/security/ask/node) hinzufügen. https://docs.clawd.bot/tools/exec - Exec approvals: Freigaben-Überprüfung auf `~/.clawdbot/exec-approvals.json` mit per-Agent-Allowlists + Skill-Auto-Allow Toggle migrieren, und Approvals-UI + Node-Exec-Lifecycle-Ereignisse hinzufügen. https://docs.clawd.bot/tools/exec-approvals - Nodes: Headless Node-Host hinzufügen (`clawdbot node start`) für `system.run`/`system.which`. https://docs.clawd.bot/cli/node - Nodes: Node-Daemon-Service Install/Status/Start/Stop/Restart hinzufügen. https://docs.clawd.bot/cli/node - Bridge: `skills.bins` RPC hinzufügen um Auto-Allow-Skill-Bins für Node-Host zu unterstützen. - Sessions: tägliche Rücksetzrichtlinie mit Typ-spezifischen Overrides und Idle-Fenstern (Standard 4 Uhr lokal), Legacy-Idle-Konfigurationen beibehalten. (#1146) https://docs.clawd.bot/concepts/session - Sessions: `sessions_spawn` zulassen, um das Thinking-Level für Sub-Agenten-Läufe zu überschreiben. https://docs.clawd.bot/tools/subagents - Channels: einheitliche Thread/Topic-Allowlist-Matching + Helper für Command/Mention-Gating über alle Core-Provider. https://docs.clawd.bot/concepts/groups - Models: Qwen Portal OAuth-Anbieter-Unterstützung hinzufügen. (#1120) https://docs.clawd.bot/providers/qwen - Onboarding: Allowlist-Eingabeaufforderungen und Username-zu-ID-Auflösung über Core- und Extensions-Channel hinweg hinzufügen. https://docs.clawd.bot/start/onboarding - Docs: klären Allowlist-Eingabetypen und Onboarding-Verhalten für Messaging-Channels. https://docs.clawd.bot/start/onboarding - Docs: Android-Knoten-Erkennungsdokumentation für Gateway WS-Service-Typ aktualisieren. https://docs.clawd.bot/platforms/android - Docs: Amazon Bedrock in Providerschlüssel-Listen aufführen und Bedrock-Auth-Umgebungsvariablen klären. (#1289) https://docs.clawd.bot/bedrock - Docs: WhatsApp-Sprachnachrichten klären. https://docs.clawd.bot/channels/whatsapp - Docs: Windows WSL Portproxy LAN-ZugriffsNotizen klären. https://docs.clawd.bot/platforms/windows - Docs: Bird-Skill-Installations-Metadaten und

Gemini CLIv0.25.0

Release v0.25.0

21. Januar 2026

## Was sich geändert hat * feat(core): verbessere activate_skill Tool und verwende Kleinbuchstaben XML-Tags von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16009 * Initiierungsmethode Telemetrie-Eigenschaft hinzufügen von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15818 * chore(release): Version auf 0.25.0-nightly.20260107.59a18e710 erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/16048 * Hx-Unterstützung von @kevinfjiang in https://github.com/google-gemini/gemini-cli/pull/16032 * [Skills] Foundation: Zentralisiere Verwaltungslogik und Feedback-Rendern von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15952 * Einführung von GEMINI_CLI_HOME für strikte Testisolation von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15907 * [Skills] Mehrere Bereiche Skill-Aktivierung und Shadowing-Fehlerbehebung von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15953 * policy: Legacy-Policy aus dem Core-Tool-Scheduler in Policy-Engine extrahieren von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15902 * Verbesserung von TestRig mit Prozessverwaltung und Timeouts von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15908 * Troubleshooting-Dokumentation aktualisieren für UNABLE_TO_GET_ISSUER_CERT_LOCALLY von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16069 * Keytar zu Abhängigkeiten hinzufügen von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/15928 * Vereinfachung des Extension-Einstellungen-Befehls von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/16001 * feat(admin): Extensions deaktiviert implementieren von @skeshive in https://github.com/google-gemini/gemini-cli/pull/16024 * Core-Datenstruktur-Updates für Rewind-Funktionalität von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/15714 * feat(hooks): Vereinfachung des Hook-Firing mit HookSystem Wrapper-Methoden von @ved015 in https://github.com/google-gemini/gemini-cli/pull/15982 * Add exp.gws_experiment Feld zu LogEventEntry von @gsquared94 in https://github.com/google-gemini/gemini-cli/pull/16062 * Rückgängigmachen von "feat(admin): implement extensions disabled" (#16082) von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/16082 * feat(core): Entkopplung des UI für das Aktivieren von Hooks vom Subsystem von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/16074 * Docs: Dokumentation für Hooks + Extensions hinzufügen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/16073 * feat(core): Vorläufige Änderungen für Subagent-Modell-Routing von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/16035 * CI-Workflow optimieren: Jobs parallelisieren und Linters cachen von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16054 * Option für Fallback bei Kapazitätsfehlern in ProQuotaDi… hinzufügen von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16050 * feat: Bestätigung Details-Unterstützung + jsonrpc vs http rest-Unterstützung von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/16079 * fix(workflows): Labels für pr-triage.sh Skript korrigieren und begrenzen von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/16096 * Fix und Umbenennung introspection agent -> cli help agent von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16097 * Docs: Changelog Aktualisierung 20260105 von @jkcinouye in https://github.com/google-gemini/gemini-cli/pull/15937 * Enable cli_help Agent standardmäßig von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16100 * Optimierte json-output Tests mit Mock-Antworten von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16102 * CI für Forks korrigieren von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16113 * Reduziere Mahnungen zu PRs, die auf Issues verweisen, aber diese nicht lösen von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/16112 * feat(cli): Pfad-Autovervollständigung nach Slash-Befehlen hinzufügen von @jasmeetsb in https://github.com/google-gemini/gemini-cli/pull/14738 * Upgrade-Option für zahlende Nutzer hinzufügen von @cayden-google in https://github.com/google-gemini/gemini-cli/pull/15978 * [Skills] UX-Polish: Transparentes Feedback und CLI-Feinabstimmungen von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15954 * Polish: Warnung „Failed to load skills“ in Debug-Logs verschieben von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16142 * feat(cli): Chat-Historie exportieren in /bug und GitHub-Issue vorbelegen von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16115 * bug(core): Problem mit Overrides zu Bases beheben von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/15255 * enableInteractiveShell für externes Tooling, das auf a2a-Server angewiesen ist von @DavidAPierce in https://github.com/google-gemini/gemini-cli/pull/16080 * Reapply "feat(admin): implement extensions disabled" (#16082) von @skeshive in https://github.com/google-gemini/gemini-cli/pull/16109 * bug(core): Spewie Getter in `hookTranslator.ts` beheben von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/16108 * feat(hooks): explicit stop and block Ausführungskontrolle in Model-Hooks hinzufügen von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15947 * Verfeinere Gemini 3 System-Anweisungen zur Reduzierung der Model-Verbosität von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16139 * chore: unbenutzte Modelle bereinigen und Konstanten verwenden von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/16246 * Immer geklammertes Einfügen aktivieren von @scidomino in https://github.com/google-gemini/gemini-cli/pull/16179 * refactor: migrate clearCommand Hook-Aufrufe zu HookSystem von @ved015 in https://github.com/google-gemini/gemini-cli/pull/16157 * refactor: migrate App-Containner Hook-Aufrufe zum Hook-System von @ishaanxgupta in https://github.com/google-gemini/gemini-cli/pull/16161 * Zeige Einstellungsquelle in Extensions-Listen an von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/16207 * feat(skills): pr-creator Skill hinzufügen und Skills aktivieren von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/16232 * fix: Shift+Space im Kitty Keyboard Protocol-Terminals behandeln von @tt-a1i in https://github.com/google-gemini/gemini-cli/pull/15767 * feat(core, ui): Addiere `/agents refresh` Befehl. von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/16204 * feat(core): lokale Experimente-Überschreibungen via GEMINI_EXP hinzufügen von @kevin-

Claude Codev2.1.14

v2.1.14

20. Januar 2026

## What's changed - Added history-based autocomplete in bash mode (`!`) - type a partial command and press Tab to complete from your bash command history - Added search to installed plugins list - type to filter by name or description - Added support for pinning plugins to specific git commit SHAs, allowing marketplace entries to install exact versions - Fixed a regression where the context window blocking limit was calculated too aggressively, blocking users at ~65% context usage instead of the intended ~98% - Fixed memory issues that could cause crashes when running parallel subagents - Fixed memory leak in long-running sessions where stream resources were not cleaned up after shell commands completed - Fixed `@` symbol incorrectly triggering file autocomplete suggestions in bash mode - Fixed `@`-mention menu folder click behavior to navigate into directories instead of selecting them - Fixed `/feedback` command generating invalid GitHub issue URLs when description is very long - Fixed `/context` command to show the same token count and percentage as the status line in verbose mode - Fixed an issue where `/config`, `/context`, `/model`, and `/todos` command overlays could close unexpectedly - Fixed slash command autocomplete selecting wrong command when typing similar commands (e.g., `/context` vs `/compact`) - Fixed inconsistent back navigation in plugin marketplace when only one marketplace is configured - Fixed iTerm2 progress bar not clearing properly on exit, preventing lingering indicators and bell sounds - Improved backspace to delete pasted text as a single token instead of one character at a time - [VSCode] Added `/usage` command to display current plan usage

OpenCodev1.1.28

v1.1.28

20. Januar 2026

## Core - Preserve tool input from running state for MCP tool results (@rynfar) - Pass arguments to commands without explicit placeholders (@remorses) - Adjust Codex prompt to discourage unnecessary questions and encourage more autonomous decision-making - Register OAuth callback before opening browser in MCP (@msvechla) - Support ACP audience by mapping to ignore and synthetic (@noamzbr) - Terminal no longer hangs on exit or Ctrl+D and closes the pane properly (@ProdigyRahul) ## TUI - Prevent sidebar height from overflowing in terminal UI (@IdrisGit) ## Desktop - Refactor Select component to use settings variant for settings modal styling - Show "View all sessions" button for active project and close hovercard on click - Remove close delay on hover cards to prevent overlapping - Fix review empty state font size - Fix race condition when editing project names (@DNGriffin) - Resolve Select children type conflict with ButtonProps - Fix prompt agent button style - Fix type error - Update keyboard shortcuts panel to match general settings styling - Add color scheme preview on hover in appearance dropdown - Update settings panel padding and make content full width - Set 32px spacing between main title and group title - Add fade gradient to settings panel headers - Reset select trigger to default state after selection - Add extra large dialog size and use it for settings modal - Add triggerStyle prop to Select component and apply it to font selector - Render font options in their respective fonts - Wrap settings row groups with styled container - Remove border and background from settings panel headers - Remove subheader from General settings panel - Add keyboard icon and use it for Shortcuts settings tab - Add sliders icon and use it for General settings tab - Update settings tabs layout and spacing - Increase icon letter size to 32px in edit project dialog - Increase session messages popover open delay to 1000ms - Fix end-to-end tests in app package - Add Windows end-to-end tests for the app - Change terminal.new keybind to Ctrl+Alt+T (@ProdigyRahul) - Cleanup PTY exited event listener on component unmount (@ProdigyRahul) - Fix duplicate session loads in app - Add tooltips to sidebar new session and workspace buttons (@ProdigyRahul) - Better pending states for workspace operations - Fix session loading loop in app - Add restart and reload menu items on macOS (@BlankParticle) - Fix broken app functionality - Add settings page to app **Thank you to 13 community contributors:** - @ProdigyRahul: - fix(app): terminal no longer hangs on exit or ctrl + D and closes the pane (#9506) - feat(app): add tooltips to sidebar new session/workspace buttons (#9652) - fix(app): cleanup pty.exited event listener on unmount (#9671) - fix(app): change terminal.new keybind to ctrl+alt+t (#9670) - @BlankParticle: - feat: add restart and reload menu items on macOS (#9212) - @noamzbr: - feat: Support ACP audience by mapping to ignore and synthetic (#9593) - @msvechla: - fix(mcp): register OAuth callback before opening browser (#9646) - @remorses: - fix: pass arguments to commands without explicit placeholders (#9606) - @zerone0x: - fix(ui): show filename in Edit/Write permission titles (#9662) - @MBanucu: - fix(opencode): Allow compatible Bun versions in packageManager field (#9597) - @DRunkPiano114: - fix(ui): prevent Enter key action during IME composition (#9564) - @IdrisGit: - fix(tui): prevent sidebar height from overflowing. (#9689) - @gigamonster256: - fix(opencode): relax bun version requirement (#9682) - @DNGriffin: - fix(app): Edit-project name race condition (#9551) - @rynfar: - fix(opencode): preserve tool input from running state for MCP tool results (#9667) - @kdcokenny: - docs(web): add KDCO plugins to ecosystem (#7694)

OpenCodev1.1.27

v1.1.27

20. Januar 2026

## Core - Remove filetime assertions from patch tool - Ensure truncation handling applies to MCP servers - Update batch tool definition to outline correct value for max tool calls (@jamesmengo) ## TUI - Add space toggle hint to tool selection prompt (@cjellick) - Fix message history cleanup to prevent memory leaks in TUI ## Desktop - Fixed flash of home page on application start - Fixed localStorage quota handling in app - Persist quota limits to prevent loss on reload - Completely disable pinch to zoom in desktop application (@Brendonovich) - Fix message list overflow and scrolling issues (@neriousy) - Add 8px padding to recent sessions popover - Remove top padding from edit project dialog form - Update session messages popover gutter to 28px - Position session messages popover at top - Retain session hover state when popover is open and update border radius - Update session hover popover styling - Update review session empty state styling - Add ability to select project directory text in web interface (@DNGriffin) **Thank you to 5 community contributors:** - @DNGriffin: - feat(app): Add ability to select project directory text to web (#9344) - @neriousy: - fix(app): message list overflow & scrolling (#9530) - @jamesmengo: - fix(batch): update batch tool definition to outline correct value for max tool calls (#9517) - @cjellick: - fix: add space toggle hint to tool selection prompt (#9535) - @Brendonovich: - fix(desktop): completely disable pinch to zoom

Gemini CLIv0.24.5

Release v0.24.5

20. Januar 2026

**Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.24.4...v0.24.5

Continue.devv1.5.35

CLI Stabile Veröffentlichung v1.5.35

20. Januar 2026

Direktes stabiles Release aus dem main-Branch. Version 1.5.35 wurde direkt aus dem neuesten main-Branch gebaut und veröffentlicht.

OpenCodev1.1.26

v1.1.26

19. Januar 2026

## Core - Increase maximum batch tool limit from 10 to 25 (@spoons-and-mirrors) - Remove favicon from application - Apply patch description - Format only edited file in Cargo workspace instead of entire workspace (@goniz) - Cargo fmt does not support formatting single files - Rename uv formatter from 'uv format' to 'uv' for config consistency (@paulclou) - Initial e2e test setup for app - Reset worktree to clean state - Delete workspace functionality - Tweak apply_patch tool description - Omit todo tools for OpenAI models - Add proper variant support to copilot - Track reasoning by output_index for Copilot compatibility (@NateSmyth) - Add scheduler and cleanup module - Auto-route GPT-5+ models to Responses API in GitHub Copilot integration (@christso) - Use streamObject when using OpenAI OAuth in agent generation (@thmalek) - Fix incorrect model variants for Grok models - Note untracked files in review documentation - Treat .fbs files as text instead of images in tool handling (@zerone0x) - Update GitLab AI provider to version 1.3.2 (@vglafirov) - Add command execute before hook (@spoons-and-mirrors) - Overhaul Nix flake and packages (@gigamonster256) - Preserve file attachment metadata during session replay (@liorshk) - Update Scoop installation method (@chawyehsu) - Fix import issue in patch module - Restore persisted model and agent when loading ACP session (@noamzbr) - Use single global event subscription and route by session ID in ACP (@noamzbr) - Increase question header and label limits (@bdwelle) - Add apply_patch tool for OpenAI models - Fix Java language server download on Windows (@pschiel) - Add vim-style line-by-line scrolling keybindings (@NathanFlurry) - Prevent environment variables in config from being replaced with actual values - Fix mDNS hostname discovery (@j0nl1) - Revert OAuth redirect URI configuration for MCP servers - Fix GPT ID handling in API responses - Add OAuth polling safety margin in Copilot device authentication (@hamir-suspect) - Open help dialog with tui/open-help route (@AksharP5) ## TUI - Clarify resume session tip in TUI (@kearns-cu) - Set input mode based on whether mouse or keyboard is in use to prevent unwanted mouse events (@jcampuza) - Don't unnecessarily wrap lines and introduce an unneeded empty line (@ariane-emory) - Add mouse support for permission buttons in TUI (@Hona) - Fix Windows evaluating text on copy (@Hona) - Uninstall OpenCode installed via Windows package managers (@itsrainingmani) - Unify command registry and derive slash commands from command definitions (@kitlangton) - Hide variants hint when list is empty - Fix model state persistence when model store is not ready - Centralize OSC 52 clipboard support for SSH sessions (@b3nw) - Keep primary model after subagent runs (@sethcarlton) - Make home/end keys work in menu list modal windows (@ariane-emory) ## Desktop - Fix favicon in app - Change keybind for cycling thinking effort in app (@neriousy) - Fixed global terminal and review pane toggle functionality - Fix hash-based navigation in the application - Give feedback when trying to paste an unsupported file type (@neriousy) - Rename favicons to v2 for cache busting (@R44VC0RP) - Reload instance after workspace reset - Make terminal sessions scoped to workspace - Add hover overlay with upload and trash icons to project icon in edit dialog - Tighten edit project color spacing - Constrain edit project dialog width - Tweak edit project icon container styling - Adjust edit project icon helper text - Update edit project dialog padding and avatar styles for consistency - New layout improvements for app - New session layout - Improved session layout - Hide settings button in app - Add more end-to-end tests for the app - Add end-to-end testing utilities for the application - Tweak share button to prevent layout shift in desktop (@itsrainingmani) - Apply getComputedStyle polyfill on all platforms in desktop app (@Slone123c) - Improve layout and styling of desktop session search button (@OpeOginni) - Remove redundant toast for thinking effort changes (@kitlangton) - Add session unshare button to desktop application (@itsrainingmani) - Remove max-width constraint from session name tooltip - Adjust project path tooltip placement - Remove max width on sidebar new buttons - Adjust session list tooltip trigger and delay - Keep project avatar hover styles while popover is open - Adjust recent sessions popover padding - Reduce prompt dock bottom spacing - Show project name in avatar hover - Remove loading text after splash screen - Fix command palette file list item spacing - Shrink project notification dot and apply masking - Update desktop initializing splash logo - Align project avatar notification dot - Don't try to open command palette if a dialog is already open (@BlankParticle) - Added Windows-only guard to make window.getComputedStyle fall back to document.documentElement (@Eric-Guo) - Fix flaky session archiving behavior - Prevent text clipping on search button descenders - Remove secondary text from commands - Limit search modal max-height to 480px - Refine search modal styling and list component - Reduce command item left padding in search modal - Add hideIcon and class options to List search and customize search modal input styling - Adjust command item left padding in search modal - Add left padding to command items in search modal - Increase keybind border radius in search modal - Display arrow keys as symbols in keybind formatting - Use Keybind component in search modal list - Use Keybind component in titlebar search button - Fix recent sessions gutter spacing ## Extensions - Add use_github_token to example **Thank you to 39 community contributors:** - @ariane-emory: - fix: make home/end keys work in menu list modal windows (resolves #7190) (#8347) - fix: Don't unnecessarily wrap lines and introduce an unneeded empty line (resolves #9489) (#9488) - @sethcarlton: - fix: keep primary model after subagent runs (#8951) - @AksharP5: - fix: open help dialog with tui/open-help route (#8596) - @b3nw: - fix: centralize OSC 52 clipboard support for SSH sessions (#8974) - @gigamonster256: - fix(nix): add desktop application entry (#8972) - feat(nix): overhaul nix flake and packages (#9032) - fix: nix ci - swapped dash/underscore (#9352) - @dbpolito: - fix(desktop): Stream bash output + strip-asni (#8961) - @hamir-suspect: - fix(opencode): add oauth polling safety margin in copilot device authentication (#8986) - @jerome-benoit: - fix(nix): resolve hash race condition in parallel matrix jobs (#8995) - @benjaminshafii: - docs: Add OpenWork to ecosystem (#8741) - @Eric-Guo: - fix(desktop): Added a Windows-only guard that makes window.getComputedStyle fall back to document.documentElement (#9054) - @j0nl1: - fix: mdns discover hostname (#9039) - @Slone123c: - fix: escape CSS selector keys to handle special characters (#9030) - fix(desktop): apply getComputedStyle polyfill on all platforms (#9369) - @cgilly2fast: - docs: add firmware provider to providers docs (#8993) - @berenar: - docs(config): explain that `autoupdate` doesn't work when installed with a package manager (#9092) - @BlankParticle: - fix: don't try to open command palette if a dialog is already open (#9116) - @kitlangton: - refactor(tui): unify command registry and derive slash commands (#9115) - fix(app): remove redundant toast for thinking effort changes (#9181) - @NathanFlurry: - feat: bind vim-style line-by-line scrolling (#8980) - @pschiel: - fix(windows): fix jdtls download on Windows (#9195) - @bdwelle: - feat(tool): increase question header and label limits (#9201) - @noamzbr: - fix(acp): use single global event subscription and route by sessionID (#5628) - fix: restore persisted model/agent when loading ACP session (#7809) - @itsrainingmani: - feat(cli): uninstall opencode installed via windows package managers (#8571) - feat(desktop): implement session unshare button (#8660) - refactor(desktop): tweak share button to prevent layout shift (#9322) - @OpeOginni: - refactor(desktop): improve layout and styling of session search button (#9251) - @chawyehsu: - refactor(installation): update scoop installation method (#9243) - @liorshk: - fix(acp): preserve file attachment metadata during session replay (#6342) - @spoons-and-mirrors: - feat(hook): command execute before hook (#9267) - tweak(batch): up restrictive max batch tool from `10` to `25` (#9275) - @vglafirov: - fix: update gitlab-ai-provider to 1.3.2 (#9279) - docs: Improve Gitlab self-hosted instances documentation (#9391) - @zerone0x: - fix(tool): treat .fbs files as text instead of images (#9276) - @alanpog: - docs: clarify agent tool access and explore vs general distinction (#9300) - @Hona: - fix: Windows evaluating text on copy (#9293) - feat(tui): use mouse for permission buttons (#9305) - @thmalek: - fix(opencode): use streamObject when using openai oauth in agent generation (#9231) - @christso: - fix(github-copilot): auto-route GPT-5+ models to Responses API (#5877) - @NateSmyth: - fix: track reasoning by output_index for copilot compatibility (#9124) - @kedMertens: - docs: update codecompanion.nvim acp doc (#9411) - @paulclou: - fix(config): rename uv formatter from 'uv format' to 'uv' for config consistency (#9409) - @jcampuza: - fix(app): list component jumping when mouse happens to be under the list and keyboard navigating. (#9435) - fix(opencode): sets input mode based on whether mouse vs keyboard is in use to prevent mouse events firing (#9449) - @goniz: - fix(opencode): `cargo fmt` is formatting whole workspace instead of edited file (#9436) - @R44VC0RP: - fix(web): rename favicons to v2 for cache busting (#9492) - @neriousy: - fix(app): give feedback when trying to paste a unsupported filetype (#9452) - fix: keyboard nav when mouse hovered over list (#9500) - fix(app): change keybind for cycling thinking effort (#9508) - @kearns-cu: - fix(tui): clarify resume session tip (#9490)

Gemini CLIv0.24.4

Release v0.24.4

19. Januar 2026

## Änderungen * fix(patch): cherry-pick 88f1ec8 to release/v0.24.0-pr-16179 [CONFLICTS] by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/16783 **Full Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.24.0...v0.24.4

Claude Codev2.1.12

v2.1.12

17. Januar 2026

## What's changed - Fixed message rendering bug

OpenClawv2026.1.16-2

clawdbot 2026.1.16-2

17. Januar 2026

### Highlights - Hooks: System zum Hinzufügen von Hooks mit gebündelten Hooks, CLI-Tools und Dokumentation. (#1028) — danke @ThomsenDrake. https://docs.clawd.bot/hooks - Media: inbound media Understanding (Bild/Audio/Video) mit Provider + CLI-Fallbacks hinzufügen. https://docs.clawd.bot/nodes/media-understanding - Plugins: Zalo Personal Plugin (`@clawdbot/zalouser`) hinzufügen und Verzeichnisstruktur der Plugin-Kanäle vereinheitlichen. (#1032) — danke @suminhthanh. https://docs.clawd.bot/plugins/zalouser - Models: Auth-Auswahl für Vercel AI Gateway hinzufügen + Onboarding-Updates. (#1016) — danke @timolins. https://docs.clawd.bot/providers/vercel-ai-gateway - Sessions: `session.identityLinks` für plattform-übergreifende DM-Session-Verlinkung hinzufügen. (#1033) — danke @thewilloftheshadow. https://docs.clawd.bot/concepts/session - Web search: Parameter `country`/`language` (Schema + Brave API) und Docs hinzufügen. (#1046) — danke @YuriNachos. https://docs.clawd.bot/tools/web ### Breaking - **BREAKING:** `clawdbot message` und Nachrichten-Tool benötigen jetzt `target` (Verwerfen von `to`/`channelId` für Ziele). (#1034) — danke @tobalsan. - **BREAKING:** Channel-Auth bevorzugt jetzt Config über Env für Discord/Telegram/Matrix (Env-Fallback nur). (#1040) — danke @thewilloftheshadow. - **BREAKING:** Drop der Legacy-Unterstützung für `chatType: "room"`; verwenden Sie `chatType: "channel"`. - **BREAKING:** Entferne legacy provider-spezifische Zielauflösungs-Fallbacks; Zielauflösung zentralisiert mit Plugin-Hints + Verzeichnis-Lookups. - **BREAKING:** `clawdbot hooks` ist jetzt `clawdbot webhooks`; Hooks befinden sich unter `clawdbot hooks`. https://docs.clawd.bot/cli/webhooks - **BREAKING:** `clawdbot plugins install <path>` kopiert jetzt in `~/.clawdbot/extensions` (verwenden Sie `--link`, um Pfad-basierte Laden zu behalten). ### Changes - Plugins: gebündelte Plugins standardmäßig deaktiviert; Überschreibungen durch installierte Versionen erlauben. (#1066) — danke @ItzR3NO. - Plugins: gebündelte Antigravity + Gemini CLI OAuth + Copilot Proxy Provider-Plugins hinzufügen. (#1066) — danke @ItzR3NO. - Tools: Verbesserung der `web_fetch`-Extraktion mittels Readability (mit Fallback). - Tools: Firecrawl-Fallback für `web_fetch`, wenn konfiguriert. - Tools: Chrome-ähnliche Header standardmäßig für `web_fetch` senden, um Extraktion auf bot-sensitiven Seiten zu verbessern. - Tools: Firecrawl-Fallback nutzt standardmäßig Bot-Umgehung + Cache; entfernen Sie den einfachen HTML-Fallback bei fehlender Extraktion. - Tools: Standardmäßige Exit-Benachrichtigungen von `exec` und automatische Migration des legacy `tools.bash` zu `tools.exec`. - Tools: `exec` PTY-Unterstützung für interaktive Sessions hinzufügen. https://docs.clawd.bot/tools/exec - Tools: tmux-Style `process send-keys` und bracketed paste-Helfer für PTY-Sitzungen. - Tools: `process submit`-Hilfsfunktion zum Senden von CR für PTY-Sitzungen. - Tools: Antworten auf PTY-Cursor-Position-Anfragen, um interaktive TUIs zu freischalten. - Tools: Tool-Ausgaben in verbose Mode einschließen und ausführliches Tool-Feedback erweitern. - Skills: Aktualisierung der Coding-Agent-Richtlinien, bevorzugte PTY-fähige Exec-Läufe und Vereinfachung der tmux-Nutzung. - TUI: Token-Anzahlen der Session nach Abschluss oder Fehlschlag aktualisieren. (#1079) — danke @d-ploutarchos. - Status: `/status` auf aktuelle Provider-Nutzung beschränken und OAuth/Token-Block entfernen. - Directory: Vereinheitlichung des `clawdbot directory` über Kanäle und Plugin-Kanäle hinweg. - UI: Löschen von Sessions im Control UI zulassen. - Skills: benutzbare Skill-Befehle hinzufügen und erweiterte Skill-Befehl-Registrierung. - Telegram: Standardreaktionsstufe auf minimal setzen und Reaktions-Benachrichtigungen standardmäßig aktivieren. - Telegram: Antworten-Ketten-Nachrichten dürfen Erwähnungs-Gating in Gruppen umgehen. (#1038) — danke @adityashaw2. - iMessage: Remote-Anhang-Unterstützung für VM/SSH-Deployments hinzufügen. - Messages: Live-Verzeichnis-Cache-Ergebnisse beim Auflösen von Zielen aktualisieren. - Messages: Delivery-Text/Media in Transcripts der Sessions spiegeln. (#1031) — danke @TSavo. - Messages: Redundante Absender-Umhüllungen für iMessage + Signal-Gruppenchats vermeiden. (#1080) — danke @tyler6204. - Media: Deepgram Audio-Upload-Bytes für Fetch-Kompatibilität normalisieren. - Cron: isolierte Cron-Jobs starten bei jedem Lauf eine frische Session-ID, um Kontextaufbau zu verhindern. - Docs: `/help`-Hub, Node/npm PATH-Anleitung und erweiterte Directory-CLI-Dokumentation hinzufügen. - Config: Umgebungsvariablen-Substitution in Config-Werten unterstützen. (#1044) — danke @sebslight. - Health: pro-Agenten-Sitzungszusammenfassungen und kontenbasierte Health-Details hinzufügen, sowie selektive Probes ermöglichen. (#1047) — danke @gumadeiras. - Hooks: Hook-Paket-Installationen (npm/Pfad/Zip/Tar) mit `clawdbot.hooks`-Manifests und `clawdbot hooks install/update`. - Plugins: Zip-Installationen hinzufügen und `--link`, um Kopieren lokaler Pfade zu vermeiden. ### Fixes - macOS: Subprozess-Pipes leeren, bevor gewartet wird, um Deadlocks zu vermeiden. (#1081) — danke @thesash. - Verbose: Tool-Zusammenfassungen/Ausgaben nur in Markdown-Channels formatieren. - Telegram: tg/Group/Telegram-Präfixe + Target-Topics für Inline-Button-Validierung akzeptieren. (#1072) — danke @danielz1z. - Telegram: Lange Bildunterschriften in Follow-up-Nachrichten aufteilen. - Config: Startvorgang bei ungültiger Config blockieren, bestmögliche Doctor-Config bewahren und rollende Config-Backups beibehalten. (#1083) — danke @mukhtharcm. - Sub-Agents: Normalisieren von Announcement-Delivery-Origin + Queue-Bucketing nach accountId, um Multi-Account-Routing stabil zu halten. (#1061, #1058) — danke @adam91holt. - Sessions: DeliveryContext in sessions.list aufnehmen und normalisierte Delivery-Routing für Announcement/Restart-Fallbacks wiederverwenden. (#1058) - Sessions: DeliveryContext in Last-Route-Updates propagieren, um Routing-Stabilität zu bewahren. (#1058) - Sessions: Overrides beim `/new`-Reset beibehalten. - Memory: Ungehandelte Rejections verhindern, wenn Watch/Interval-Sync fehlschlägt. (#1076) — danke @roshanasingh4. - Memory: Gateway-Absturz vermeiden, wenn Embeddings 429/insufficient_quota zurückgeben (Tool deaktivieren + Fehler anzeigen). (#1004) - Gateway: Explizite Delivery Targets beachten, kein impliziter accountId-Fallback; letzte AccountId für implizites Routing beibehalten. - Gateway: Vermeide Wiederverwendung von Last-zu/accountId, wenn der angeforderte Kanal abweicht; DeliveryContext mit letzten Route-Feldern synchronisieren. - Build: `@lydell/node-pty`-Builds auf unterstützten Plattformen zulassen. - Repo: Oxlint-Konfigurations-Dateiname korrigieren und Ignore-Muster in Konfig verschieben. (#1064) — danke @connorshea. - Messages: `/stop` bricht jetzt hart laufende Follow-ups und Sub-Agenten-Läufe ab; Null-Zähl-Stop-Notizen unterdrücken. - Messages: Berücksichtige Channel, wenn Nachrichten-Tools zusammengeführt werden. - Messages: Sender-Labels für Live-Gruppen-Nachrichten über alle Kanäle hinweg einbeziehen, passend zu Queued/Historic-Formatierung. (#1059) - Sessions: `compactionCount` bei `/new` und `/reset` zurücksetzen und Dateimodus von `sessions.json` beibehalten (0600). - Sessions: Nicht mehr verbundene Benutzer-Turns vor eingebetteten Prompts reparieren. - Sessions: harten Stop von

Claude Codev2.1.11

v2.1.11

17. Januar 2026

## What's changed - Fixed excessive MCP connection requests for HTTP/SSE transports

OpenCodev1.1.25

v1.1.25

16. Januar 2026

## Core - Improve wording - Revert tool attachment fix to prevent them being sent as user messages - Reinstall plugins when cache module is missing (@kenryu42) ## Desktop - Remember last opened project in app - Truncate workspace title in app - Handle new session correctly in app - Center dialog on page instead of session **Thank you to 3 community contributors:** - @kenryu42: - fix(bun): reinstall plugins when cache module missing (#8815) - @ariane-emory: - fix: make home/end keys work in menu list modal windows (resolves #7190) (#8347)

OpenCodev1.1.24

v1.1.24

16. Januar 2026

## Core - Upgrade OpenTUI to v0.1.74 and fix tmux kitty keyboard regression - Ensure that tool attachments aren't sent as user messages - Use dynamic import for session event in config.ts to avoid circular dependency - Fix broken URL in documentation (@berenar) - Fix itemId stripping logic for GPT models - Revert test fix for image handling - Update GPT models prompt - Add OAuth redirect URI configuration for MCP servers (@christso) - Adjust websearch tool to emphasize current date and clarify it's no longer 2024 - Clarify question tool guidance in documentation (@kitlangton) - Set OpenCode as user agent for most inference requests - Show auth URL when browser cannot open in remote sessions (@danlapid) - Strip item IDs in more cases - Stop changing main model/agent from subtasks invocation (@dbpolito) - Fix default agent selection in ACP and headless mode (@assagman) - Add LiteLLM proxy provider option for explicit compatibility (@seilk) ## TUI - Add version to session header and /status dialog (@AksharP5) - Correct theme count tip in TUI (@kitlangton) - Dim question option prefixes in terminal UI (@kitlangton) - Add tab navigation support for questions in terminal interface (@kitlangton) - Add Carbonfox theme (@devatnull) ## Desktop - Add ability to edit project and session titles - Open workspace if navigating to session in workspace - Can't expand workspaces - Fixed scroll jumping when expanding workspaces - Truncate the workspace name on hover - Fix alignment of load more label - Update panel minimum size and button maximum width - Fix session icon and label alignment - Fixed desktop hamburger menu shifting issue - Center hamburger menu with project avatars - Fix mobile menu on web - Fix responsive menu on desktop - Moved createMemo down to fix bug - Updated project and sessions list width - Fix styles for view all sessions state - Add default icon to sessions - Fix search bar sizing, padding, and keyboard shortcut styling - Fixed avatar border radius and current project display - Persist workspace order and collapsed state - Add unified search for commands and files in app - Add skeleton loader for sessions - Fix project avatar border radius ## SDK - Update OpenAPI TypeScript SDK generator to version 0.90.4 - Revert SDK OpenAPI TypeScript generator update to 0.90.4 - Update @hey-api/openapi-ts to 0.90.4 (@coleleavitt) **Thank you to 13 community contributors:** - @seilk: - feat: add litellmProxy provider option for explicit LiteLLM compatibility (#8658) - @devatnull: - feat: add Carbonfox theme (#8723) - @johnnyapu15: - fix:subagent reasoningEffort not being applied (#8646) - @aaroniker: - feat(console): Fix /black page View Transition Safari issue (#8755) - feat(console): /black shader improvements, performance, details (#8871) - @assagman: - fix(agent): default agent selection in acp and headless mode (#8678) - @dbpolito: - fix: stop changing main model/agent from subtasks invocation (#7681) - feat(desktop): Terminal Splits (#8767) - @ricardo-valero: - fix: nix desktop workflow (#8747) - @danlapid: - fix(mcp): show auth URL when browser cannot open in remote sessions (#7884) - @kitlangton: - docs: clarify question tool guidance (#8778) - fix(tui): add tab navigation in questions (#8777) - fix(tui): dim question option prefixes (#8776) - fix(tui): correct theme count tip (#8779) - @AksharP5: - feat: add version to session header and /status dialog (#8802) - @christso: - feat(mcp): add OAuth redirect URI configuration for MCP servers (#7379) - @coleleavitt: - chore(sdk): update @hey-api/openapi-ts to 0.90.4 (#8921) - @berenar: - fix(docs): Broken URL (#8918)

Codex CLIvrust-v0.87.0

0.87.0

16. Januar 2026

## New Features - User message metadata (text elements and byte ranges) now round-trips through protocol/app-server/core so UI annotations can survive history rebuilds. (#9331) - Collaboration wait calls can block on multiple IDs in one request, simplifying multi-thread coordination. (#9294) - User shell commands now run under the user snapshot so aliases and shell config are honored. (#9357) - The TUI now surfaces approval requests from spawned/unsubscribed threads. (#9232) ## Bug Fixes - Token estimation during compaction is now accurate, improving budgeting during long sessions. (#9337) - MCP CallToolResult now includes `threadId` in both `content` and `structuredContent`, and returns a defined output schema for compatibility. (#9338) - The TUI “Worked for” separator only appears after actual work has occurred. (#8958) - Piped non-PTY commands no longer hang waiting on stdin. (#9369) ## Documentation - MCP interface docs updated to reflect structured output schema and `threadId` behavior. (#9338) ## Chores - Windows builds enable the PowerShell UTF-8 feature by default. (#9195) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.86.0...rust-v0.87.0 - #9331 Add text element metadata to protocol, app server, and core @charley-oai - #9195 chore(windows) Enable Powershell UTF8 feature @dylan-hurd-oai - #8958 fix(tui): only show 'Worked for' separator when actual work was performed @ThanhNguyxn - #9338 fix(mcp): include threadId in both content and structuredContent in CallToolResult @bolinfest - #9337 Fix token estimate during compaction @etraut-openai - #9336 rename model turn to sampling request @aibrahim-oai - #9232 feat: propagate approval request of unsubscribed threads @jif-oai - #9357 feat: run user commands under user snapshot @jif-oai - #9294 feat: collab wait multiple IDs @jif-oai - #9366 feat: CODEX_CI @jif-oai - #9367 prompt collab @jif-oai - #9369 chore: close pipe on non-pty processes @jif-oai

OpenClawv2026.1.15

clawdbot 2026.1.15

16. Januar 2026

### Highlights - Plugins: Provide-Auth-Registry hinzufügen + `clawdbot models auth login` für plugin-gesteuerte OAuth/API-Key-Flows. - Browser: remote CDP/Browserless Support verbessern (Auth-Passthrough, `wss`-Upgrade, Timeouts, klarere Fehler). - Heartbeat: pro-Agent-Konfiguration + 24h Duplikat-Suppressions. (#980) — Danke an @voidserf. - Security: Audit-warnt vor schwachen Modell-Tiers; App-Nodes speichern Auth-Tokens verschlüsselt (Keychain/SecurePrefs). ### Breaking - **BREAKING:** iOS-Minimumversion ist jetzt 18.0, um native Chat-Textual-Markdown-Darstellung zu unterstützen. (#702) - **BREAKING:** Microsoft Teams ist jetzt ein Plugin; installiere `@clawdbot/msteams` über `clawdbot plugins install @clawdbot/msteams`. ### Changes - CLI: Prozess-Titel auf `clawdbot-<command>` setzen für klarere Prozesslisten. - CLI/macOS: entfernte SSH-Targets/Identitäten mit der config synchronisieren und `gateway status` SSH-Targets automatisch ableiten lassen (ssh-config-aware). - Heartbeat: Prompt-Anleitung straffen + Duplikat-Warnungen für 24h unterdrücken. (#980) — Danke an @voidserf. - Repo: lokale Identity-Dateien ignorieren, um versehentliches Committen zu vermeiden. (#1001) — Danke an @gerardward2007. - Sessions/Security: `session.dmScope` für Multi-User-DM-Isolation hinzufügen und Audit-Warnungen. (#948) — Danke an @Alphonse-arianee. - Plugins: Provider-Auth-Registry hinzufügen + `clawdbot models auth login` für plugin-gesteuerte OAuth/API-Key-Flows. - Onboarding: Kanal-Einrichtung auf eine Single-Select-Schleife umgestellt mit per-Kanal-Aktionen und disable-Hints im Picker. - TUI: Provider-/Model-Labels für die aktive Sitzung und das Standardmodell anzeigen. - Heartbeat: pro-Agenten-Heartbeat-Konfiguration und Multi-Agenten-Dokumentations-Beispiel hinzufügen. - UI: Anleitungs-Beistellung für Gateway-Auth + Doc-Link bei unautorisierten Control-UI-Verbindungen anzeigen. - Security: Warnung vor schwachen Modell-Tiers (Haiku, unter GPT-5, unter Claude 4.5) in `clawdbot security audit`. - Apps: Node-Auth-Tokens verschlüsselt speichern (Keychain/SecurePrefs). - Daemon: Freigabe von Profil-/State-Dir-Auflösung über Service-Helfer und Berücksichtigung von `CLAWDBOT_STATE_DIR` für Windows-Task-Skripte. - Docs: Klarstellung zu Multi-Gateway Rescue-Bot-Anleitungen. (#969) — Danke an @bjesuiter. - Agents: Current Date & Time System-Prompt-Sektion mit konfigurierbarem Zeitformat (auto/12/24). - Tools: Normalisieren von Slack/Discord-Nachrichten-Timestamps mit `timestampMs`/`timestampUtc` bei Erhaltung roher Provider-Felder. - macOS: `system.which` für prompt-freie Remote-Skill-Erkennung hinzufügen (mit Gateway-Fallback zu `system.run`). - Docs: Datums- & Zeitleitfaden hinzufügen und Prompt-/Timezone-Konfigurationsdokumentation aktualisieren. - Messages: Debounce schneller eingehender Nachrichten über Kanäle mit per-Connector-Overrides. (#971) — Danke an @juanpablodlc. - Messages: Media-only-Sendungen (CLI/Tool) zulassen und Telegram-Sprachnachrichten-Status für Sprachnotizen anzeigen. (#957) — Danke an @rdev. - Auth/Status: Auth-Profile pro Sitzung sticky halten (Rotieren bei Kompaktierung/Neu), Provider-Nutzungs-Header in `/status` und `clawdbot models status` anzeigen und Docs aktualisieren. - CLI: `--json`-Ausgabe für `clawdbot daemon` Lifecycle/Installationsbefehle hinzufügen. - Memory: `node-llama-cpp` zu einer optionalen Abhängigkeit machen (Installationsfehler bei Node 25 vermeiden) und lokale-Embeddings-Fallback/Fehler verbessern. - Browser: `snapshot refs=aria` (Playwright aria-ref-IDs) für selbstauflösende Referenzen über `snapshot` → `act` hinzufügen. - Browser: `profile="chrome"` dient jetzt standardmäßig der Host-Steuerung und gibt klarere “Tab anhängen”-Fehler zurück. - Browser: stabilen Chrome für Auto-Erkennung bevorzugen, Brave/Edge-Fallbacks und aktualisierte Docs. (#983) — Danke an @cpojer. - Browser: Remote-CDP-Erreichbarkeits-Timeouts erhöhen + `remoteCdpTimeoutMs`/`remoteCdpHandshakeTimeoutMs` hinzufügen. - Browser: Auth-/Abfrage-Tokens für Remote-CDP-Endpunkte bewahren und Basic-Auth für CDP HTTP/WS übergeben. (#895) — Danke an @mukhtharcm. - Telegram: bidirektionale Reaktionsunterstützung mit konfigurierbaren Benachrichtigungen und Agentenführung. (#964) — Danke an @bohdanpodvirnyi. - Telegram: benutzerdefinierte Befehle im Bot-Menü zulassen (mit Native verschmolzen; Konflikte ignoriert). (#860) — Danke an @nachoiacovino. - Discord: whitelist-Guilds ohne Channel-Listen empfangen Nachrichten erlauben, wenn `groupPolicy="allowlist"`. — Danke an @thewilloftheshadow. - Discord: Emoji/Sticker-Uploads + Channel-Aktionen in Config-Standards zulassen. (#870) — Danke an @JDIVE. ### Fixes - Fix: Modelauswahl-Entries als Provider/Model-Paare anzeigen für klare Auswahl. (#970) — Danke an @mcinteerj. - Fix: OpenAI Image-Gen-Standards an DALL-E 3 Standardqualität angleichen und Ausgabeformate dokumentieren. (#880) — Danke an @mkbehr. - Fix: Persistieren von `gateway.mode=local` nach Auswahl des Local-Run-Modus in `clawdbot configure`, auch wenn keine anderen Abschnitte gewählt werden. - Daemon: fehlertolerante Profil-abhängige Service-Label-Auflösung (umweltgesteuert) korrigieren und Abdeckung für launchd/systemd/schtasks hinzufügen. (#969) — Danke an @bjesuiter. - Agents: false Positives vermeiden bei Logging-Schema-Schlüsselwörter der Google-Tools. - Agents: Gemini-Historie-Niederrangstellungen für google-antigravity überspringen, um Tool-Aufrufe zu bewahren. (#894) — Danke an @mukhtharcm. - Status: Nutzungszusammenfassungszeile für den aktuellen Provider wiederherstellen, wenn keine OAuth-Profile existieren. - Fix: Modell-Fallback gegen undefinierte Provider/Model-Werte absichern. (#954) — Danke an @roshanasingh4. - Fix: Sitzungs-Store-Updates refactoren, chat.inject hinzufügen, Subagent-Cleanup-Flow härten. (#944) — Danke an @tyler6204. - Fix: suspendierte CLI-Prozesse über Backends bereinigen. (#978) — Danke an @Nachx639. - Fix: MiniMax-Kodierungsplan-Antworten mit `model_remains`/`current_interval_*` Payloads unterstützen. - Fix: WhatsApp-Pairing-Antworten für historische Catch-up DMs beim ersten Link unterdrücken. (#904) - Browser: Extensionsmodus wiederhergestellt, wenn nur ein Tab verbunden ist (veralteter targetId-Fallback). - Browser: `tab not found` für Extension-Relay-Snapshots/Aktionen beheben, wenn Playwright `newCDPSession` blockiert (verwende die einzige verfügbare Page). - Browser: `ws` auf `wss` aktualisieren, wenn remote CDP `https` verwendet (Behebung des Browserless-Handshakes). - Telegram: `message_thread_id=1` für General-Themen-Sends überspringen, während Typing-Indikatoren beibehalten werden. (#848) — Danke an @azade-c. - Fix: benutzerorientierte Fehlermeldung bereinigen + `<final>`-Tags über Reply-Pipelines hinweg entfernen. (#975) — Danke an @ThomsenDrake. - Fix: Pairing-CLI-Aliase normalisieren, Extensions-Kanäle erlauben und Zalo-Webhook-Payload-Parsing härten. (#991) — Danke an @longmaba. - Fix: lokale Tailscale-Serve-Hostnamen zulassen, ohne Tailnet-Clients als Direkt-Verbindung zu behandeln. (#885) — Danke an @oswalpalash. - Fix: Sitzungen nach Konflikten bei Rollen-Reihenfolge zurücksetzen, um auf aufeinanderfolgende Benutzerzüge zu reagieren. (#998)

Claude Codev2.1.9

v2.1.9

16. Januar 2026

## What's changed - Added `auto:N` syntax for configuring the MCP tool search auto-enable threshold, where N is the context window percentage (0-100) - Added `plansDirectory` setting to customize where plan files are stored - Added external editor support (Ctrl+G) in AskUserQuestion "Other" input field - Added session URL attribution to commits and PRs created from web sessions - Added support for `PreToolUse` hooks to return `additionalContext` to the model - Added `${CLAUDE_SESSION_ID}` string substitution for skills to access the current session ID - Fixed long sessions with parallel tool calls failing with an API error about orphan tool_result blocks - Fixed MCP server reconnection hanging when cached connection promise never resolves - Fixed Ctrl+Z suspend not working in terminals using Kitty keyboard protocol (Ghostty, iTerm2, kitty, WezTerm)

Codex CLIvrust-v0.86.0

0.86.0

16. Januar 2026

## New Features - Skill metadata can now be defined in `SKILL.toml` (names, descriptions, icons, brand color, default prompt) and surfaced in the app server and TUI (#9125) - Clients can explicitly disable web search and signal eligibility via a header to align with server-side rollout controls (#9249) ## Bug Fixes - Accepting an MCP elicitation now sends an empty JSON payload instead of null to satisfy servers expecting content (#9196) - Input prompt placeholder styling is back to non-italic to avoid terminal rendering issues (#9307) - Empty paste events no longer trigger clipboard image reads (#9318) - Unified exec cleans up background processes to prevent late End events after listeners stop (#9304) ## Chores - Refresh the orchestrator prompt to improve internal routing behavior (#9301) - Reduce noisy `needs_follow_up` error logging (#9272) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.85.0...rust-v0.86.0 - #9301 chore: better orchestrator prompt @jif-oai - #9304 nit: clean unified exec background processes @jif-oai - #9307 Revert recent styling change for input prompt placeholder text @etraut-openai - #9125 Support SKILL.toml file. @xl-openai - #9249 [search] allow explicitly disabling web search @sayan-oai - #9272 remove needs_follow_up error log @pap-openai - #9318 Revert empty paste image handling @aibrahim-oai - #9196 fix: send non-null content on elicitation Accept @yuvrajangadsingh

Windsurfv1.13.9

GPT-5.2-Codex-Verbesserungen

16. Januar 2026

Verbesserungen am GPT-5.2-Codex Harness und Windows Group Policy-Unterstützung.

Continue.devv1.5.34

CLI Stable Release v1.5.34

15. Januar 2026

Direktes stabiles Release vom Hauptzweig. Version 1.5.34 wird direkt aus dem neuesten Hauptzweig gebaut und veröffentlicht.

Continue.devv1.5.33

CLI Stable Release v1.5.33

15. Januar 2026

Direktes stabiles Release vom Hauptzweig. Version 1.5.33 wurde direkt aus dem neuesten Hauptzweig gebaut und veröffentlicht.

OpenCodev1.1.23

v1.1.23

15. Januar 2026

## Core - Ensure agent exists before processing title in session summary (@0x404) - More defensive handling of project list to prevent errors - Filter dead worktrees from git worktree list - Load models.dev in development mode - Ensure frontmatter can process same content as other agents - Show toast error message when ConfigMarkdown parsing fails (@vvarp) - Prettify retry duration display in TUI - Handle broken symlinks gracefully in grep tool (@andrew-kramer-inno) - Honor per-server MCP timeouts - Upgrade OpenTUI to v0.1.73, fixing CJK word wrapping and Thai text rendering in non-tmux environments - Add input limit for compaction (@brandon93s) - Fix Codex ID issue - Revert provider-level store option - Allow provider-level store option (@cyhhao) - Make dialog keybinds configurable (@CasualDeveloper) ## TUI - Update black plan selection with light rays effect and mobile styles (@aaroniker) - Use native text truncation for sidebar diff paths - Center the initially selected session in the session list (@ariane-emory) - Restore brand integrity of TUI wordmark (@kitlangton) ## Desktop - Fix sidebar toggle on desktop - Persist workspace branch selection across sessions - Persist workspace order and collapsed state in sidebar - Fix thinking animation opacity and improve design - Fix session icon and name alignment - Show session as busy even when it's the active session - Stop killing opencode-cli process on dev mode restart (@dbpolito) - Update desktop question tools (@dbpolito) - Fix current session background color - Remove the active state from load more button after press - Fix new session tooltip position and add keyboard shortcut - Remove more options tooltip - Fix workspace name color in UI - Hide "View all sessions" button when on active project page - Fix avatar button states - Reduce font size of load more button - Fix recent sessions title color - Fix project dropdown labels and sorting order - Move left panel toggle button to the right - Fix avatar background color - Show project options on hover of row - Increase delay on session list tooltips - Fix tooltip gutter spacing on session items and archive buttons - Archive icon replaces diff count on hover - Fix hover state for recent sessions in gutter - Fix search bar functionality in header - Fixed layout issues in the app - Fix stale sessions in new layout - New app layout - Open external links in system browser instead of webview in desktop app (@turculaurentiu91) **Thank you to 15 community contributors:** - @CasualDeveloper: - feat(tui): make dialog keybinds configurable (#6143) (#6144) - @kitlangton: - fix: restore brand integrity of TUI wordmark (#8584) - @cyhhao: - feat: allow provider-level store option (#8000) - @ariane-emory: - fix(tui): Center the initially selected session in the session_list (resolves #8558) (#8560) - @brandon93s: - fix(opencode): add input limit for compaction (#8465) - @R44VC0RP: - fix: enable sticky header on changelog and download pages (#8556) - @turculaurentiu91: - fix(desktop): open external links in system browser instead of webview (#7360) - @LAG-4: - docs: fix typos in documentation (#8703) - @andrew-kramer-inno: - fix: handle broken symlinks gracefully in grep tool (#8612) - @vvarp: - fix: show toast error message on ConfigMarkdown parse error (#8049) - @dbpolito: - chore(desktop): Question Tools Updates - chore(desktop): Stop Killing opencode-cli on dev - @aaroniker: - feat(console): Update /black plan selection, light rays effect. mobile styles (#8731) - feat(console): Style improvements for /black, View Transition fixes (#8739) - @nathannewyen: - docs: add /thinking command documentation (#8722) - @0x404: - fix(session): ensure agent exists before processing title in session summary (#8662) - @ricardo-valero: - fix: update hix hashes for all systems (#8732)

Codex CLIvrust-v0.85.0

0.85.0

15. Januar 2026

## New Features - App-server v2 now emits collaboration tool calls as item events in the turn stream, so clients can render agent coordination in real time. (#9213) - Collaboration tools gained richer agent control: `spawn_agent` accepts an agent role preset, and `send_input` can optionally interrupt a running agent before delivering the message. (#9275, #9276) - `/models` metadata now includes upgrade migration markdown so clients can display richer guidance when suggesting model upgrades. (#9219) ## Bug Fixes - [revert] Linux sandboxing now falls back to Landlock-only restrictions when user namespaces are unavailable, and sets `no_new_privs` before applying sandbox rules. (#9250) - `codex resume --last` now respects the current working directory, with `--all` as an explicit override. (#9245) - Stdin prompt decoding now handles BOMs/UTF-16 and provides clearer errors for invalid encodings. (#9151) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.84.0...rust-v0.85.0 - #9219 Add `migration_markdown` in `model_info` @aibrahim-oai - #9250 fix: fallback to Landlock-only when user namespaces unavailable and set PR_SET_NO_NEW_PRIVS early @viyatb-oai - #9213 feat: collab tools app-server event mapping @jif-oai - #9275 feat: add agent roles to collab tools @jif-oai - #9276 feat: add interrupt capabilities to `send_input` @jif-oai - #9209 feat: basic tui for event emission @jif-oai - #9245 Changed `codex resume --last` to honor the current cwd @etraut-openai - #9207 Propagate MCP disabled reason @gt-oai - #9151 fix(exec): improve stdin prompt decoding @liqiongyu - #9300 revert: remove pre-Landlock bind mounts apply @viyatb-oai

OpenClawv2026.1.14-1

clawdbot 2026.1.14-1

15. Januar 2026

### Install - npm i -g clawdbot@2026.1.14-1 ### Highlights - Web-Suche: `web_search`/`web_fetch` Tools (Brave API) + Ersteinrichtung im Onboarding/Setup. - Browser-Steuerung: Chrome-Erweiterungs-Relay-Übernahme-Modus + Remote-Browser-Steuerung über `clawdbot browser serve`. - Plugins: Kanal-Plugins (Gateway HTTP-Hooks) + Zalo-Plugin + Onboarding-Installations-Flow. (#854) — thanks @longmaba. - Sicherheit: erweitertes `clawdbot security audit` (+ `--fix`), detect-secrets CI-Scan und eine `SECURITY.md` Berichts-Richtlinie. ### Changes #### Web Tools - Tools: hinzufügen `web_search`/`web_fetch` (Brave API), einschließlich hilfreicher Setup-Hinweise, wenn der Schlüssel fehlt. - Tools: Standardmäßiges Aktivieren von `web_fetch` (sofern nicht explizit in der Konfiguration deaktiviert). - CLI/Docs: hinzufügen `clawdbot configure --section web` zum Speichern Brave-API-Schlüssel und Onboarding-Tipps aktualisieren. #### Browser / Control UI - Browser: Chrome-Erweiterungs-Relay-Übernahme-Modus (Symbolleiste) + `clawdbot browser serve` Remote-Steuerung + `browser.controlToken`. - Browser: integriertes `chrome`-Profil für Extension-Relay liefern und das Relay beim lokalen Start automatisch starten. - Browser: Standardwert `browser.defaultProfile` auf `chrome` setzen (vorhandener Chrome-Übernahme-Modus). - Browser: hinzufügen `clawdbot browser extension install/path` und Erweiterungs-Pfad in die Zwischenablage kopieren. - Control UI: rohe Any-Map-Einträge in den Config-Ansichten anzeigen; Docs-Link in die linke Navigation verschieben. #### Plugins - Plugins: hinzufügen HTTP-Hooks von Plugins + Loader-Updates zur Unterstützung von Channel-Plugins. (#854) — thanks @longmaba. - Plugins: onboarding Plugin-Installations-Flow hinzufügen. (#854) — thanks @longmaba. - Channels: Matrix-Plugin (extern) hinzufügen mit Docs + Onboarding-Hooks. - Voice Call: Plivo-Anbieter hinzufügen (kein SDK-Abhängigkeit). (#846) — thanks @vrknetha. #### Security - Security: Prüfung von `clawdbot security audit` erweitern und Richtlinie in `SECURITY.md` veröffentlichen. - Security: erweitern `clawdbot security audit --fix`, um sensiblere Pfade stärker abzusichern. - Security: detect-secrets CI-Scan und Baseline-Anleitungen hinzufügen. (#227) — thanks @Hyaxia. #### Onboarding / Daemon - Onboarding: Sicherheits-Checkpunkt-Abfrage hinzufügen (Docs-Link + Sandboxing-Hinweis); für `--non-interactive` `--accept-risk` erforderlich. - Daemon: unterstütze profilabhängige Servicenamen für Multi-Gateway-Setups. (#671) — thanks @bjesuiter. #### Auth / Usage / Config - Usage: MiniMax-Codierungsplan-Nutzungs-Tracking hinzufügen. - Auth: Claude Code CLI Auth-Optionen kennzeichnen. (#915) — thanks @SeanZoR. - Agents: optionale Auth-Profile-Kopie-Eingabe bei `agents add` hinzufügen und Auth-Fehlermeldungen verbessern. - Auth: dynamische Template-Variablen zu `messages.responsePrefix` hinzufügen. (#928) — thanks @sebslight. - Config: hinzufügen `channels.<provider>.configWrites` Gatekeeper für channel-initiierte Config-Schreibvorgänge; Slack-Channel-IDs migrieren. #### Channels - Telegram: Aktion zum Löschen von Nachrichten im Nachricht-Tool hinzufügen. (#903) — thanks @sleontenko. - WhatsApp: `channels.whatsapp.sendReadReceipts` hinzufügen, um automatische Lesebestätigungen zu deaktivieren. (#882) — thanks @chrisrodz. #### Docs - Docs: Klarstellung zu pro-Agent Auth-Stores, sandboxed Skill-Binaries und elevierten Semantiken. - Docs: FAQ-Einträge zu fehlendem Provider-Auth nach dem Hinzufügen von Agents und Gemini Thinking Signatur-Fehlern. - Docs: Erweiterte Gateway-Sicherheits-Härtungshinweise und Incident-Response-Checkliste. - Docs: DM-Historie-Limits für Channel-DMs dokumentieren. (#883) — thanks @pkrmf. - Docs: Claude Code CLI-Namenskonventionen in allen Docs und Prompts standardisieren. (Fortsetzung zu #915) - Docs: pro-Befehl CLI-Dokumentationsseiten hinzufügen und von `clawdbot <command> --help` verlinken. - Docs: Multi-Gateway-Guide hinzufügen (Sidebar + Navigation). ### Fixes #### Gateway / Daemon / Sessions - Gateway: Terminationssignale an neu gestartete CLI-Unterprozesse weiterleiten, um verwaiste systemd-Runs zu vermeiden. (#933) — thanks @roshanasingh4. - Gateway/UI: Sitzungs-Defaults im Hello-Snapshot liefern, sodass die Control-UI Haupt-Sitzungsschlüssel canonicalisiert (kein bloßes `main`-Alias). - Agents: Denk- und Final-Tag-Stripping innerhalb von Markdown-Code-Spans überspringen. (#939) — thanks @ngutman. - Browser: Tests für Snapshot-Labels/efficient Query-Params und gelabelte Bildantworten hinzufügen. - Browser: Rollen-Snapshot-Refs pro CDP-Target persistieren, sodass `snapshot` → `act` Klicks auch dann funktionieren, wenn Playwright eine andere Page-Instanz zurückgibt. - macOS: sicherstellen, dass das Launchd-Protokollverzeichnis existiert (Test-Override). (#909) — thanks @roshanasingh4. - macOS: ConnectionsStore-Konfig formatieren, um SwiftFormat-Lint zu erfüllen. (#852) — thanks @mneves75. - Packaging: `pnpm build` bei `prepack` ausführen, damit npm Publikationen frische `dist/`-Ausgaben enthalten. - Telegram: native Befehls-Command mit Unterstrichen registrieren, um `BOT_COMMAND_INVALID` zu vermeiden (#929, fixes #901) — thanks @grp06. - Google: unsignierte Thinking-Blöcke vor dem Senden verringern, um fehlende Signaturfehler zu vermeiden. - Agents: Benutzerzeitzone und 24-Stunden-Zeit explizit im Systemprompt festlegen. (#859) — thanks @CashWilliams. - Agents: Text des downgrade Tools entfernen, Adjacent-Replies nicht verschlingen und Thinking-Tag-Lecks filtern. (#905) — thanks @erikpr1994. - Agents: Tool-Call-IDs für OpenAI/OpenRouter begrenzen, um Anforderungsablehnungen zu vermeiden. (#875) — thanks @j1philli. - Doctor: vermeiden, WhatsApp-Konfiguration erneut hinzuzufügen, wenn nur Legacy-Ack-Reaktionen gesetzt sind. (#927, fixes #900) — thanks @grp06. - Agents: Tupel-Items-Schemas für Gemini-Tool-Aufrufe säubern. (#926, fixes #746) — thanks @grp06. - Agents: Sub-Agent-Announce-Status aus Laufzeitergebnissen stabilisieren und Result/Notes normalisieren. (#835) — thanks @roshanasingh4. - Apps: canonical main Session-Keys aus Gateway-Defaults MAC-/iOS-/Android-weite verwenden, um bloße `main`-Sitzungen zu vermeiden. - Embedded runner: rohe API-Fehlerpayloads in Antworten unterdrücken. (#924) — thanks @grp06. - Auth: Claude Code CLI-Profilmodus auf OAuth normalisieren und Config automatisch migrieren. (#855) — thanks @sebslight. - Daemon: persistierte Launchd-Disabled-State vor Bootstrap löschen (fixes `daemon install` nach Deinstallation). (#849) — thanks @ndraiman. - Sessions: tiefe Klone (`structuredClone`) zurückgeben, damit zwischengespeicherte Sitzungseinträge nicht mutiert werden können. (#934) — thanks @ronak-guliani. - Heartbeat: `updatedAt` monoton halten beim Restore von Heartbeat-Sessions. (#934) — thanks @ronak-guliani. - Agent: Ausführungskontext nach CLI-Lauf löschen (`clearAgentRunContext`), um entlaufende Kontexte zu vermeiden. (#934) — thanks @ronak-guliani. - Gateway/Dev: sicherstellen, dass `pnpm gateway:dev` immer das Dev-Profil-Config + State verwendet (`~/.clawdbot-dev`). #### CLI / Onboarding - Onboarding: Web-Suche-Setup am Ende anzeigen (nicht am Anfang). - Onboarding: Anzeige von Daemon-Installations-/Neu-starting-Fortschritt (kein “Blinken des Cursors”) und Ausgabe-Formatierung der Daemon-Installation korrigieren. - Health: Farben für “nicht konfiguriert” Provider-Zeilen zum leichteren Scannen. #### Control UI / TUI - Control UI: Cron-Lauf-Histor

Continue.devv@continuedev/config-yaml@1.39.0

@continuedev/config-yaml@1.39.0

15. Januar 2026

# [1.39.0](https://github.com/continuedev/continue/compare/@continuedev/config-yaml@1.38.0...@continuedev/config-yaml@1.39.0) (2026-01-15) ### Bug Fixes * add GH_TOKEN to continue-agents workflow ([dd93919](https://github.com/continuedev/continue/commit/dd939190701be55917d4d224f012017e8ea17be9)), schließt [#9493](https://github.com/continuedev/continue/issues/9493) * break circular dependency by extracting tool names ([6a5f80b](https://github.com/continuedev/continue/commit/6a5f80b0fc739ed36448b62bd036c01476b91d49)) * **cli:** Kompaktierung für fehlende Tool-Ergebnisse ([0792b94](https://github.com/continuedev/continue/commit/0792b949bdf629fd14dac45478b2027962b77454)) * Erkennung von Längenfehlern im Kontext ([1bfb72e](https://github.com/continuedev/continue/commit/1bfb72edf3f6c0eccb66454eb9e3ad423f58bee5)) * Korrektur der Liste der integrierten Tool-Namen ([26bef77](https://github.com/continuedev/continue/commit/26bef77a8859788b870e42036e0958bdbaa9d93a)) * Korrektur der Import-Reihenfolge in ToolPermissionService ([2003128](https://github.com/continuedev/continue/commit/2003128f084517bdf5343d6e2205edc85db8ff64)) * URI-entschlüsselten Pfadnamen für WSL-Arbeitsbereichspfadnamen dekodieren ([074a6c7](https://github.com/continuedev/continue/commit/074a6c74baf35b98e0092f0e6b37027bb6561eba)) * YAML.schemas-Einstellungen zusammenführen ([a0729bd](https://github.com/continuedev/continue/commit/a0729bd6ec7f5a7b8cdf187baac8abf3156b9f75)), schließt [#7080](https://github.com/continuedev/continue/issues/7080) * WSL2-Arbeitsbereichspfadnamen korrekt auflösen in runTerminalCommand ([eab2ac1](https://github.com/continuedev/continue/commit/eab2ac1e045f2e4ba994cb7371b8e3a8e8aaf4b7)) * kreisförmige Abhängigkeit im Subagent-Executor auflösen ([87fd9b6](https://github.com/continuedev/continue/commit/87fd9b67df4661a261881a2cb028ab651c695128)) * kreisförmige Abhängigkeit im Subagent-Tool auflösen ([1c448c4](https://github.com/continuedev/continue/commit/1c448c479d1e2f4c50c586d494288d6b8889c178)) * upgrade @sentry/profiling-node von 9.46.0 auf 9.47.1 ([9448930](https://github.com/continuedev/continue/commit/9448930f6094cf46dff5001664c9033312072edc)) * upgrade lru-cache von 11.0.2 auf 11.2.4 ([269be8b](https://github.com/continuedev/continue/commit/269be8bb5fdb4651a2e52f832e92f5dc8b250b55)) * upgrade monaco-editor von 0.45.0 auf 0.55.1 ([cc1c70d](https://github.com/continuedev/continue/commit/cc1c70d246b124a592cf94b17cb1fe7ba8fa2bc2)) * Verwende try/finally für Env-Cleanup in Tests ([27adb1d](https://github.com/continuedev/continue/commit/27adb1d677653904c8a2efad1fc1065deb7993e7)) * wrap fileURLToPath in try-catch für WSL2-Kompatibilität ([788e253](https://github.com/continuedev/continue/commit/788e25338a1f21fe0b15db92397cb7fc1c236c6a)), schließt [#8091](https://github.com/continuedev/continue/issues/8091) ### Features * Subagents ([32af25f](https://github.com/continuedev/continue/commit/32af25fc9980c505c16889b47e78feb7feafbc95))

OpenCodev1.1.21

v1.1.21

15. Januar 2026

## Core - Add error message to prompt users to reauthenticate with Copilot - Official Copilot plugin - Adjust LSP wording to better encourage users to fix issues - Fix plan path permissions - Add check in case provider doesn't exist in models list - Adjust GitLab logic in provider configuration - Add AWS Web Identity Token File support for Bedrock (@bainos) - Add Chocolatey and Scoop to OpenCode upgrade methods (@itsrainingmani) - Remove duplicate path import in session module (@CasualDeveloper) ## TUI - Improve question prompt UX in TUI (@kitlangton) **Thank you to 8 community contributors:** - @kitlangton: - feat(tui): improve question prompt UX (#8339) - @R44VC0RP: - fix(homepage): Update FAQ to include desktop and web links (#8453) - docs: update screenshot images (#8479) - feat(console): add /changelog page (#8476) - docs: add Web usage page (#8482) - fix(console): make logo on /black link back to homepage (#8498) - feat(console): add OG image and SEO meta tags for /black page (#8506) - @CasualDeveloper: - fix(session): remove typo'd duplicate path import (#8408) (#8412) - @itsrainingmani: - feat: add choco and scoop to opencode upgrade methods (#8439) - @aaroniker: - feat(console/app): Style changes, view transitions, small improvements (#8481) - @bainos: - feat(opencode): add AWS Web Identity Token File support for Bedrock (#8461) - @samiulsami: - feat(install): respect ZDOTDIR for zsh config detection (#8511) - @IdrisGit: - docs: fix permission system documentation in agents section (#7652)

Codex CLIvrust-v0.84.0

0.84.0

15. Januar 2026

## New Features - Extend the Rust protocol/types to include additional metadata on text elements, enabling richer client rendering and schema evolution (#9235) ## Chores - Reduce flaky Rust release pipelines (notably on Windows) by increasing the release build job timeout (#9242) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.83.0...rust-v0.84.0 - #9242 fix: increase timeout for release builds from 30 to 60 minutes @bolinfest - #9235 Add text element metadata to types @charley-oai

Continue.devv1.2.14-vscode

v1.2.14-vscode

14. Januar 2026

## Änderungen * docs: YT zur Fußzeile hinzufügen von @BekahHW in https://github.com/continuedev/continue/pull/8566 * chore(deps): bump actions/upload-artifact von 4 auf 5 von @dependabot[bot] in https://github.com/continuedev/continue/pull/8459 * Add description truncation for slash commands in CLI UI von @sestinj in https://github.com/continuedev/continue/pull/8172 * docs: Configs-Sidebar-Icon von Robot auf Buch ändern von @continue[bot] in https://github.com/continuedev/continue/pull/8543 * docs: YouTube-Video in Understanding-configs-Guide einbetten von @continue[bot] in https://github.com/continuedev/continue/pull/8425 * feat: grok-Code-Fast-1 upgraden von @RomneyDa in https://github.com/continuedev/continue/pull/8475 * Throw error and exit on MCP failure in headless mode von @continue[bot] in https://github.com/continuedev/continue/pull/8559 * update docs header text von @TyDunn in https://github.com/continuedev/continue/pull/8570 * feat(cli): Update-Updates-Scheck in Dev-Modus überspringen von @uinstinct in https://github.com/continuedev/continue/pull/8564 * Add snyk-bot zur CLA-Whitelist von @sestinj in https://github.com/continuedev/continue/pull/8571 * [Snyk] Upgrade @continuedev/config-yaml von 1.14.0 auf 1.26.0 von @sestinj in https://github.com/continuedev/continue/pull/8505 * fix: CLI-Kontext-Verschmutzung zwischen Sitzungen von @continue[bot] in https://github.com/continuedev/continue/pull/8420 * feat(cli): immer Kontext-Prozentsatz anzeigen von @uinstinct in https://github.com/continuedev/continue/pull/8563 * docs: README-Agenten-Update von @TyDunn in https://github.com/continuedev/continue/pull/8572 * docs: JetBrains README für Agents aktualisieren von @TyDunn in https://github.com/continuedev/continue/pull/8573 * docs: VS Code README für Agents aktualisieren von @TyDunn in https://github.com/continuedev/continue/pull/8574 * fix: Sofortige Ablehnung aller VS Code durch @RomneyDa in https://github.com/continuedev/continue/pull/8489 * feat: MCP-Warnung mit konfigurierten MCPs anzeigen von @uinstinct in https://github.com/continuedev/continue/pull/8478 * docs: Entferne Beta-Warnung zur CLI-Einführung von @uinstinct in https://github.com/continuedev/continue/pull/8463 * docs: Umfassende Agenten-Dokumentation hinzufügen von @BekahHW in https://github.com/continuedev/continue/pull/8509 * docs: Cookbook-Anleitungen aktualisieren, um Agents zu verwenden von @BekahHW in https://github.com/continuedev/continue/pull/8486 * Add script to analyze file hotspots from git history von @sestinj in https://github.com/continuedev/continue/pull/8591 * Add x-continue-unique-id-Header zu LLM-Anfragen und Header-Groß-/Kleinschreibung normalisieren von @sestinj in https://github.com/continuedev/continue/pull/8589 * chore: Version auf 1.3.25 erhöhen von @sestinj in https://github.com/continuedev/continue/pull/8592 * perf(intellij): GsonService-Singleton implementieren zur Reduzierung von Speicherallokationen von @houssemzaier in https://github.com/continuedev/continue/pull/8524 * Alle Umgebungsvariablen an MCP-Server weiterreichen statt nur PATH von @sestinj in https://github.com/continuedev/continue/pull/8602 * docs: Verbesserte Dokumentations-Schreibanleitung mit Video und klareren Überschriften von @continue[bot] in https://github.com/continuedev/continue/pull/8424 * delete Source-Control-Seite aus Docs von @TyDunn in https://github.com/continuedev/continue/pull/8605 * [Snyk] Upgrade @c15t/react von 1.4.4 auf 1.6.1 von @sestinj in https://github.com/continuedev/continue/pull/8580 * fix: Typprüfung Schreib-Datei-Argumente von @RomneyDa in https://github.com/continuedev/continue/pull/8585 * fix: Apply-Prompt so ändern, dass Kommentare nicht entfernt werden von @uinstinct in https://github.com/continuedev/continue/pull/8608 * chore: continue.telemetryEnabled in VSCode-Einstellungen veralten lassen von @uinstinct in https://github.com/continuedev/continue/pull/8497 * Remove beta-Text aus CLI-Einführung von @continue[bot] in https://github.com/continuedev/continue/pull/8610 * refactor: ide.readFile statt fs.readFile im Static Context Service verwenden von @uinstinct in https://github.com/continuedev/continue/pull/8422 * chore: Hub-Links aktualisieren von @Patrick-Erichsen in https://github.com/continuedev/continue/pull/8621 * [Snyk] Upgrade zod von 3.24.2 auf 3.25.76 von @sestinj in https://github.com/continuedev/continue/pull/8579 * Update macOS-Runner von macos-13 auf macos-latest in CI-Workflows von @sestinj in https://github.com/continuedev/continue/pull/8593 * fix(cli): benötigte Parameter beim Tool-Aufruf validieren von @uinstinct in https://github.com/continuedev/continue/pull/8598 * feat: entfernte angewandte Regeln aus Chat-Verlauf anzeigen von @RomneyDa in https://github.com/continuedev/continue/pull/8594 * fix: useSlashCommand-Telemetrie von @RomneyDa in https://github.com/continuedev/continue/pull/8584 * fix: Headless-Ladefehler beheben von @sestinj in https://github.com/continuedev/continue/pull/8633 * fix: prepended fim_prefix-Tag in Mercury-Coder entfernen von @uinstinct in https://github.com/continuedev/continue/pull/8624 * Bump-Version von 1.3.25 auf 1.3.26 von @sestinj in https://github.com/continuedev/continue/pull/8635 * chore: Discord-Link durch GitHub-Discussions ersetzen von @uinstinct in https://github.com/continuedev/continue/pull/8667 * [Snyk] Upgrade @anthropic-ai/sdk von 0.62.0 auf 0.67.0 von @sestinj in https://github.com/continuedev/continue/pull/8641 * [Snyk] Upgrade @continuedev/config-yaml von 1.15.1 auf 1.29.0 von @sestinj in https://github.com/continuedev/continue/pull/8638 * [Snyk] Upgrade @c15t/react von 1.6.1 auf 1.7.0 von @sestinj in https://github.com/continuedev/continue/pull/8637 * fix: verhindern, dass sensible Dateien in nächste Edit-Diffs auftauchen von @uinstinct in https://github.com/continuedev/continue/pull/8658 * [Snyk] Upgrade @aws-sdk/credential-providers von 3.890.0 auf 3.913.0 von @sestinj in https://github.com/continuedev/continue/pull/8639 * fix: Befehlstitel von View History zu View Logs aktualisieren von @lixiaoyang1992 in https://github.com/continuedev/continue/pull/8455 * Fix: MCP-Fehlerausgabe und Bugfixes von @RomneyDa in https://github.com/continuedev/continue/pull/8655 * Add guidance to Bash tools to prefer edit tools over bash commands von @continue[bot] in https://github.com/continuedev/continue/pull/8674 * Slack-Integrations-Dokumentation von @continue[bot] in https://github.com/continuedev/continue/pull/8546 * [Snyk] Upgrade tar von 7.4.3 auf 7.5.1 von @sestinj in https://github.com/continuedev/continue/pull/8682 * Remove hub/source-control und customize/deep-dives/slash-commands aus

Continue.devv@continuedev/openai-adapters@1.37.0

@continuedev/openai-adapters@1.37.0

14. Januar 2026

# [1.37.0](https://github.com/continuedev/continue/compare/@continuedev/openai-adapters@1.36.0...@continuedev/openai-adapters@1.37.0) (2026-01-14) ### Bug Fixes * fehlende Hub-Mock-Verweigerung im Test hinzugefügt ([53a9fa4](https://github.com/continuedev/continue/commit/53a9fa4cb34d97a9f3e2af1f26a02b10bbd0a9a1)) * auch Umgebungsvariablen in Write-Job-Zusammenfassungs-Schritt verwenden ([8969f42](https://github.com/continuedev/continue/commit/8969f422619f63becf2f778c4465cf192ee83e18)) * 'Agent:' im Einleitungsnachrichten in 'Config:' ändern und Tests aktualisieren ([c1e4668](https://github.com/continuedev/continue/commit/c1e466836acd0b2a5e81f2925c5d2e14691543d9)) * klargestellen, dass paralleles Bearbeiten mit sich selbst ebenfalls nicht möglich ist ([46b31af](https://github.com/continuedev/continue/commit/46b31af330aa669d93c777d5fa577b2ca7c72225)) * **cli:** Kompaktion für fehlende Tool-Ergebnisse ([0792b94](https://github.com/continuedev/continue/commit/0792b949bdf629fd14dac45478b2027962b77454)) * **cli:** cli für manuelles Update neu starten ([595986a](https://github.com/continuedev/continue/commit/595986a6fb46ccd4e023078e9e3a8aeb911629c7)) * Erkennung von Längenfehlern im Kontext ([1bfb72e](https://github.com/continuedev/continue/commit/1bfb72edf3f6c0eccb66454eb9e3ad423f58bee5)) * Kontextlängen­fehlerbehebungen, Abschneiden, usw. ([613ffb8](https://github.com/continuedev/continue/commit/613ffb8ca9847fcb3ff99beec6f191b25cd85f9b)) * Fallback auf relativen Pfad vermeiden, wenn keine Markdown-Datei vorliegt ([d93b432](https://github.com/continuedev/continue/commit/d93b4321a550acf7151949d1fee1c436945e2ee0)) * Stub-Umgebungen für cli-Plattform-Tests ([6b3c0be](https://github.com/continuedev/continue/commit/6b3c0be2f1443e47db3e985560cb859b8e42f6a9)) * Lint-Fehler beheben und Tests für Markdown-Only-Fallback aktualisieren ([4c0c532](https://github.com/continuedev/continue/commit/4c0c53257baf7edb943c76155fa6b0bde79b554f)) * Eininjectierte Blocks-Tests zur Lösung von Secrets in CLI ([219f566](https://github.com/continuedev/continue/commit/219f5661c9999daa5e185740dd2fa1d24f0187b2)) * packages/continue-sdk/package.json & packages/continue-sdk/package-lock.json zur Reduzierung von Sicherheitslücken ([2f128be](https://github.com/continuedev/continue/commit/2f128beb2ce1c69097a5eb21e1a2d7834fc16d41)) * verhindern String-Interpolation-Probleme in verbleibenden Workflow-Schritten ([2d21472](https://github.com/continuedev/continue/commit/2d2147230c28fc8d91a0d4df8a27fbb88c6b0ed5)) * Prompt-Feinabstimmung ([41c4f29](https://github.com/continuedev/continue/commit/41c4f297f3bef3a43f12833d3d2e0d6c84888ec7)) * Hub-Fehler erneut werfen für Nicht-Markdown-Hub-Slugs ([f644c88](https://github.com/continuedev/continue/commit/f644c8873e09f8efa912eefeb1a03e5d8708f24f)) * Paket-Lock-Datei neu generieren, um fehlende @types/node@25.0.3 einzuschließen ([10b457b](https://github.com/continuedev/continue/commit/10b457b0eae361ed18f9128980f5f6bfb1c9ff1a)) * Symbolik-Logik aus Produktions-Blueprint-Vorlage entfernen ([a396ed2](https://github.com/continuedev/continue/commit/a396ed2acccfa523c64a45a94428c3698fdf9601)) * Symbolik-Logik aus Staging-Blueprint-Vorlage entfernen ([ce07084](https://github.com/continuedev/continue/commit/ce07084d2dcfddebf7e28b6193ecf660b84bfd09)) * neue Regel nach Erstellung anzeigen ([bf6ee2f](https://github.com/continuedev/continue/commit/bf6ee2f71da3008bf01bd5be030f5bff62410d0e)) * Tests und cubic-Feedback ([d7b91f8](https://github.com/continuedev/continue/commit/d7b91f884c2be22fb6805dbcc5a9ad738bb443ee)) * LRU-Cache von 11.0.2 auf 11.2.4 aktualisieren ([269be8b](https://github.com/continuedev/continue/commit/269be8bb5fdb4651a2e52f832e92f5dc8b250b55)) * Mocha von 11.7.1 auf 11.7.5 aktualisieren ([c65d6df](https://github.com/continuedev/continue/commit/c65d6dfb9ac9daac866fc55b62a55de6fd9a402c)) * Monaco-Editor von 0.45.0 auf 0.55.1 aktualisieren ([cc1c70d](https://github.com/continuedev/continue/commit/cc1c70d246b124a592cf94b17cb1fe7ba8fa2bc2)) * Umgebungsvariablen für sichere String-Verarbeitung in continue-agents-Workflow verwenden ([fe1a35a](https://github.com/continuedev/continue/commit/fe1a35a33e0a54ae32a2fc97991a26a21b53ea09)) * Proxy für ungerenderte eingefügte Block-Geheimnisse verwenden ([91ddbb8](https://github.com/continuedev/continue/commit/91ddbb8c2e6686f91595132c42fafc6c7518e930)) ### Features * Wiederverwendbaren Continue Agents-Workflow hinzufügen ([526b0b6](https://github.com/continuedev/continue/commit/526b0b6672abe22567a7f54ed8274e00db19824e)) * Staging-Blueprint (cn-staging) für isoliertes Testen hinzufügen ([3dc36ba](https://github.com/continuedev/continue/commit/3dc36ba00ea14ec939b34ed841c436fdca43c582)) * Auswahl von Bildern mit cmd+a zulassen ([e31605a](https://github.com/continuedev/continue/commit/e31605aaf06353dfbf34559595b4664b576fd5d6)) * **assets:** Xiaomi MiMo-Logo hinzufügen ([3f61af5](https://github.com/continuedev/continue/commit/3f61af520d43e949ef77821f56558cb3d15cc76b)) * CLI-Tool-Zusammenfassung und Dokumentation ([34537d9](https://github.com/continuedev/continue/commit/34537d926fca7017c748fe69c9bf888aee0d8b95)) * **neues Modell:** MiMo-V2-Flash hinzufügen ([a2d8443](https://github.com/continuedev/continue/commit/a2d8443a35b90cc909e888e2e4d503c1525ff9da)) * verfeinerte CLI-Bash-Tool-Abschneidung ([9a630a0](https://github.com/continuedev/continue/commit/9a630a025de85dd11069afb04a37f6295aca8029))

Codex CLIvrust-v0.81.0

0.81.0

14. Januar 2026

## New Features - Default API model moved to gpt-5.2-codex. (#9188) - The `codex` tool in `codex mcp-server` now includes the `threadId` in the response so it can be used with the `codex-reply` tool, fixing #3712. The documentation has been updated at https://developers.openai.com/codex/guides/agents-sdk/. (#9192) - Headless runs now switch to device-code login automatically so sign-in works without a browser. (#8756) - Linux sandbox can mount paths read-only to better protect files from writes. (#9112) - Support partial tool calls rendering in `tui` ## Bug Fixes - Alternate-screen handling now avoids breaking Zellij scrollback and adds a config/flag to control it. (#8555) - Windows correctly prompts before unsafe commands when running with a read-only sandbox policy. (#9117) - Config.toml and rules parsing errors are reported to app-server clients/TUI instead of failing silently. (#9182, #9011) - Worked around a macOS system-configuration crash in proxy discovery. (#8954) - Invalid user image uploads now surface an error instead of being silently replaced. (#9146) ## Documentation - Published a generated JSON Schema for `config.toml` in `docs/` to validate configs. (#8956) - Documented the TUI paste-burst state machine for terminals without reliable bracketed paste. (#9020) ## Chores - Added Bazel build support plus a `just bazel-codex` helper for contributors. (#8875, #9177) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.80.0...rust-v0.81.0 - #8756 [device-auth] When headless environment is detected, show device login flow instead. @mzeng-openai - #8930 feat: first pass on clb tool @jif-oai - #8966 nit: rename session metric @jif-oai - #8969 chore: non mutable btree when building specs @jif-oai - #8968 chore: move otel provider outside of trace module @jif-oai - #8973 chore: add mcp call metric @jif-oai - #8970 chore: add approval metric @jif-oai - #8975 chore: metrics tool call @jif-oai - #8901 chore: update metrics temporality @jif-oai - #8954 Work around crash in system-configuration library @etraut-openai - #8873 fix(app-server): set originator header from initialize JSON-RPC request @owenlin0 - #8909 Add config to disable /feedback @gt-oai - #8985 chore: nuke telemetry file @jif-oai - #8986 Revert "fix(app-server): set originator header from initialize JSON-RPC request" @jif-oai - #8978 nit: rename to analytics_enabled @jif-oai - #8963 renaming: task to turn @jif-oai - #8555 fix: add tui.alternate_screen config and --no-alt-screen CLI flag for Zellij scrollback @hjanuschka - #8875 feat: add support for building with Bazel @zbarsky-openai - #8988 fix(app-server): set originator header from initialize (re-revert) @owenlin0 - #8766 fix: harden arg0 helper PATH handling @viyatb-oai - #8949 Log unhandled sse events @pakrym-oai - #8996 Add hierarchical agent prompt @pakrym-oai - #9003 Delete announcement_tip.toml @pakrym-oai - #8940 Refactor remote models tests to use TestCodex builder @aibrahim-oai - #8981 Add model provider info to /status if non-default @gt-oai - #8984 Add URL to responses error messages @gt-oai - #9008 fix: add .git to .bazelignore @bolinfest - #9010 fix: include AGENTS.md as repo root marker for integration tests @bolinfest - #8950 Label attached images so agent can understand in-message labels @charley-oai - #9032 Revert "Delete announcement_tip.toml" @jif-oai - #9018 fix: support remote arm64 builds, as well @zbarsky-openai - #8983 feat: testing harness for collab 1 @jif-oai - #9088 feat: add wait tool implementation for collab @jif-oai - #9090 feat: add close tool implementation for collab @jif-oai - #8971 add static mcp callback uri support @WhammyLeaf - #9099 nit: add docstring @jif-oai - #8994 feat: wire fork to codex cli @apanasenko-oai - #9071 chore(deps): bump tui-scrollbar from 0.2.1 to 0.2.2 in /codex-rs @dependabot - #9072 chore(deps): bump ts-rs from 11.0.1 to 11.1.0 in /codex-rs @dependabot - #9074 chore(deps): bump which from 6.0.3 to 8.0.0 in /codex-rs @dependabot - #9075 chore(deps): bump clap from 4.5.53 to 4.5.54 in /codex-rs @dependabot - #9076 chore(deps): bump tokio-util from 0.7.16 to 0.7.18 in /codex-rs @dependabot - #8957 feat: hot reload mcp servers @shijie-oai - #9107 Remove unused conversation_id header @pakrym-oai - #9080 Add some tests for image attachments @charley-oai - #9114 Extract single responses SSE event parsing @pakrym-oai - #8961 Assemble sandbox/approval/network prompts dynamically @aibrahim-oai - #9102 Add model client sessions @pakrym-oai - #8246 fix(tui): show in-flight coalesced tool calls in transcript overlay @Chriss4123 - #9127 Reuse websocket connection @pakrym-oai - #9117 fix: prompt for unsafe commands on Windows @bolinfest - #9128 Websocket append support @pakrym-oai - #9077 Send message by default mid turn. queue messages by tab @aibrahim-oai - #9049 Handle image paste from empty paste events @aibrahim-oai - #8952 Use markdown for migration screen @aibrahim-oai - #9109 Updated heuristic for tool call summary to detect file modifications @etraut-openai - #9138 Show tab queue hint in footer @aibrahim-oai - #9140 Use thread rollback for Esc backtrack @aibrahim-oai - #9143 chore: better error handling on collab tools @jif-oai - #9147 nit: collab send input cleaning @jif-oai - #9129 Support response.done and add integration tests @pakrym-oai - #8798 ollama: default to Responses API for built-ins @drifkin - #8956 add generated jsonschema for config.toml @sayan-oai - #9122 Fix queued messages during /review @charley-oai - #9126 fix: drop session span at end of the session @apanasenko-oai - #9101 Restrict MCP servers from `requirements.toml` @gt-oai - #9020 fix(tui): document paste-burst state machine @joshka-oai - #9118 feat(app-server): add an --analytics-default-enabled flag @owenlin0 - #9130 Fresh tooltips @mzeng-openai - #9134 fix(windows-sandbox-rs) bump SETUP_VERSION @dylan-hurd-oai - #9121 test(tui): add deterministic paste-burst tests @joshka-oai - #8661 Fix spinner/Esc interrupt when MCP startup completes mid-turn @2mawi2 - #9165 Allow close skill popup with esc. @xl-openai - #9169 Fix flakiness in WebSocket tests @pakrym-oai - #9011 fix: report an appropriate error in the TUI for malformed rules @bolinfest - #9166 fix: integration test for #9011 @bolinfest - #9177 feat: add bazel-codex entry to justfile @bolinfest - #9175 WebSocket test server script @pakrym-oai - #9136 Update models.json @github-actions - #9168 clean models manager @aibrahim-oai - #8933 [CODEX-4427] improve parsed commands @aibrahim-oai - #9174 Renew cache ttl on etag match @aibrahim-oai - #9124 fix(tui): harden paste-burst state transitions @joshka-oai - #9186 Use offline cache for tui migrations @aibrahim-oai - #9182 Improve handling of config and rules errors for app server clients @etraut-openai - #8997 chore: clarify default shell for unified_exec @sayan-oai - #9192 feat: add threadId to MCP server messages @bolinfest - #9188 change api default model @aibrahim-oai - #9150 feat: add sourcing of rc files to shell snapshot @jif-oai - #9155 fix: shell snapshot clean-up @jif-oai - #9146 feat: return an error if the image sent by the user is a bad image @jif-oai - #9197 feat: only source shell snapshot if the file exists @jif-oai - #9145 fix: drop double waiting header in TUI @jif-oai - #9194 Render exec output deltas inline @aibrahim-oai - #9156 chore: clamp min yield time for empty write_stdin @jif-oai - #9105 feat: add auto refresh on thread listeners @jif-oai - #9112 feat: add support for read-only bind mounts in the linux sandbox @viyatb-oai - #9179 Use current model for review @pakrym-oai

OpenCodev1.1.20

v1.1.20

14. Januar 2026

## Core - Fixed "load more" button behavior in desktop sidebar (@shanebishop1) - Fix plan mode when not in git worktree - Skip duplicate system prompt for Codex OAuth sessions (@zerone0x) ## TUI - Fixed Docker image after SST rename in tips (@goniz) - Show connected providers in /connect dialog (@AksharP5) - Deduplicate file references in sent prompts (@felixzsh) - Track all timeouts in Footer component to prevent memory leaks (@sauerdaniel) - Make TUI work when OPENCODE_SERVER_PASSWORD is set (@LeonMueller-OneAndOnly) ## SDK - Add missing metadata() and ask() definitions to ToolContext type (@danielmbrasil) **Thank you to 21 community contributors:** - @LeonMueller-OneAndOnly: - fix(TUI): make tui work when OPENCODE_SERVER_PASSWORD is set (#8179) - fix(github): add persist-credentials: false to workflow templates (#8202) - fix(TUI): make tui work when OPENCODE_SERVER_PASSWORD is set (#8179) - @Eduard-Voiculescu: - docs: Update plan mode restrictions (#8290) - docs: Update plan mode restrictions (#8290) - @mthezi: - docs: add 302ai provider (#8142) - docs: add 302ai provider (#8142) - @danielmbrasil: - fix: add missing metadata() and ask() defintions to ToolContext type (#8269) - fix: add missing metadata() and ask() defintions to ToolContext type (#8269) - @sauerdaniel: - fix(tui): track all timeouts in Footer to prevent memory leak (#8255) - fix(mcp): close existing client before reassignment to prevent leaks (#8253) - fix(state): delete key from recordsByKey on instance disposal (#8252) - fix(tui): track all timeouts in Footer to prevent memory leak (#8255) - @felixzsh: - fix: deduplicate file refs in sent prompts (#8303) - fix: deduplicate file refs in sent prompts (#8303) - @usvimal: - fix(desktop): correct health check endpoint URL to /global/health (#8231) - @dbpolito: - feat(desktop): Adding Provider Icons (#8215) - fix(desktop): Revert provider icon on select model dialog (#8245) - feat(desktop): Ask Question Tool Support (#8232) - @Brendonovich: - console: reduce desktop download cache ttl to 5 minutes - @cmdr-chara: - feat: add Undertale and Deltarune built-in themes (#8240) - @zeke: - docs: document ~/.claude/CLAUDE.md compatibility behavior (#8268) - @vglafirov: - feat: Add GitLab Duo Agentic Chat Provider Support (#7333) - fix: Add Plugin Mocks to Provider Tests (#8276) - @dmmulroy: - fix(cli): mcp auth duplicate radio button icon (#8273) - @josephbharrison: - fix(prompt-input): handle Shift+Enter before IME check to prevent stuck state (#8275) - @AlanThiec: - fix: update User-Agent string to latest Chrome version in webfetch (#8284) - @zerone0x: - fix(session): skip duplicate system prompt for Codex OAuth sessions (#8357) - @AksharP5: - feat: show connected providers in /connect dialog (#8351) - @goniz: - fix(opencode): fix docker image after sst rename in tips (#8376) - @neriousy: - fix(app): file listing (#8309) - @ajzbc: - fix(ui): layout-bottom icons (#8330) - @shanebishop1: - fix(desktop): "load more" button behavior in desktop sidebar (#8430)

Gemini CLIv0.24.0

Release v0.24.0

14. Januar 2026

## Was sich geändert hat * chore(core): refactor model resolution and cleanup fallback logic von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15228 * Add Folder Trust Support To Hooks von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15325 * Record timestamp with code assist metrics. von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15439 * feat(policy): implement dynamic mode-aware policy evaluation von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15307 * fix(core): use debugLogger.debug for startup profiler logs von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15443 * feat(ui): Add security warning and improve layout for Hooks list von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15440 * fix #15369, prevent crash on unhandled EIO error in readStdin cleanup von @ElecTwix in https://github.com/google-gemini/gemini-cli/pull/15410 * chore: improve error messages for --resume von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15360 * chore: remove clipboard file von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15447 * Implemented unified secrets sanitization and env. redaction options von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15348 * feat: automatic `/model` persistence across Gemini CLI sessions von @niyasrad in https://github.com/google-gemini/gemini-cli/pull/13199 * refactor(core): remove deprecated permission aliases from BeforeToolHookOutput von @StoyanD in https://github.com/google-gemini/gemini-cli/pull/14855 * fix: add missing `type` field to MCPServerConfig von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15465 * Make schema validation errors non-fatal von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15487 * chore: limit MCP resources display to 10 by default von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15489 * Add experimental in-CLI extension install and uninstall subcommands von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/15178 * feat: Add A2A Client Manager and tests von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15485 * feat: terse transformations of image paths in text buffer von @psinha40898 in https://github.com/google-gemini/gemini-cli/pull/4924 * Security: Project-level hook warnings von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15470 * Added modifyOtherKeys protocol support for tmux von @ved015 in https://github.com/google-gemini/gemini-cli/pull/15524 * chore(core): fix comment typo von @Mapleeeeeeeeeee in https://github.com/google-gemini/gemini-cli/pull/15558 * feat: Show snowfall animation for holiday theme von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15494 * do not persist the fallback model von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15483 * Resolve unhandled promise rejection in ide-client.ts von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/15587 * fix(core): handle checkIsRepo failure in GitService.initialize von @Mapleeeeeeeeeee in https://github.com/google-gemini/gemini-cli/pull/15574 * fix(cli): add enableShellOutputEfficiency to settings schema von @Mapleeeeeeeeeee in https://github.com/google-gemini/gemini-cli/pull/15560 * Manual nightly version bump to 0.24.0-nightly.20251226.546baf993 von @galz10 in https://github.com/google-gemini/gemini-cli/pull/15594 * refactor(core): extract static concerns from CoreToolScheduler von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15589 * fix(core): enable granular shell command allowlisting in policy engine von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15601 * chore/release: bump version to 0.24.0-nightly.20251227.37be16243 von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15612 * refactor: deprecate legacy confirmation settings and enforce Policy Engine von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15626 * Migrate console to coreEvents.emitFeedback oder debugLogger von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/15219 * Exponential back-off retries for retryable error without a specified … von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15684 * feat(agents): add support for remote agents and multi-agent TOML files von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15437 * Update wittyPhrases.ts von @segyges in https://github.com/google-gemini/gemini-cli/pull/15697 * refactor(auth): Refactor non-interactive mode auth validation & refresh von @skeshive in https://github.com/google-gemini/gemini-cli/pull/15679 * Revert "Update wittyPhrases.ts (#15697)" von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15719 * fix(hooks): deduplicate agent hooks and add cross-platform integration tests von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15701 * Implement support for tool input modification von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15492 * Add instructions to the extensions update info notification von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/14907 * Add extension settings info to /extensions list von @chrstnb in https://github.com/google-gemini/gemini-cli/pull/14905 * Agent Skills: Implement Core Skill Infrastructure & Tiered Discovery von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15698 * chore: remove cot style comments von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15735 * feat(agents): Add remote agents to agent registry von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15711 * feat(hooks): implement STOP_EXECUTION and enhance hook decision handling von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15685 * Fix build issues caused by year-specific linter rule von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15780 * fix(core): handle unhandled promise rejection in mcp-client-manager von @kamja44 in https://github.com/google-gemini/gemini-cli/pull/14701 * log fallback mode von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15817 * Agent Skills: Implement Autonomous Activation Tool & Context Injection von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15725 * fix(core): improve shell command with redirection detection von @galz10 in https://github.com/google-gemini/gemini-cli/pull/15683 * Add security docs von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15739 * feat: add folder suggestions to `/dir add` command von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15724 * Agent Skills: Implement Agent Integration and System Prompt Awareness von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15728 * chore: cleanup old smart edit settings von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15832 * Agent Skills: Status Bar Integration for Skill Counts von @NTaylorMullen in https://github.com

Continue.devv1.5.32

CLI Stable Release v1.5.32

14. Januar 2026

Direktes stabiles Release aus dem main-Branch. Version 1.5.32 wurde direkt aus dem neuesten main-Branch gebaut und veröffentlicht.

OpenCodev1.1.19

v1.1.19

14. Januar 2026

## Core - Do not allow agent to ask custom-less questions - Improve prompt for explore agent - Add plugin mocks to provider tests (@vglafirov) - Update User-Agent string to latest Chrome version in webfetch (@AlanThiec) - Add plan mode with enter and exit tools - Add family to GPT 5.2 Codex in Codex plugin - Add GitLab Duo Agentic Chat provider support (@vglafirov) - Delete key from recordsByKey on instance disposal (@sauerdaniel) - Close existing MCP client before reassignment to prevent resource leaks (@sauerdaniel) - Added Ask Question tool support for desktop application (@dbpolito) ## TUI - Tweak external directory permission rendering in TUI - Ensure external directory and Bash tool invocations render working directory details - Fix duplicate radio button icon in MCP auth CLI interface (@dmmulroy) ## Desktop - Handle Shift+Enter before IME check to prevent stuck state in prompt input (@josephbharrison) - Revert provider icon on select model dialog in desktop app (@dbpolito) **Thank you to 8 community contributors:** - @dbpolito: - fix(desktop): Revert provider icon on select model dialog (#8245) - feat(desktop): Ask Question Tool Support (#8232) - @cmdr-chara: - feat: add Undertale and Deltarune built-in themes (#8240) - @sauerdaniel: - fix(mcp): close existing client before reassignment to prevent leaks (#8253) - fix(state): delete key from recordsByKey on instance disposal (#8252) - @zeke: - docs: document ~/.claude/CLAUDE.md compatibility behavior (#8268) - @vglafirov: - feat: Add GitLab Duo Agentic Chat Provider Support (#7333) - fix: Add Plugin Mocks to Provider Tests (#8276) - @dmmulroy: - fix(cli): mcp auth duplicate radio button icon (#8273) - @josephbharrison: - fix(prompt-input): handle Shift+Enter before IME check to prevent stuck state (#8275) - @AlanThiec: - fix: update User-Agent string to latest Chrome version in webfetch (#8284)

Claude Codev2.1.7

v2.1.7

14. Januar 2026

## What's changed - Added `showTurnDuration` setting to hide turn duration messages (e.g., "Cooked for 1m 6s") - Added ability to provide feedback when accepting permission prompts - Added inline display of agent's final response in task notifications, making it easier to see results without reading the full transcript file - Fixed security vulnerability where wildcard permission rules could match compound commands containing shell operators - Fixed false "file modified" errors on Windows when cloud sync tools, antivirus scanners, or Git touch file timestamps without changing content - Fixed orphaned tool_result errors when sibling tools fail during streaming execution - Fixed context window blocking limit being calculated using the full context window instead of the effective context window (which reserves space for max output tokens) - Fixed spinner briefly flashing when running local slash commands like `/model` or `/theme` - Fixed terminal title animation jitter by using fixed-width braille characters - Fixed plugins with git submodules not being fully initialized when installed - Fixed bash commands failing on Windows when temp directory paths contained characters like `t` or `n` that were misinterpreted as escape sequences - Improved typing responsiveness by reducing memory allocation overhead in terminal rendering - Enabled MCP tool search auto mode by default for all users. When MCP tool descriptions exceed 10% of the context window, they are automatically deferred and discovered via the MCPSearch tool instead of being loaded upfront. This reduces context usage for users with many MCP tools configured. Users can disable this by adding `MCPSearch` to `disallowedTools` in their settings. - Changed OAuth and API Console URLs from console.anthropic.com to platform.claude.com - [VSCode] Fixed `claudeProcessWrapper` setting passing the wrapper path instead of the Claude binary path

Windsurfv1.13.8

GPT-5.2-Codex

14. Januar 2026

Unterstützung für GPT-5.2-Codex mit vier Reasoning-Aufwendungen hinzugefügt.

Continue.devv1.5.31

CLI Stable Release v1.5.31

13. Januar 2026

Direktes stabiles Release aus dem main-Branch. Version 1.5.31 wurde direkt aus dem neuesten main-Branch gebaut und veröffentlicht.

Continue.devv@continuedev/config-yaml@1.38.0

@continuedev/config-yaml@1.38.0

13. Januar 2026

# [1.38.0](https://github.com/continuedev/continue/compare/@continuedev/config-yaml@1.37.0...@continuedev/config-yaml@1.38.0) (2026-01-13) ### Bug Fixes * Fallun abweisende Matching-Strategie für Suchen und Ersetzen hinzufügen (case-insensitive) ([ac476cc](https://github.com/continuedev/continue/commit/ac476cc02884bf6d3dd52adf67a0c6eb89ab9d0a)) * Fehlende Hub-Mock-Verweigerung im Test hinzufügen ([53a9fa4](https://github.com/continuedev/continue/commit/53a9fa4cb34d97a9f3e2af1f26a02b10bbd0a9a1)) * Env-Variablen auch im Write-Job-Zusammenfassungs-Schritt verwenden ([8969f42](https://github.com/continuedev/continue/commit/8969f422619f63becf2f778c4465cf192ee83e18)) * **anthropic:** Azure-gehostete Anthropic-Endpunkte unterstützen ([9e8bfcd](https://github.com/continuedev/continue/commit/9e8bfcd9396b9abf7643c79bb53f7263da8138c6)) * Änderung von 'Agent:' zu 'Config:' in Einführungsnachricht und Tests aktualisieren ([c1e4668](https://github.com/continuedev/continue/commit/c1e466836acd0b2a5e81f2925c5d2e14691543d9)) * klären, dass paralleles Editieren mit sich selbst auch nicht möglich ist ([46b31af](https://github.com/continuedev/continue/commit/46b31af330aa669d93c777d5fa577b2ca7c72225)) * **cli:** Modellerkennung erweitert, um Llama, Nemotron und Mistral abzudecken ([#8845](https://github.com/continuedev/continue/issues/8845)) ([528a8ab](https://github.com/continuedev/continue/commit/528a8abceb69a6afb12e0a32ea5eb8acbc1e8241)), schließt [#1](https://github.com/continuedev/continue/issues/1) * **cli:** Testannahmen und Importpfad für uriUtils korrigiert ([3ccaa09](https://github.com/continuedev/continue/commit/3ccaa092e65fdd520e3e8b157feaf0eb3d18ba25)) * **cli:** Pfad-Konvertierung von URI und umgekehrt ([d09b3e2](https://github.com/continuedev/continue/commit/d09b3e2126d7349f23e50e90107e154838bc06ae)) * **cli:** Kreisförmige Abhängigkeit im UploadArtifact-Tool auflösen ([b5da44d](https://github.com/continuedev/continue/commit/b5da44ddd665d571b9669a9832e2d5b34a9f743c)) * **cli:** CLI-Neustart für manuelles Update ([595986a](https://github.com/continuedev/continue/commit/595986a6fb46ccd4e023078e9e3a8aeb911629c7)) * **cli:** Logger.info statt console.log für Konsistenz verwenden ([00b665f](https://github.com/continuedev/continue/commit/00b665f089e1b1c1166bf8c7d275465b6fb8f5e2)) * Root-ESLint auf v8 beschränken, um Core-Abhängigkeit zu entsprechen ([ad6ab80](https://github.com/continuedev/continue/commit/ad6ab8024b888816dcf0c8eb39921a98777ec5ab)) * Kontextlängenfixes, Trunkierung, etc. ([613ffb8](https://github.com/continuedev/continue/commit/613ffb8ca9847fcb3ff99beec6f191b25cd85f9b)) * Korrekte Import-Reihenfolge für ESLint-Konformität ([657c5fe](https://github.com/continuedev/continue/commit/657c5fe023cf25473d4365112f0162af38922d50)) * Falls nicht Markdown-Datei, nicht auf relativen Pfad zurückfallen ([d93b432](https://github.com/continuedev/continue/commit/d93b4321a550acf7151949d1fee1c436945e2ee0)) * Sicherstellen, dass cross-target LanceDB-Binärdateien korrekt kopiert werden ([#9100](https://github.com/continuedev/continue/issues/9100)) ([291f8f5](https://github.com/continuedev/continue/commit/291f8f5dd2efec7a53e49cc7e2a1b7e6a0833466)) * Env-Stubs für CLI-Plattform-Tests ([6b3c0be](https://github.com/continuedev/continue/commit/6b3c0be2f1443e47db3e985560cb859b8e42f6a9)) * Lint-Fehler beheben und Tests für Markdown-only-Fallback aktualisieren ([4c0c532](https://github.com/continuedev/continue/commit/4c0c53257baf7edb943c76155fa6b0bde79b554f)) * JSON-Inhalte von create_new_file behandeln ([5f14a2f](https://github.com/continuedev/continue/commit/5f14a2fa664f3c9b0df23eb01c2f774a69cfd764)), schließt [#8972](https://github.com/continuedev/continue/issues/8972) * Nutzungsfeld in createSession und startNewSession initialisieren ([5c50e91](https://github.com/continuedev/continue/commit/5c50e911d203fc0a7ae6407f84bdfafd8c1f19f4)) * Inject-Blocks-Tests für CLI-Geheimnisauflösung ([219f566](https://github.com/continuedev/continue/commit/219f5661c9999daa5e185740dd2fa1d24f0187b2)) * **intellij:** Verhindern von ConcurrentModificationException beim Zugriff auf Keymaps ([5d1cd00](https://github.com/continuedev/continue/commit/5d1cd001039eb2b8b037db7dbdeaf91a11be59b0)) * Linting und Tests ([3adcac2](https://github.com/continuedev/continue/commit/3adcac22bbe34748126e54bd22207a135807416a)) * Baum-Sitter-Lookup in .js- und .ts-Dateien nur das zuletzt vorhergehende Kommentar vor Code-Block erfassen lassen ([cb4db87](https://github.com/continuedev/continue/commit/cb4db878b89828d01880251cb490af50d15dbd44)) * Hauptzweig zusammenführen ([48c7616](https://github.com/continuedev/continue/commit/48c76160d05563b1aebe18ae53eea61cb0f5d614)) * my config not myproject ([23b359a](https://github.com/continuedev/continue/commit/23b359aab0dc964d6e96a2bb641538d13b285e7f)) * **openai-adapters:** Defensiv-Typprüfungen für stream.usage Promise hinzufügen ([7d3fa6d](https://github.com/continuedev/continue/commit/7d3fa6daa97519df87c70b32fbbe1f46e6fa3bc7)) * **openai-adapters:** Fallback zu stream.usage Promise für Usage-Tokens hinzufügen ([bbeec4b](https://github.com/continuedev/continue/commit/bbeec4b1bf7225c0159cccb5562ff2c7701e356e)) * **openai-adapters:** addressing 4 PR-Review-Probleme ([93d9c12](https://github.com/continuedev/continue/commit/93d9c123d348a1a79829299b9756781a8839b1ae)) * **openai-adapters:** Nicht-Ausgabe von Usage aus Voll-Stream-Finish-Ereignis ([a89187b](https://github.com/continuedev/continue/commit/a89187b409314fc4e502290b41d0071d459102ac)) * **openai-adapters:** Timing des API-Initialisierungs-Turns für Multi-Turn-Tools im Test timing korr

OpenCodev1.1.18

v1.1.18

13. Januar 2026

## TUI - Add persist-credentials: false to workflow templates to prevent credential exposure (@LeonMueller-OneAndOnly) - Add fullscreen view to permission prompt ## Desktop - Add provider icons to desktop application (@dbpolito) - Correct desktop health check endpoint URL to /global/health (@usvimal) **Thank you to 4 community contributors:** - @usvimal: - fix(desktop): correct health check endpoint URL to /global/health (#8231) - @dbpolito: - feat(desktop): Adding Provider Icons (#8215) - @Brendonovich: - console: reduce desktop download cache ttl to 5 minutes - @LeonMueller-OneAndOnly: - fix(github): add persist-credentials: false to workflow templates (#8202)

OpenCodev1.1.17

v1.1.17

13. Januar 2026

## Core - Remove the symlinkBinary function call that replaces the wrapper script - Fix Homebrew auto-upgrade functionality ## TUI - Prevent question tool keybindings when dialog is open (@zerone0x) ## Desktop - Kill existing OpenCode CLI processes on macOS when launching desktop app (@Brendonovich) - Set serverPassword in desktop app (@Brendonovich) - Provide PTY socket authentication if available from desktop (@Brendonovich) - Tie desktop and CLI to the same Windows JobObject (@Brendonovich) **Thank you to 3 community contributors:** - @zerone0x: - fix(tui): prevent question tool keybindings when dialog is open (#8147) - @Brendonovich: - feat(desktop): Tie desktop & CLI to the same Windows JobObject (#8153) - fix(app): provide pty socket auth if available from desktop (#8210) - fix(desktop): set serverPassword - desktop: macos killall opencode-cli on launch - @OpeOginni: - fix(ui): track memo-based items in useFilteredList without affecting async function based lists (#8216)

OpenClawv2026.1.12-2

clawdbot 2026.1.12-2

13. Januar 2026

### Fixes - Packaging: include `dist/memory/**` im npm tarball (behebt `ERR_MODULE_NOT_FOUND` für `dist/memory/index.js`).

OpenClawv2026.1.12

clawdbot 2026.1.12

13. Januar 2026

### Highlights - **BREAKING:** Umbenennung von Chat-“Providers” (Slack/Telegram/WhatsApp/…) zu **Channels** über CLI/RPC/Config; Legacy-Config-Schlüssel migrieren beim Laden automatisch (und werden beim Schreiben wieder als `channels.*` gespeichert). - Memory: Speicher für Agenten-Gedächtnisse um eine vektorbasierte Suche (Markdown-only) mit SQLite-Index, Chunking, lazy Sync + Dateiwache, und per-Agentaktivierung/-fallback. - Plugins: vollständige Parität des Voice-Call-Plugins wiederherstellen (Telnyx/Twilio, Streaming,Inbound-Richtlinien, Tools/CLI). - Models: synthetischen Provider hinzufügen plus Moonshot Kimi K2 0905 + turbo/Thinking-Varianten (mit Docs). (#811) — danke @siraht; (#818) — danke @mickahouan. - Cron: One-Shot-Schedules akzeptieren ISO-Timestamps (UTC) mit optionalem Delete-after-Run; Cron-Jobs können auf einen bestimmten Agenten abzielen (CLI + macOS/Control UI). - Agents: Kompaktions-Modus-Konfig hinzufügen mit optionaler Sicherheitszusammenfassung und Per-Agent-Modell-Fallbacks. (#700) — danke @thewilloftheshadow; (#583) — danke @mitschabaude-bot. ### New & Improved - Memory: benutzerdefinierte OpenAI-kompatible Embedding-Endpunkte hinzufügen; OpenAI/local `node-llama-cpp` Embeddings mit per-Agent-Overrides und Provider-Metadaten in tools/CLI unterstützen. (#819) — danke @mukhtharcm. - Memory: neuer `clawdbot memory` CLI plus Tools `memory_search`/`memory_get` mit Snippets + Zeilenbereichen; Index gespeichert unter `~/.clawdbot/memory/{agentId}.sqlite` mit standardmäßig aktivem Watch. - Agents: Gedächtnisabruf-Richtlinien verstärken; Workspace-Bootstrap-Trunkierung konfigurierbar machen (Standard 20k) mit Warnungen; Standard-Sub-Agent-Modellkonfiguration hinzufügen. - Tools/Sandbox: Tool-Profile + Gruppen-Abkürzungen hinzufügen; Unterstützung von Tool-Policy-Gruppen in `tools.sandbox.tools`; Legacy-`memory`-Abkürzung entfernen; Docker-Bind-Mounts über `docker.binds` zulassen. (#790) — danke @akonyer. - Tools: Browser-`scrollintoview`-Aktion hinzufügen; Aliase für Claude/Gemini Tool-Parameter zulassen; Thinking-Modus `xhigh` für GPT-5.2/Codex mit sicherem Downgrade zulassen. (#793) — danke @hsrvc; (#444) — danke @grp06. - Gateway/CLI: Tailscale Binary Discovery, benutzerdefinierter Bind-Modus, und Probe-Auth-Retry hinzufügen; `clawdbot dashboard`-Auto-Open-Flow hinzufügen; Standard-native Slash-Befehle auf `"auto"` mit provider-spezifischen Overrides. (#740) — danke @jeffersonwarrior. - Auth/Onboarding: Chutes OAuth (PKCE + Refresh + Onboarding-Choice) hinzufügen; API-Key-Eingaben normalisieren; Default TUI-Onboarding auf `deliver: false`. (#726) — danke @FrieSei; (#791) — danke @roshanasingh4. - Providers: `discord.allowBots` hinzufügen; legacy MiniMax M2 aus Standard-Katalogen trimmen; MiniMax Vision zum Coding Plan VLM-Endpunkt routen (nimmt auch Inputs wie `@/path/to/file.png` an). (#802) — danke @zknicker. - Gateway: Tailscale Serve Identity-Header zulassen, um Token-Auth zu befriedigen; Control-UI-Assets neu aufbauen, wenn das Protokoll-Schema neuer ist. (#823) — danke @roshanasingh4; (#786) — danke @meaningfool. - Heartbeat: Standard `ackMaxChars` auf 300 setzen, damit kurze `HEARTBEAT_OK`-Antworten intern bleiben. ### Installer - Install: Führe nach Git-Installationen/-Updates `clawdbot doctor --non-interactive` aus und benachrichtige Daemon-Neustarts, wenn erkannt. ### Fixes - Doctor: Warnungen bei pnpm-Workspaces-Mismatches, fehlenden Control-UI-Assets und fehlenden tsx-Binärdateien; UI-Neubaukosten anbieten. - Models/Providers: Behandlung von Credential-Validierungsfehlern als Auth-Fehler zur Triggerung von Fallbacks; Normalisierung von `${ENV_VAR}`-apiKey-Werten und Auto-Fill fehlender Provider-Schlüssel; explizite GitHub Copilot-Anbieter-Konfig + Agent-DIR-Auth-Profile beibehalten. (#822) — danke @sebslight; (#705) — danke @TAGOOZ. - Auth: Ungültige Auth-Profile aus der Reihenfolge entfernen, damit Umgebungsvariablen weiterhin für Provider wie MiniMax nutzbar bleiben. - Gemini: Gemini 3-IDs zu Preview-Varianten normalisieren; Gemini CLI-Tool-Aufrufs-/Antwort-IDs entfernen; fehlende `thought_signature` downgraden; Claude `msg_*`-Thought-Signature-Felder entfernen, um Base64-Dekodierungsfehler zu vermeiden. (#795) — danke @thewilloftheshadow; (#783) — danke @ananth-vardhan-cn; (#793) — danke @hsrvc; (#805) — danke @marcmarg. - Agents: Auto-Wiederherstellung aus Kompaktions-Kontext-Overflows durch Zurücksetzen der Sitzung und erneuten Versuch; Overflow-Details aus eingebetteten Durchläufen an Weitergeber weiterreichen. - MiniMax: Fehlformatierte Tool-Aufruf-XML entfernen; `MiniMax-VL-01` im impliziten Provider für Image-Paarung einschließen. (#809) — danke @latitudeki5223. - Onboarding/Auth: Berücksichtigung von `CLAWDBOT_AGENT_DIR` / `PI_CODING_AGENT_DIR` beim Schreiben von Auth-Profilen (MiniMax). (#829) — danke @roshanasingh4. - Anthropic: `overloaded_error` mit freundlicher Meldung handeln und Failover-Klassifikation; zusammenführen aufeinanderfolgender Benutzerspuren (neueste Metadaten bewahren) vor Validierung, um falsche Rollen-Fehler zu vermeiden. (#832) — danke @danielz1z; (#804) — danke @ThomsenDrake. - Messaging: Kontext-Isolation für Nachrichten-Tool-Sends erzwingen; Tipp-Indikatoren während der Tool-Ausführung aktiv halten. (#793) — danke @hsrvc; (#450, #447) — danke @thewilloftheshadow. - Auto-Reply: `/status`-Whitelist-Verhalten, Begründungs-Tag-Verfolgung bei Fallback, und System-Ereignis-Verarbeitung für erhöhte/Begründungs-Toggle. (#810) — danke @mcinteerj. - System-Ereignisse: Lokale Zeitstempel einbeziehen, wenn Ereignisse in Prompts injiziert werden. (#245) — danke @thewilloftheshadow. - Auto-Reply: Unklare `/model`-Treffe vermeiden; Streaming-Block-Antwort-Mediendateien korrigieren; Heartbeat-Antworten >300 Zeichen beibehalten statt Abbruch. - Discord/Slack: Zentralisieren der Reply-Thread-Planung; AutoThread-Routing korrigieren + pro-Channel AutoThread hinzufügen; doppelte Listener vermeiden; Behalte Begründungs-Format-Italics; Kanal-Eltern über Nachrichten-Tool löschen zulassen. (#800, #807) — danke @davidguttman; (#744) — danke @thewilloftheshadow. - Telegram: Foren-Topic-Thread-IDs beibehalten, Polling-Offsets persistieren, Bindings im Webhook-Modus respektieren und Tipp-Indikator in General-Themen anzeigen. (#727, #739) — danke @thewilloftheshadow; (#821) — danke @gumadeiras; (#779) — danke @azade-c. - Slack: Slash-Befehle mit oder ohne führendem `/` für benutzerdefinierte Befehls-Konfigurationen akzeptieren. (#798) — danke @thewilloftheshadow. - Cron: Fehlende Jobs korrekt persistieren; Alias `jobId` für Update/Run/Remove-Param verwenden. (#205, #252) — danke @thewilloftheshadow. - Gateway/CLI: Berücksichtigen von `CLAWDBOT_LAUNCHD_LABEL` / `CLAWDBOT_SYSTEMD_UNIT` Overrides; `agents.list` respektiert explizite Config; weniger noisy Loopback-WS-Logs während Tests; führe `clawdbot doctor --non-interactive` bei Updates aus. (#781) — danke @ronyrus. - Onboarding/Control UI: Ungültige Configs ablehnen (erst Doctor ausführen); Windows-Browser-URLs für OAuth quoten; Chat-Scroll

OpenCodev1.1.16

v1.1.16

13. Januar 2026

## Core - Handle write errors when updating schema in OpenCode config (@ShoeBoom) - Ensure clear_thinking is disabled for zai model - Improved edit and write tool feedback to prevent confusion about whether changes were successfully applied - Prevent [object Object] error display in console output (@lengmodkx) - Respect agent task permission for nested sub-agents (@malhashemi) - Fix max completion tokens error for Cloudflare provider (@roerohan) - Use reasoningConfig for Anthropic models on Bedrock (@Adam8234) - Store API URL in session metadata (@jkoelker) - Add nova-2 to region prefix models - Tweak truncation directory permissions - Make environment file reads ask user for permission instead of blocking - Use Biome check instead of format to include import sorting (@kynnyhsap) - Ensure plurals are properly handled ## TUI - Support debug tool calling directly in CLI (@Eric-Guo) - Don't submit prompt when switching sessions (@LeonMueller-OneAndOnly) - Await auth.set to prevent race condition when setting API key (@darshkpatel) - Actually modify OpenCode config when adding MCP servers with `mcp add` command (@paoloricciuti) ## Desktop - Spawn local server with password in desktop app (@Brendonovich) - Add missing args to Windows Tauri CLI spawn (@pschiel) - Implement better dynamic terminal numbering (@OpeOginni) - Add icon button for server removal in dialog select server (@OpeOginni) **Thank you to 15 community contributors:** - @kynnyhsap: - fix(formatter): use biome `check` instead of `format` to include import sorting (#8057) - @OpeOginni: - feat(dialog-select-server): add icon button for server removal (#8053) - feat(terminal): implement better dynamic terminal numbering (#8081) - @jkoelker: - fix(session): store api url in metadata (#8090) - @paoloricciuti: - fix: actually modify opencode config with `mcp add` (#7339) - @jerome-benoit: - fix(nix): initialize hashes.json with per-system format (#8095) - @darshkpatel: - fix: await auth.set to prevent race condition when setting API key (#8075) - @Adam8234: - fix(bedrock): use reasoningConfig for Anthropic models on Bedrock (#8092) - @roerohan: - fix: max completion tokens error for cloudflare (#7970) - @malhashemi: - fix(task): respect agent task permission for nested sub-agents (#8111) - @pschiel: - fix: add missing args to windows tauri cli spawn (#8084) - @lengmodkx: - fix: prevent [object Object] error display in console output (#8116) - @LeonMueller-OneAndOnly: - fix(TUI): dont submit prompt when switching sessions (#8016) - @Eric-Guo: - feat(cli): Support debug tool calling directly in CLI. (#6564) - @ShoeBoom: - fix(config): handle write errors when updating schema in opencode config (#8125) - @Brendonovich: - feat(desktop): spawn local server with password (#8139)

Claude Codev2.1.6

v2.1.6

13. Januar 2026

## What's changed - Added search functionality to `/config` command for quickly filtering settings - Added Updates section to `/doctor` showing auto-update channel and available npm versions (stable/latest) - Added date range filtering to `/stats` command - press `r` to cycle between Last 7 days, Last 30 days, and All time - Added automatic discovery of skills from nested `.claude/skills` directories when working with files in subdirectories - Added `context_window.used_percentage` and `context_window.remaining_percentage` fields to status line input for easier context window display - Added an error display when the editor fails during Ctrl+G - Fixed permission bypass via shell line continuation that could allow blocked commands to execute - Fixed false "File has been unexpectedly modified" errors when file watchers touch files without changing content - Fixed text styling (bold, colors) getting progressively misaligned in multi-line responses - Fixed the feedback panel closing unexpectedly when typing 'n' in the description field - Fixed rate limit warning appearing at low usage after weekly reset (now requires 70% usage) - Fixed rate limit options menu incorrectly auto-opening when resuming a previous session - Fixed numpad keys outputting escape sequences instead of characters in Kitty keyboard protocol terminals - Fixed Option+Return not inserting newlines in Kitty keyboard protocol terminals - Fixed corrupted config backup files accumulating in the home directory (now only one backup is created per config file) - Fixed `mcp list` and `mcp get` commands leaving orphaned MCP server processes - Fixed visual artifacts in ink2 mode when nodes become hidden via `display:none` - Improved the external CLAUDE.md imports approval dialog to show which files are being imported and from where - Improved the `/tasks` dialog to go directly to task details when there's only one background task running - Improved @ autocomplete with icons for different suggestion types and single-line formatting - Updated "Help improve Claude" setting fetch to refresh OAuth and retry when it fails due to a stale OAuth token - Changed task notification display to cap at 3 lines with overflow summary when multiple background tasks complete simultaneously - Changed terminal title to "Claude Code" on startup for better window identification - Removed ability to @-mention MCP servers to enable/disable - use `/mcp enable <name>` instead - [VSCode] Fixed usage indicator not updating after manual compact

OpenCodev1.1.15

v1.1.15

12. Januar 2026

## Core - Rename OPENCODE_PASSWORD to OPENCODE_SERVER_PASSWORD for clearer authentication configuration - Fix HTTP exception handling order to prevent NamedError from masking HTTP exceptions - Add password authentication and improve server security - Improve error handling for built-in plugin installation failures - Allow wildcard permissions to match both the command with and without arguments - Fix model selection in title generation to use user's model instead of assistant's ## TUI - Prevent layout jump when toggling tips visibility in TUI (@R44VC0RP) - Add hint for variants toggle in TUI - Fix autocomplete reopening after backspace deletes space in terminal UI (@Raviguntakala) - Rename kvStore to store for consistency in TUI - Redesign tips display on home screen - Redesign tips display on home screen - Remove update complete toast notification from TUI ## Desktop - Disable AppImage build (@Brendonovich) - Improve server detection and connection logic in desktop application (@Brendonovich) - Resolve defaultServerUrl inside AppInterface to improve server URL handling (@Brendonovich) **Thank you to 7 community contributors:** - @Brendonovich: - app: resolve defaultServerUrl inside AppInterface - fix(desktop): improve server detection & connection logic (#7962) - disable appimage - @shubh73: - docs: fix typo (#8041) - @jerome-benoit: - fix(nix): filter optional dependencies by target platform (#8033) - @Raviguntakala: - fix(tui): reopen autocomplete after backspace deletes space (#6031) - @dbpolito: - fix(desktop): Improve User Message Badges for Big Titles and Small Screens (#8037) - @kdcokenny: - docs: fix permission rule ordering in examples (#7010) - @R44VC0RP: - fix(tui): prevent layout jump when toggling tips visibility (#8068)

Claude Codev2.1.5

v2.1.5

12. Januar 2026

## What's changed - Added `CLAUDE_CODE_TMPDIR` environment variable to override the temp directory used for internal temp files, useful for environments with custom temp directory requirements

OpenClawv2026.1.11-3

clawdbot 2026.1.11-3

12. Januar 2026

### Fixes - CLI: vermeiden Top-Level-Await-Warnungen im Entry-Point bei frischen Installationen. - CLI: zeige einen Commit-Hash im Banner bei npm-Installationen an (package.json gitHead Fallback).

OpenClawv2026.1.11-2

clawdbot 2026.1.11-2

12. Januar 2026

### Fixes - Installer: ensure the CLI entrypoint is executable after npm installs. - Packaging: include `dist/plugins/` in the npm package to avoid missing module errors.

OpenClawv2026.1.11-1

clawdbot 2026.1.11-1

12. Januar 2026

## Fixes - Installer: `patches/` in das npm-Paket aufnehmen, damit das Patchen nach der Installation bei npm/bun-Installationen funktioniert.

OpenClawv2026.1.11

clawdbot 2026.1.11

12. Januar 2026

### Highlights - Plugins sind jetzt erstklassig: Loader + CLI-Management, plus das neue Voice Call Plugin. - Config: modulares `$include`-Support für geteilte Config-Dateien. (#731) — thanks @pasogott. - Agents/Pi: Reserviere Kompressions-Spielraum, damit Pre-Compaction-Memory-Schreibungen vor der Auto-Compaction laufen können. - Agents: automatische Pre-Compaction-Memory-Flush-Umstellung, um durable memories vor der Kompaktion zu speichern. ### Changes - Deps: aktualisiere pi-agent-core/pi-ai/pi-coding-agent/pi-tui und aktualisiere den pi-ai Patch. - Dev: @types/node erhöhen. - macOS: Wizard-Debug-CLI hinzufügen und Wizard-Parsing-Hilfen teilen. - CLI/Onboarding: MiniMax-Authetik-Auswahl auf eine einzige M2.1-Option vereinfachen. - CLI: Konfigurationsabschnittsauswahl läuft jetzt in einer Schleife bis Continue. - Docs: MiniMax vs MiniMax Lightning (schneller vs Kosten) erläutern und LM Studio-Beispiel wiederherstellen. - Docs: Cerebras GLM 4.6/4.7 Config-Beispiel (OpenAI-kompatibler Endpoint) hinzufügen. - Onboarding/CLI: Modell-/Auth-Auswahl pro Anbieter gruppieren und Z.AI als GLM 4.7 kennzeichnen. - Onboarding/Docs: Moonshot AI (Kimi K2) Auth-Auswahl + Config-Beispiel hinzufügen. - CLI/Onboarding: Aufforderung zur Wiederverwendung erkannter API-Keys für Moonshot/MiniMax/Z.AI/Gemini/Anthropic/OpenCode. - CLI/Onboarding: MiniMax an den Anfang der Anbieterliste verschieben. - CLI/Onboarding: MiniMax M2.1 Lightning Auth-Auswahl hinzufügen. - CLI/Onboarding: Schlüsselvorschauen anzeigen, wenn erkannte API-Keys wiederverwendet werden. - Auto-Reply: kompakter `/model`-Picker (Modelle + verfügbare Provider) hinzufügen und Provider-Endpunkte in `/model status` anzeigen. - Control UI: Config-Tab mit Modell-Presets (MiniMax M2.1, GLM 4.7, Kimi) für One-Click-Setup. - Plugins: Erweiterungslader (tools/RPC/CLI/services), Discover-Pfade, Config-Schema + Control UI Labels (uiHints). - Plugins: `clawdbot plugins install` (Path/tgz/npm) hinzufügen, sowie UX für `list|info|enable|disable|doctor`. - Plugins: Voice-Call-Plugin jetzt real (Twilio/Log), start/status RPC/CLI/Tool + Tests hinzufügen. - Docs: Plugins-Dokumentation + Querverweise von Tools/Skills/Gateway-Config. - Docs: Speicherlöschverhalten + schreibbare Arbeitsbereich-Anforderung in Memory/Session/FAQ klären. - Docs: anfängerfreundlichen Plugin-Quickstart hinzufügen + Voice-Call-Plugin-Dokumentation erweitern. - Tests: Docker Plugin-Loader + tgz-Install-Smoketest hinzufügen. - Tests: Docker Plugin E2E erweitern, um Installationen aus lokalen Ordnern (`plugins.load.paths`) und `file:` npm-Spezifikationen abzudecken. - Tests: Abdeckung für Pre-Compaction-Memory-Flush-Einstellungen (einschließlich readonly/CLI-Skip) hinzufügen. - Tests: Modernisierung der Live-Model-Smokeselektion für aktuelle Releases und Durchsetzung von Tools/Images/Thinking-High-Abdeckung. (#769) — thanks @steipete. - Agents/Tools: `apply_patch`-Tool für Multi-Datei-Bearbeitungen (experimentell; gated by tools.exec.applyPatch; OpenAI-only). - Agents/Tools: Bash-Tool in Exec umbenennen (Konfig-Alias beibehalten). (#748) — thanks @myfunc. - Agents: Pre-Compaction-Memory-Flush-Konfiguration (`agents.defaults.compaction.*`) mit soft Threshold + System Prompt hinzufügen. - Config: `$include`-Direktiv für modulare Config-Dateien hinzufügen. (#731) — thanks @pasogott. - Build: Mindestalter für pnpm-Releases auf 2880 Minuten (2 Tage) setzen. (#718) — thanks @dan-dr. - macOS: Aufforderung zur Installation der globalen `clawdbot` CLI, wenn sie im Local-Modus fehlt; Installation über `clawd.bot/install-cli.sh` (kein Onboarding) und externes Launchd/CLI statt eingebettetem Gateway-Runtime verwenden. - Docs: gog Kalender-Ereignis-Farb-IDs aus `gog calendar colors` hinzufügen. (#715) — thanks @mjrussell. - Cron/CLI: `--model`-Flag zu Cron Add/Edit-Befehlen hinzufügen. (#711) — thanks @mjrussell. - Cron/CLI: Modell-Overrides bei Cron-Edits trimmen und Haupt-Sitzungsleitfaden dokumentieren. (#711) — thanks @mjrussell. - Skills: Bündelung von `skill-creator` zur Anleitung beim Erstellen und Verpacken von Skills. - Providers: pro-DM History-Overridess (`dmHistoryLimit`) mit provider-spezifischer Config hinzufügen. (#728) — thanks @pkrmf. - Discord: Kanal/Kategorie-Verwaltungsaktionen im Message-Tool freigeben. (#730) — thanks @NicholasSpisak. - Docs: README “macOS app” Abschnitt in “Apps” umbenennen. (#733) — thanks @AbhisekBasu1. - Gateway: `client.id` in WebSocket-Connect-Paramen verlangen; `client.instanceId` für Presence-Deduplikation verwenden; Docs/Tests aktualisieren. - macOS: Attach-only Gateway-Einstellung entfernen; Local-Modus verwaltet nun Launchd immer noch, falls ein Gateway vorhanden ist. ### Installer - Postinstall: `git apply` durch eingebauten JS-Patcher ersetzen (funktioniert mit npm/pnpm/bun; kein Git-Dependency) плюс Regressionstests. - Postinstall: PNPM-Patch-Fallback überspringen, wenn der neue Patcher aktiv ist. - Installer-Tests: Root+Non-Root Docker Smoke-Tests hinzufügen, CI-Workflow zum Abrufen von clawd.bot-Skripten und Ausführen von Install-Sh/CLI mit übersprungenem Onboarding. - Installer-UX: Unterstützung `CLAWDBOT_NO_ONBOARD=1` für nicht-interaktive Installationen; npm-Prefix unter Linux korrigieren und Git automatisch installieren. - Installer-UX: `install.sh --help` mit Flags/Env und Git-Installationshinweis hinzufügen. - Installer-UX: `--install-method git|npm` hinzufügen und Quell-Checkouts automatisch erkennen (Aufforderung zum Update des Git-Checkouts vs Migration zu npm). ### Fixes - Control UI: Navigation in eine einzige horizontale Scroll-Reihe auf Tablet/Mobile flatten (und immer Collapsed Group Items anzeigen). (#771) — thanks @carlulsoe. - macOS: Starte lokale Gateway-Dienste und warte, bevor der Onboarding-Wizard beginnt. - macOS: Onboarding-Wizard beim Schließen abbrechen, wiederherstellen, falls das Gateway die Session trennt, und Timeouts bei hängenden Gateway-Verbindungen. - macOS: Wizard Debug-CLI zeigt nun FehlStatus statt als abgeschlossen zu beenden. - Models/Onboarding: MiniMax (minimax.io) standardmäßig über Anthropic-kompatiblen `/anthropic`-Endpoint konfigurieren (Beibehaltung von `minimax-api` als Legacy-Alias). - Agents/Browser: Playwright AI-Snapshots für Tool-Aufrufe begrenzen (maxChars); CLI-Snapshots bleiben vollständig. (#763) — thanks @thesash. - Models: Gemini 3 Pro/Flash IDs auf Preview-Namen für Live-Model- Lookups normalisieren. (#769) — thanks @steipete. - CLI: GuardCancel-Typisierung für Configure-Prompts korrigieren. (#769) — thanks @steipete. - Providers: Standard-Gruppenpolitik auf Across-Provider-Allowlist setzen und bei offenen Gruppen in Doctor warnen. - MS Teams: Gruppenpolitik/GroupAllowFrom-Gating für Gruppen-Chats hinzufügen und warnen, wenn Gruppen offen sind. - Providers: Tool-Call/Result-IDs aus Gemini CLI-Payloads entfernen, um API-400s zu vermeiden. (#756) - Gateway/WebChat: Handshake-Validierungsdetails im WebSocket-Close-Reason speichern für einfacheres Debugging; Close-Codes beibehalten. - Gateway/Auth: Ungültige Connect-Antworten vor dem Schließen des Handshakes senden; ungültiger-Connect-Auth-Test stabilisieren. - Gateway: Gateway-Listener-Erkennung verschärfen. - Control UI: Onboarding-Chat ausblenden, wenn konfiguriert, und mobile Chat-Sidebar-Overlay absichern. - Auth: Codex Keychain Credentials lesen und die Lookup plattform-aware machen. - macOS/Release: Verteilungsartefakte in Relay-Builds vermeiden und Appcasts aus zip-only-Quellen generieren. - Doctor: Plugin-Diagnostik im Report

OpenCodev1.1.14

v1.1.14

12. Januar 2026

## Core - Support resource content type in MCP tool output (@antigremlin) - Check worktree for external_directory permission in subdirectories (@dmmulroy) - Bump Copilot plugin version ## TUI - Add clickable navigation to subagent header - Add mouse hover and click support to questions tool (@indeep99) - Fix prompt ref initialization to prevent undefined reference errors - Refactor event streaming in TUI to use SDK instead of manual RPC subscription - Fix cursor position when pasting single-line text in TUI (@Octane0411) - Add mouse hover and click support to autocomplete (@kitlangton) - Copy OAuth URL when device code is unavailable (@itskingori) - Fix slash command autocomplete highlighted row jumping (@kitlangton) ## Desktop - Add image preview support for image attachments in desktop (@dbpolito) **Thank you to 10 community contributors:** - @kitlangton: - fix(tui): slash command autocomplete highlighted row jumping (#7815) - feat(tui): add mouse hover and click support to autocomplete (#7820) - @itskingori: - fix(tui): copy oauth url when no device code (#7812) - @zeke: - docs: fix typos on the providers page (#7829) - docs: fix typos in settings doc (#7892) - @dbpolito: - feat(desktop): Image Preview support for Image Attachments (#7841) - feat(desktop): User Message Badges on Hover (#7835) - @jacobwgillespie: - chore(lander): fix spacing in desktop app banner (#7822) - @dmmulroy: - fix: check worktree for external_directory permission in subdirs (#7811) - @Octane0411: - Fix: TUI single-line paste cursor position (#7277) - @indeep99: - feat(tui): add mouse hover and click support to questions tool (#7905) - @antigremlin: - fix(mcp): support `resource` content type in MCP tool output (#7879) - @eyalch: - docs: fix scroll_speed default value (#7867)

Windsurfv1.13.6

Agenten-Fähigkeiten für Cascade

12. Januar 2026

Windsurf unterstützt jetzt Agent Skills für Cascade.

OpenCodev1.1.13

v1.1.13

11. Januar 2026

## Core - Ensure Codex built-in plugin is always available - Ensure /connect works for OpenAI Business plans by adding missing field - Track file modifications in ACP (@wid4t) - Pass sessionID to chat.system.transform (@spoons-and-mirrors) - Extract external directory validation to shared utility to reduce code duplication across tools - Exclude 'none' variant for gpt-5.2-codex (@Alejandro-CSt) - Follow symlinks by default in ripgrep searches (@kdcokenny) - Disable telemetry in the PHP language server (@madflow) ## TUI - Move props.ref to onMount in TUI component (@itsrainingmani) - Fixed command palette incorrectly highlighting items on mouse hover (@kitlangton) ## Desktop - Add ability to fork sessions in desktop app (@dbpolito) - Make mouse hover and click work on prompt input autocomplete (@dbpolito) - Add missing StoreExt import and fix unused mut warning in desktop (@jerome-benoit) - Fix missing import causing desktop build to fail (@Leka74) - Support configuring default server URL for desktop (@athal7) - Remove split operation in serverDisplayName function (@OpeOginni) **Thank you to 13 community contributors:** - @madflow: - fix: disable telemetry in the php lsp server (#7649) - @dbpolito: - fix(desktop): Allow Selecting Card Errors (#7506) - feat(deskop): Add Copy to Messages (#7658) - feat(desktop): Make mouse hover / click work on prompt input autocomplete (#7661) - feat(desktop): Fork Session (#7673) - @zerone0x: - fix(ui): allow text editing shortcuts in search dialogs on macOS (#7419) - @OpeOginni: - fix(desktop): remove split operation in serverDisplayName function (#7397) - @athal7: - feat: support configuring default server URL for desktop (#7363) - @kdcokenny: - fix(grep): follow symlinks by default in ripgrep searches (#7501) - @Alejandro-CSt: - fix: exclude 'none' variant for gpt-5.2-codex (#7682) - @Leka74: - fix(desktop): build failing - missing import (#7697) - @jerome-benoit: - fix(desktop): add missing StoreExt import and fix unused mut warning (#7707) - @spoons-and-mirrors: - feat: pass sessionID to chat.system.transform (#7718) - @kitlangton: - fix(opencode): command palette mouse hover highlights wrong item (#7721) - @wid4t: - feat(acp): track file modified (#7723) - @itsrainingmani: - fix(tui): move props.ref to onMount (#7702)

OpenClawv2026.1.10

clawdbot 2026.1.10

11. Januar 2026

### Highlights - CLI: `clawdbot status` jetzt tabelliert + zeigt OS/Update/Gateway/Daemon/Agents/Sessions; `status --all` fügt einen vollständigen read-only Debug-Bericht hinzu ( Tabellen, Log-Tails, Tailscale-Zusammenfassung und Scan-Fortschritt über OSC-9 + Spinner). - CLI Backends: Kodex-CLI-Fallback mit Resume-Unterstützung (Textausgabe) und JSONL-Parsing für neue Runs, plus eine Live CLI-Resume-Probe. - CLI: füge `clawdbot update` (Safe-ish Git-Checkout-Update) + `--update` Kurzform hinzu. (#673) — danke an @fm1randa. - Gateway: füge OpenAI-kompatiblen HTTP-Endpunkt `/v1/chat/completions` hinzu (Auth, SSE-Streaming, pro-Agent-Routing). (#680) — danke an @steipete. ### Changes - Onboarding/Models: erstklassige Z.AI (GLM) Auth-Auswahl hinzufügen (`zai-api-key`) + `--zai-api-key` Flag. - CLI/Onboarding: OpenRouter API-Key-Auth-Option in configure/onboard hinzufügen. (#703) — danke an @mteam88. - Agents: menschliche Verzögerungs-Tempo zwischen Block-Antworten hinzufügen (Modi: aus/natürlich/kundenspezifisch, pro-Agent konfigurierbar). (#446) — danke an @tony-freedomology. - Agents/Browser: `browser.target` (Sandbox/Host/Custom) mit Sandbox-Host-Control-Gating über `agents.defaults.sandbox.browser.allowHostControl`, Whitelists für benutzerdefinierte Kontroll-URLs/Hosts/Ports, und Erweiterung der Browser-Tool-Dokumentation (Remote-Control, Profiles, Internals). - Onboarding/Models: katalog-gestützter Default-Modell-Auswahl-Assistent beim Onboarding + Configure. (#611) — danke an @jonasjancarik. - Agents/OpenCode Zen: Fallback-Modelle + Defaults aktualisieren, Legacy-Alias-Mappings beibehalten. (#669) — danke an @magimetal. - Providers: gruppenübergreifende History-Kontext-Wrapper über Providers hinweg vereinheitlichen mit pro-Provider/pro-Account `historyLimit` Overrides (Fallback zu `messages.groupChat.historyLimit`). Setze `0` zum Deaktivieren. (#672) — danke an @steipete. - Gateway/Heartbeat: optional Heartbeat-Ausgabe `Reasoning:` liefern (`agents.defaults.heartbeat.includeReasoning`). (#690) - Docker: optionale Home-Volume + zusätzliche Bind-Mounts in `docker-setup.sh` zulassen. (#679) — danke an @gabriel-trigo. ### Fixes - CLI/Status: Tabellen auf volle Terminalbreite erweitern; Klarstellung bei Provider-Setup vs Runtime-Warnungen; reichhaltigere Provider-Details; Token-Vorschau in `status` bei Redaction in `status --all`; Troubleshooting-Link-Footer hinzufügen; Log-Tails kopierbar behalten; Gateway-Auth bei Erreichbarkeit anzeigen; Provider-Laufzeitfehler (Signal/iMessage/Slack) surfaces; harte Verbesserung der `tailscale status --json`-Parserung; `status --all` Scan-Fortschritt determiniert gestalten; Footer durch eine 3-zeilige “Nächste Schritte”-Empfehlung ersetzen (teilen/debug/probe). - CLI/Gateway: klarstellen, dass `clawdbot gateway status` RPC-Gesundheit (Connect + RPC) meldet und RPC-Fehler separat von Verbindungsfehlern zeigt. - CLI/Update: Fortschritt-Spinner an stdout TTY binden und Label des Clean-Check-Schritts angleichen. (#701) — danke an @bjesuiter. - Telegram: `/whoami` + `/id` Befehle hinzufügen, um Absender-ID für Allowlists offenzulegen; `@username` und prefixed IDs in `allowFrom`-Prompts zulassen (mit Stabilitäts-Warnung). - Control UI: verhindern, dass automatisch `telegram.groups["*"]` geschrieben wird; Warnung/Bestätigung vor Aktivierung von Wildcard-Gruppen. - WhatsApp: Ack-Reaktionen nur für bearbeitete Nachrichten senden und Legacy `messages.ackReaction` ignorieren (Arzt-Kopien in `whatsapp.ackReaction`). (#629) — danke an @pasogott. - Sandbox/Skills: Skills in Sandbox-Arbeitsbereiche duplizieren (Read-Only-Mounts), damit SKILL.md zugänglich bleibt. - Terminal/Table: ANSI-sichere Wraps zur Vermeidung von Tabellen-Clipping/Color-Verlust; Regression-Abdeckung hinzufügen. - Docker: optionale apt-Pakete während Build zulassen und Build-Arg dokumentieren. (#697) — danke an @gabriel-trigo. - Gateway/Heartbeat: Reasoning auch liefern, wenn die Haupt-Heartbeat-Antwort `HEARTBEAT_OK` ist. (#694) — danke an @antons. - macOS: unsigned launchd-Overrides bei signierten Neustarts löschen und beim Attach-Modus/Deaktivierung-Marker Doctor-mäßig warnen. (#695) — danke an @jeffersonwarrior. - Agents: Ein-Writers-Sessions-Locks erzwingen und orphan Tool-Ergebnisse droppen, um Fehler bei Tool-Call IDs zu verhindern (MiniMax/Anthropic-kompatible APIs). - Docs: `clawdbot status` als ersten Diagnostik-Schritt festlegen, Verhalten von `status --deep` klären und `/whoami` + `/id` dokumentieren. - Docs/Testing: Live-Tools+Image-Probes klären und auflisten, welche `provider/model`-IDs testbar sind. - Tests/Live: Gateway-Bash+Read-Probes robust gegen Provider-Formatierungen machen, während reale Tool-Aufrufe weiter validiert werden. - WhatsApp: Erkennung von @lid Erwähnungen in Gruppen via authDir-Reverse-Mapping + Auflösung der eigenen JID E.164 für Filterung von Erwähnungen. (#692) — danke an @peschee. - Gateway/Auth: Standardmäßig Token-Auth auf Loopback während Onboarding, Doctor-Token-Generierungs-Flow hinzufügen, Audio-Transkriptions-Konfig zu Whisper-only straffen. - Providers: eingehende Nachrichten über Provider hinweg deduplizieren, um doppelte LLM-Durchläufe bei erneuter Lieferung/Neuverbindung zu vermeiden. (#689) — danke an @adam91holt. - Agents: `<thought>`/`<antthinking>`-Tags aus versteckten Denkprozessen entfernen und Tag-Varianten in Tests abdecken. (#688) — danke an @theglove44. - macOS: Modell-Auswahl-Selektoren als normalisierte Provider/Model-IDs speichern und manuelle Einträge angleichen. (#683) — danke an @benithors. - Agents: Erkenne „usage limit“-Fehler als Rate Limits für Failover. (#687) — danke an @evalexpr. - CLI: Erfolgsnachrichten vermeiden, wenn Daemon-Neustart übersprungen wird. (#685) — danke an @carlulsoe. - Commands: Standardmäßig `/config` + `/debug` deaktivieren; via `commands.config`/`commands.debug` absichern und aus native Registration/Hilfe-Ausgabe ausblenden. - Agents/System: Klarstellen, dass Sub-Agents sandboxed bleiben und keinen erhöhten Host-Zugriff verwenden können. - Gateway: OpenAI-kompatiblen `/v1/chat/completions` Endpunkt standardmäßig deaktivieren; über `gateway.http.endpoints.chatCompletions.enabled=true` aktivieren. - macOS: Bridge-Tunnels stabilisieren, Invoke-Sender bei Disconnect schützen, stdout/stderr abfließen lassen, um Deadlocks zu vermeiden. (#676) — danke an @ngutman. - Agents/System: Klarstellen, dass Sandbox-Runtime im System-Prompt vorhanden ist und erhöhte Verfügbarkeit bei Sandboxing anzeigen. - Auto-reply: Bevorzugen von `RawBody` zur Parsing von Commands/Direktiven (WhatsApp + Discord) und verhindern, dass Fallback-Läufe gleichzeitige Session-Updates überschreiben. (#643) — danke an @mcinteerj. - WhatsApp: Gruppen-Reaktionen durch Beibehaltung von Nachrichten-IDs und Sender-JIDs in der Verlaufshistorie korrigieren; Teilnehmer-Telefonnummern zu JIDs in ausgehenden Reaktionen normalisieren. (#640) — danke an @mcinteerj. - WhatsApp: Gruppen-Teilnehmer-IDs dem Modell zugänglich machen, damit Reaktionen den richtigen Sender adressieren können. - Cron: `wakeMode: "now"` wartet auf Abschluss des Heartbeats (undRetry, wenn die Hauptspur beschäftigt ist). (#666) — danke an @roshanasingh4. - Agents/OpenAI: Fehlerbehebung bei Responses-Tool-nur → Folge-Turn-Behandlung (vermeiden eigenständiger `reasoning`-Items, die 400 “required following item” auslösen) und Reasoning-Items in Responses/Codex-Responses-Verlauf für tool-call-only Turns erneut abspielen. - Sandbox: füge `clawdbot sandbox explain

Claude Codev2.1.4

v2.1.4

11. Januar 2026

## What's changed - Added `CLAUDE_CODE_DISABLE_BACKGROUND_TASKS` environment variable to disable all background task functionality including auto-backgrounding and the Ctrl+B shortcut - Fixed "Help improve Claude" setting fetch to refresh OAuth and retry when it fails due to a stale OAuth token

OpenCodev1.1.12

v1.1.12

10. Januar 2026

## Core - Add ChatGPT-Account-Id header for organization subscriptions (@fdematos) **Thank you to 2 community contributors:** - @vtemian: - docs(ecosystem): add micode and octto plugins (#7327) - @fdematos: - fix: add ChatGPT-Account-Id header for organization subscriptions (#7603)

OpenClawv2026.1.9

clawdbot 2026.1.9

10. Januar 2026

### Highlights - Microsoft Teams provider: polling, attachments, outbound CLI send, per-channel policy. - Models/Auth expansion: OpenCode Zen + MiniMax API onboarding; token auth profiles + auth order; OAuth health in doctor/status. - CLI/Gateway UX: message subcommands, gateway discover/status/SSH, /config + /debug, sandbox CLI. - Provider reliability sweep: WhatsApp contact cards/targets, Telegram audio-as-voice + streaming, Signal reactions, Slack threading, Discord stability. - Auto-reply + status: block-streaming controls, reasoning handling, usage/cost reporting. - Control UI/TUI: queued messages, session links, reasoning view, mobile polish, logs UX. ### Breaking - CLI: `clawdbot message` now subcommands (`message send|poll|...`) and requires `--provider` unless only one provider configured. - Commands/Tools: `/restart` und gateway restart tool disabled by default; enable with `commands.restart=true`. ### New Features and Changes - Models/Auth: OpenCode Zen onboarding (#623) — thanks @magimetal; MiniMax Anthropic-compatible API + hosted onboarding (#590, #495) — thanks @mneves75, @tobiasbischoff. - Models/Auth: setup-token + token auth profiles; `clawdbot models auth order {get,set,clear}`; per-agent auth candidates in `/model status`; OAuth expiry checks in doctor/status. - Agent/System: claude-cli runner; `session_status` tool (und sandbox allowed); adaptive context pruning default; system prompt messaging guidance + no auto self-update; eligible skills list injection; sub-agent context trimmed. - Commands: `/commands` list; `/models` alias; `/usage` alias; `/debug` runtime overrides + effective config view; `/config` chat updates + `/config get`; `config --section`. - CLI/Gateway: unified message tool + message subcommands; gateway discover (local + wide-area DNS-SD) with JSON/timeout; gateway status human-readable + JSON + SSH loopback; wide-area records include gatewayPort/sshPort/cliPath + tailnet DNS fallback. - CLI UX: logs output modes (pretty/plain/JSONL) + colorized health/daemon output; global `--no-color`; lobster palette in onboarding/config. - Dev ergonomics: gateway `--dev/--reset` + dev profile auto-config; C-3PO dev templates; dev gateway/TUI helper scripts. - Sandbox/Workspace: sandbox list/recreate commands; sync skills into sandbox workspace; sandbox browser auto-start. - Config/Onboarding: inline env vars; OpenAI API key flow to shared `~/.clawdbot/.env`; Opus 4.5 default prompt for Anthropic auth; QuickStart auto-install gateway (Node-only) + provider picker tweaks + skip-systemd flags; TUI bootstrap prompt (`tui --message`); remove Bun runtime choice. - Providers: Microsoft Teams provider (polling, attachments, outbound sends, requireMention, config reload/DM policy). (#404) — thanks @onutc - Providers: WhatsApp broadcast groups for multi-agent replies (#547) — thanks @pasogott; inbound media size cap configurable (#505) — thanks @koala73; identity-based message prefixes (#578) — thanks @p6l-richard. - Providers: Telegram inline keyboard buttons + callback payload routing (#491) — thanks @azade-c; cron topic delivery targets (#474/#478) — thanks @mitschabaude-bot, @nachoiacovino; `[[audio_as_voice]]` tag support (#490) — thanks @jarvis-medmatic. - Providers: Signal reactions + notifications with allowlist support. - Status/Usage: /status cost reporting + `/cost` lines; auth profile snippet; provider usage windows. - Control UI: mobile responsiveness (#558) — thanks @carlulsoe; queued messages + Enter-to-send (#527) — thanks @YuriNachos; session links (#471) — thanks @HazAT; reasoning view; skill install feedback (#445) — thanks @pkrmf; chat layout refresh (#475) — thanks @rahthakor; docs link + new session button; drop explicit `ui:install`. - TUI: agent picker + agents list RPC; improved status line. - Doctor/Daemon: audit/repair flows, permissions checks, supervisor config audits; provider status probes + warnings for Discord intents and Telegram privacy; last activity timestamps; gateway restart guidance. - Docs: Hetzner Docker VPS guide + cross-links (#556/#592) — thanks @Iamadig; Ansible guide (#545) — thanks @pasogott; provider troubleshooting index; hook parameter expansion (#532) — thanks @mcinteerj; model allowlist notes; OAuth deep dive; showcase refresh. - Apps/Branding: refreshed iOS/Android/macOS icons (#521) — thanks @fishfisher. ### Fixes - Packaging: include MS Teams send module in npm tarball. - Sandbox/Browser: auto-start CDP endpoint; proxy CDP out of container for attachOnly; relax Bun fetch typing; align sandbox list output with config images. - Agents/Runtime: gate heartbeat prompt to default sessions; /stop aborts between tool calls; require explicit system-event session keys; guard small context windows; fix model fallback stringification; sessions_spawn inherits provider; failover on billing/credits; respect auth cooldown ordering; restore Anthropic OAuth tool dispatch + tool-name bypass; avoid OpenAI invalid reasoning replay; harden Gmail hook model defaults. - Agent history/schema: strip/skip empty assistant/error blocks to prevent session corruption/Claude 400s; scrub unsupported JSON Schema keywords + sanitize tool call IDs for Cloud Code Assist; simplify Gemini-compatible tool/session schemas; require raw for config.apply. - Auto-reply/Streaming: default audioAsVoice false; preserve audio_as_voice propagation + buffer audio blocks + guard voice notes; block reply ordering (timeout) + forced-block fence-safe; avoid chunk splits inside parentheses + fence-close breaks + invalid UTF-16 truncation; preserve inline directive spacing + allow whitespace in reply tags; filter NO_REPLY prefixes + normalize routed replies; suppress <think> leakage with separate Reasoning; block streaming defaults (off by default, minChars/idle tuning) + coalesced blocks; dedupe followup queue; restore explicit responsePrefix default. - Status/Commands: provider prefix in /status model display; usage filtering + provider mapping; auth label + usage snapshots (claude-cli fallback + optional claude.ai); show Verbose/Elevated only when enabled; compact usage/cost line + restore emoji-rich status; /status in directive-only + multi-directive handling; mention-bypass elevated handling; surface provider usage errors; wire /usage to /status; restore hidden gateway-daemon alias; fallback /model list when catalog unavailable. - WhatsApp: vCard/contact cards (prefer FN, include numbers, show all contacts, keep summary counts, better empty summaries); preserve group JIDs + normalize targets; resolve @lid mappings/JIDs (Baileys/auth-dir) + inbound mapping; route queued replies to sender; improve web listener errors + remove provider name from errors; record outbound activity account id; fix web media fetch errors; broadcast group history consistency. - Telegram: keep streamMode draft-only; long-poll conflict retries + update dedupe; grammY fetch mismatch fixes + restrict native fetch to Bun; suppress getUpdates stack traces; include user id in pairing; audio_as_voice handling fixes. - Discord/Slack: thread context helpers + forum thread starters; avoid category parent overrides; gateway reconnect logs + HELLO timeout + stop provider after reconnect exhaustion; DM recipient parsing for numeric IDs; remove incorrect limited warning; reply threading + mrkdwn edge cases; remove ack reactions after reply; gateway debug event visibility. - Signal: reaction handling safety; own-reaction matching (uuid+phone); UUID-only senders accepted; ignore reaction-only messages. - MS Teams: download image attachments reliably; fix top-level replies; stop on shutdown + honor chunk limits; normalize poll providers/deps; pairing label fixes. - iMessage: isolate group-ish threads by chat_id. - Gateway/Daemon/Doctor: atomic config writes; repair gateway service entrypoint + install switches; non-interactive legacy migrations; systemd unit alignment + KillMode=process; node bridge keepalive/pings; Launch at Login persistence; bundle ClawdbotKit resources + Swift 6.2 compat dylib; relay version check + remove smoke test; regen Swift GatewayModels + keep agent provider string; cron jobId alias + channel alias migration + main session key normalization; heartbeat Telegram accountId resolution; avoid WhatsApp fallback for internal runs; gateway listener error wording; serveBaseUrl param; honor gateway --dev; fix wide-area discovery updates; align agents.defaults schema; provider account metadata in daemon status; refresh Carbon patch for gateway fixes; restore doctor prompter initialValue handling. - Control UI/TUI: persist per-session verbose off + hide tool cards; logs tab opens at bottom; relative asset paths + landing cleanup; session labels lookup/persistence; stop pinning main session in recents; start logs at bottom; TUI status bar refresh + timeout handling + hide reasoning label when off. - Onboarding/Configure: QuickStart single-select provider picker; avoid Codex CLI false-expiry warnings; clarify WhatsApp owner prompt; fix Minimax hosted onboarding (agents.defaults + msteams heartbeat target); remove configure Control UI prompt; honor gateway --dev flag. ### Maintenance - Dependencies: bump pi-* stack

OpenCodev1.1.11

v1.1.11

10. Januar 2026

## Core - Ensure GPT-5.2-codex has variants - Adjust Codex styling and fix hint display - Update docs and authentication methods for OpenAI - Improve Codex instructions prompt for better task planning - Add Codex authentication support ## TUI - Fix instance disposal to prevent memory leaks and dangling references **Thank you to 1 community contributor:** - @spoons-and-mirrors: - docs: add url based instructions to web docs (#7216)

Continue.devv1.5.30

CLI Stable Release v1.5.30

10. Januar 2026

Direktes stabiles Release aus dem Main-Branch. Version 1.5.30 wurde direkt aus dem neuesten Main-Branch gebaut und veröffentlicht.

Claude Codev2.1.3

v2.1.3

9. Januar 2026

## What's changed - Merged slash commands and skills, simplifying the mental model with no change in behavior - Added release channel (`stable` or `latest`) toggle to `/config` - Added detection and warnings for unreachable permission rules, with warnings in `/doctor` and after saving rules that include the source of each rule and actionable fix guidance - Fixed plan files persisting across `/clear` commands, now ensuring a fresh plan file is used after clearing a conversation - Fixed false skill duplicate detection on filesystems with large inodes (e.g., ExFAT) by using 64-bit precision for inode values - Fixed mismatch between background task count in status bar and items shown in tasks dialog - Fixed sub-agents using the wrong model during conversation compaction - Fixed web search in sub-agents using incorrect model - Fixed trust dialog acceptance when running from the home directory not enabling trust-requiring features like hooks during the session - Improved terminal rendering stability by preventing uncontrolled writes from corrupting cursor state - Improved slash command suggestion readability by truncating long descriptions to 2 lines - Changed tool hook execution timeout from 60 seconds to 10 minutes - [VSCode] Added clickable destination selector for permission requests, allowing you to choose where settings are saved (this project, all projects, shared with team, or session only)

OpenCodev1.1.10

v1.1.10

9. Januar 2026

## Core - Disable server unless explicitly opted in - Report config errors instead of silent termination (@ariane-emory) - Upgrade OpenTUI to v0.1.72 - Resolve broken forked sessions with compactions due to missing parent-child message references (@ryanwyler) - Upgrade OpenTUI to v0.1.71, fixing capability detection and Kitty keyboard support - Deduplicate plugins by name with priority-based resolution (@devxoul) - Bump Copilot plugin version - Fix model dialog search issues in favorites/recents list and prevent deprecated models from appearing ## TUI - Enable API key prompt for Bedrock in auth login (@Raviguntakala) ## Desktop - Remove inline JavaScript from the app - Remove custom URL parameter from app - Prevent model picker overflow with Zen mode in new sessions (@0xajka) - Complete symlink support implementation and enable CI for PRs in desktop app (@jerome-benoit) - Store terminal and review pane visibility per session - Change terminal cursor to bar and stop blinking when not focused (@dbpolito) **Thank you to 8 community contributors:** - @RD201: - docs: Add Scaleway to provider docs (#7389) - @dbpolito: - feat(desktop): Change Terminal's Cursor to Bar and Stop Blinking when not focused (#7402) - @devxoul: - feat(config): deduplicate plugins by name with priority-based resolution (#5957) - @jerome-benoit: - fix(desktop): complete symlink support implementation and enable CI for PRs (#7113) - @0xajka: - fix: prevent model picker overflow with Zen in new sessions (#7495) - @ryanwyler: - fix: resolve broken forked sessions with compactions due to missing parent-child message references (#6445) - @Raviguntakala: - fix(cli): enable API key prompt for Bedrock in auth login (#7332) - @ariane-emory: - fix: report config errors instead of silent termination (#7522)

Codex CLIvrust-v0.80.0

0.80.0

9. Januar 2026

## New Features - Add conversation/thread fork endpoints in the protocol and app server so clients can branch a session into a new thread. (#8866) - Expose requirements via `requirement/list` so clients can read `requirements.toml` and adjust agent-mode UX. (#8800) - Introduce metrics capabilities with additional counters for observability. (#8318, #8910) - Add elevated sandbox onboarding with prompts for upgrade/degraded mode plus the `/elevate-sandbox` command. (#8789) - Allow explicit skill invocations through v2 API user input. (#8864) ## Bug Fixes - Codex CLI subprocesses again inherit env vars like `LD_LIBRARY_PATH`/`DYLD_LIBRARY_PATH` to avoid runtime issues. As explained in #8945, failure to pass along these environment variables to subprocesses that expect them (notably GPU-related ones), was causing 10x+ performance regressions! Special thanks to @johnzfitch for the detailed investigation and write-up in #8945. (#8951) - `/review <instructions>` in TUI/TUI2 now launches the review flow instead of sending plain text. (#8823) - Patch approval “allow this session” now sticks for previously approved files. (#8451) - Model upgrade prompt now appears even if the current model is hidden from the picker. (#8802) - Windows paste handling now supports non-ASCII multiline input reliably. Special thanks to @occurrent laying the groundwork for this fix in #8021! (#8774) - Git apply path parsing now handles quoted/escaped paths and `/dev/null` correctly to avoid misclassified changes. (#8824) ## Documentation - App-server README now documents skills support and usage. (#8853) - Skill-creator docs clarify YAML frontmatter formatting and quoting rules. (#8610) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.79.0...rust-v0.80.0 - #8734 fix: do not propose to add multiline commands to execpolicy @tibo-openai - #8802 Enable model upgrade popup even when selected model is no longer in picker @charley-oai - #8805 chore: stabilize core tool parallelism test @tibo-openai - #8820 chore: silent just fmt @jif-oai - #8824 fix: parse git apply paths correctly @tibo-openai - #8823 fix: handle /review arguments in TUI @tibo-openai - #8822 chore: rename unified exec sessions @jif-oai - #8825 fix: handle early codex exec exit @tibo-openai - #8830 chore: unify conversation with thread name @jif-oai - #8840 Move tests below auth manager @pakrym-oai - #8845 fix: upgrade lru crate to 0.16.3 @bolinfest - #8763 Merge Modelfamily into modelinfo @aibrahim-oai - #8842 remove unnecessary todos @aibrahim-oai - #8846 Stop using AuthManager as the source of codex_home @pakrym-oai - #8844 Fix app-server `write_models_cache` to treat models with less priority number as higher priority. @aibrahim-oai - #8850 chore: drop useless feature flags @jif-oai - #8848 chore: drop some deprecated @jif-oai - #8853 [chore] update app server doc with skills @celia-oai - #8451 fix: implement 'Allow this session' for apply_patch approvals @owenlin0 - #8856 Override truncation policy at model info level @aibrahim-oai - #8849 Simplify error managment in `run_turn` @aibrahim-oai - #8767 Add feature for optional request compression @cconger - #8610 Clarify YAML frontmatter formatting in skill-creator @darlingm - #8847 Warn in /model if BASE_URL set @gt-oai - #8801 Support symlink for skills discovery. @xl-openai - #8800 Feat: appServer.requirementList for requirement.toml @shijie-oai - #8861 fix: update resource path resolution logic so it works with Bazel @bolinfest - #8868 fix: use tokio for I/O in an async function @bolinfest - #8867 add footer note to TUI @iceweasel-oai - #8879 feat: introduce find_resource! macro that works with Cargo or Bazel @bolinfest - #8864 Support UserInput::Skill in V2 API. @xl-openai - #8876 add ability to disable input temporarily in the TUI. @iceweasel-oai - #8884 fix: make the find_resource! macro responsible for the absolutize() call @bolinfest - #8774 fix: windows can now paste non-ascii multiline text @dylan-hurd-oai, @occurrent - #8855 chore: add list thread ids on manager @jif-oai - #8318 feat: metrics capabilities @jif-oai - #8826 fix: stabilize list_dir pagination order @tibo-openai - #8892 chore: drop metrics exporter config @jif-oai - #8896 chore: align error limit comment @tibo-openai - #8899 fix: include project instructions in /review subagent @tibo-openai - #8894 chore: add small debug client @jif-oai - #8888 fix: leverage find_resource! macro in load_sse_fixture_with_id @bolinfest - #8691 Avoid setpgid for inherited stdio on macOS @seeekr - #8887 fix: leverage codex_utils_cargo_bin() in codex-rs/core/tests/suite @bolinfest - #8907 chore: drop useless interaction_input @jif-oai - #8903 nit: drop unused function call error @jif-oai - #8910 feat: add a few metrics @jif-oai - #8911 gitignore bazel-* @zbarsky-openai - #8843 config requirements: improve requirement error messages @gt-oai - #8914 fix: reduce duplicate include_str!() calls @bolinfest - #8902 feat: add list loaded threads to app server @jif-oai - #8870 [fix] app server flaky thread/resume tests @celia-oai - #8916 clean: all history cloning @jif-oai - #8915 otel test: retry WouldBlock errors @gt-oai - #8792 Update models.json @github-actions - #8897 fix: preserve core env vars on Windows @tibo-openai - #8913 Add `read-only` when backfilling requirements from managed_config @gt-oai - #8926 add tooltip hint for shell commands (!) @fps7806 - #8857 Immutable CodexAuth @pakrym-oai - #8927 nit: parse_arguments @jif-oai - #8932 fix: increase timeout for tests that have been flaking with timeout issues @bolinfest - #8931 fix: correct login shell mismatch in the accept_elicitation_for_prompt_rule() test @bolinfest - #8874 [fix] app server flaky send_messages test @celia-oai - #8866 feat: fork conversation/thread @apanasenko-oai - #8858 remove `get_responses_requests` and `get_responses_request_bodies` to use in-place matcher @aibrahim-oai - #8939 [chore] move app server tests from chat completion to responses @celia-oai - #8880 Attempt to reload auth as a step in 401 recovery @pakrym-oai - #8946 fix: increase timeout for wait_for_event() for Bazel @bolinfest - #8789 Elevated sandbox NUX @iceweasel-oai - #8917 fix: treat null MCP resource args as empty @tibo-openai - #8942 Add 5s timeout to models list call + integration test @aibrahim-oai - #8951 fix: remove existing process hardening from Codex CLI @bolinfest

OpenCodev1.1.8

v1.1.8

9. Januar 2026

## Core - Upgrade OpenTUI to v0.1.70 - Support multi-select questions - Document cargofmt in docs (@marc2332) ## TUI - Stop escape key propagation from dialogs - Use full file path for fuzzy matching in autocomplete (@bagatka) ## Desktop - Revert fix for opening external links in default browser - Cache session-scoped stores and add optional context gating - Incrementally render turns with markdown caching and lazy diff rendering - Gate hydration side-effects and handoff while stores load - Prefetch adjacent sessions in the app - Implement chunked message loading and lazy diff loading for improved performance - Fix project sidebar folder name display issue (@0xajka) **Thank you to 4 community contributors:** - @marc2332: - fix(docs): Document `cargofmt` (#7383) - @bagatka: - fix(ui): use full file path for fuzzy matching in autocomplete (#6705) - @0xajka: - Fix for #7229 Fix/project sidebar folder name (#7232) - @Anthuang: - feat(desktop): add Vesper theme (#7331)

Claude Codev2.1.2

v2.1.2

9. Januar 2026

## What's changed - Added source path metadata to images dragged onto the terminal, helping Claude understand where images originated - Added clickable hyperlinks for file paths in tool output in terminals that support OSC 8 (like iTerm) - Added support for Windows Package Manager (winget) installations with automatic detection and update instructions - Added Shift+Tab keyboard shortcut in plan mode to quickly select "auto-accept edits" option - Added `FORCE_AUTOUPDATE_PLUGINS` environment variable to allow plugin autoupdate even when the main auto-updater is disabled - Added `agent_type` to SessionStart hook input, populated if `--agent` is specified - Fixed a command injection vulnerability in bash command processing where malformed input could execute arbitrary commands - Fixed a memory leak where tree-sitter parse trees were not being freed, causing WASM memory to grow unbounded over long sessions - Fixed binary files (images, PDFs, etc.) being accidentally included in memory when using `@include` directives in CLAUDE.md files - Fixed updates incorrectly claiming another installation is in progress - Fixed crash when socket files exist in watched directories (defense-in-depth for EOPNOTSUPP errors) - Fixed remote session URL and teleport being broken when using `/tasks` command - Fixed MCP tool names being exposed in analytics events by sanitizing user-specific server configurations - Improved Option-as-Meta hint on macOS to show terminal-specific instructions for native CSIu terminals like iTerm2, Kitty, and WezTerm - Improved error message when pasting images over SSH to suggest using `scp` instead of the unhelpful clipboard shortcut hint - Improved permission explainer to not flag routine dev workflows (git fetch/rebase, npm install, tests, PRs) as medium risk - Changed large bash command outputs to be saved to disk instead of truncated, allowing Claude to read the full content - Changed large tool outputs to be persisted to disk instead of truncated, providing full output access via file references - Changed `/plugins` installed tab to unify plugins and MCPs with scope-based grouping - Deprecated Windows managed settings path `C:\ProgramData\ClaudeCode\managed-settings.json` - administrators should migrate to `C:\Program Files\ClaudeCode\managed-settings.json` - [SDK] Changed minimum zod peer dependency to ^4.0.0 - [VSCode] Fixed usage display not updating after manual compact

OpenCodev1.1.7

v1.1.7

8. Januar 2026

## Core - Add kind, title, and rawInput fields to ACP tool_call_update events (@noamzbr) - Add help text to debug command and subcommands (@assagman) - Allow disabling .claude prompt and skills loading (@freespace8) - Add interactive question tool for gathering user preferences and clarifying instructions - Fix issue with normal transform options conflicting with small model options when generating title - Write truncated tool outputs to files - Simplify task tool subagent filtering (@malhashemi) - Improved title generation to avoid repetitive "Analyzing..." prefixes - Use correct authorization server URL for OAuth discovery in MCP (@rscarvalho) - Config precedence now correctly allows local config to override remote (@elithrar) - Update Homebrew formula for anomalyco tap ## TUI - Restore showDetails check removed in Permission rework (@ryanwyler) - Cleanly restore terminal state on fatal error exit (@itsrainingmani) - Improve responsive layout by hiding header and footer when sidebar is visible - Simplified session settings management (@maxstevens-nl) - Remove unused imports and variables from header component - Add overlay sidebar for narrow screens to improve mobile experience - Add expandable bash output for long commands to improve readability - Remove username visibility toggle and simplify user message metadata display in TUI - Improve indentation and label clarity for TUI answer options - Ensure forked message text is inserted in prompt when forking conversations (@ariane-emory) - Use selected foreground color for permission selection text (@ariane-emory) - Constrain autocomplete height to available screen space in TUI (@Raviguntakala) - Add selectedListItemText to orange themes for button text visibility (@ThanhNguyxn) ## Desktop - Show custom models without valid release_date in web UI model selector (@zerone0x) - Open external links in default browser on desktop (@athal7) **Thank you to 15 community contributors:** - @elithrar: - fix: config precedence now correctly allows local config to override remote (#7141) - @ThanhNguyxn: - fix(theme): add selectedListItemText to orng themes for button text visibility (#7169) - @rscarvalho: - fix(mcp): use correct authorization server URL for OAuth discovery (#7234) - @Raviguntakala: - fix(tui): constrain autocomplete height to available screen space (#7181) - @malhashemi: - refactor: simplify task tool subagent filtering (#7165) - @ariane-emory: - fix: use selectedForeground's computer colour (or theme's selectedForeground value) for the colour of text in permission selection (resolves #7246) (#7251) - fix(tui): ensure forked message text is inserted in prompt (resolves #7257) (#7259) - @athal7: - fix(desktop): open external links in default browser (#7221) - @freespace8: - feat(flags): allow disabling .claude prompt and skills loading (#7205) - @aaroniker: - feat: Polish dialog, popover shadows & borders, add missing provider logos (#7326) - @assagman: - fix(cli): add help text to debug command and subcommands (#7328) - @maxstevens-nl: - refactor: session settings simplification (#7342) - @noamzbr: - fix: Add kind, title, rawInput to ACP tool_call_update events (#7368) - @zerone0x: - fix(app): show custom models without valid release_date in web UI mod… (#7349) - @itsrainingmani: - fix: cleanly restore terminal state on fatal error exit (#7359) - @ryanwyler: - fix(tui): restore showDetails check removed in Permission rework (#7285)

OpenClawv2026.1.8

clawdbot 2026.1.8

8. Januar 2026

## 2026.1.8 ### Highlights - Security: DMs standardmäßig über alle Anbieter hinweg eingeschränkt; pairing-first + Anwhitelist-Anleitung. - Sandbox: standardisierte Bereichsgrenzen pro Agent + Arbeitsbereichszugriffssteuerungen; Tool-/Session-Isolierung angepasst. - Agenten-Schleife: Kompaktion, Bereinigung, Streaming und Fehlerbehandlung härter. - Provider: Zuverlässigkeit, Threading, Reaktionen, Medien und Wiederholungen bei Telegram/WhatsApp/Discord/Slack verbessert. - Kontroll-UI: Logs-Tab, Streaming-Stabilität, Fokusmodus und Rendering großer Ausgaben behoben. - CLI/Gateway/Doctor: Daemon/Logs/Status, Auth-Migration und Diagnosen deutlich erweitert. ### Breaking - **SICHERHEIT (so schnell wie möglich aktualisieren):** eingehende DMs sind jetzt standardmäßig auf Telegram/WhatsApp/Signal/iMessage/Discord/Slack eingeschränkt. - Zuvor konnte dein Bot offen für jedermann sein (insbesondere auffindbare Telegram-Bots), wenn du keine Allowlist konfiguriert hattest. - Neue Standardsetzung: DM-Paarung (`dmPolicy="pairing"` / `discord.dm.policy="pairing"` / `slack.dm.policy="pairing"`). - Um das alte Verhalten „offen für jedermann“ beizubehalten: setze `dmPolicy="open"` und füge `"*"` in das relevante `allowFrom` ein (Discord/Slack: `discord.dm.allowFrom` / `slack.dm.allowFrom`). - Genehmige Anfragen über `clawdbot pairing list --provider <provider>` + `clawdbot pairing approve --provider <provider> <code>` (Telegram unterstützt auch `clawdbot telegram pairing ...`). - Sandbox: Standard `agent.sandbox.scope` auf `"agent"` (ein Container/Arbeitsbereich pro Agent). Verwende `"session"` für Isolation pro Sitzung; `"shared"` deaktiviert die länderübergreifende Isolation. - Zeitstempel in Agentenumschlägen jetzt UTC (kompakt `YYYY-MM-DDTHH:mmZ`); entfernt `messages.timestampPrefix`. Füge `agent.userTimezone` hinzu, um dem Modell die lokale Zeit des Nutzers mitzuteilen (nur System-Prompt). - Modellerconfig-Schemaänderungen (Auth-Profile + Modelllisten); Doctor migriert automatisch und Gateway schreibt Legacy-Konfigurationen beim Start neu. - Befehle: Alle Slash-Befehle auf autorisierte Sender beschränken; füge `/compact` hinzu, um Sitzungs-Kontext manuell zu komprimieren. - Gruppen: `whatsapp.groups`, `telegram.groups` und `imessage.groups` wirken jetzt als Allowlisten, wenn gesetzt. Füge `"*"` hinzu, um das Ganze offen zu halten. - Auto-Reply: Entfernt `autoReply` aus Discord/Slack/Telegram Kanal-Konfigurationen; stattdessen `requireMention` verwenden (Telegram-Themen unterstützen jetzt Overrides für `requireMention`). - CLI: entferne `update`, `gateway-daemon`, `gateway {install|uninstall|start|stop|restart|daemon status|wake|send|agent}`, und `telegram` Befehle; verschiebe `login/logout` zu `providers login/logout` (Top-Level-Aliasse versteckt); nutze `daemon` zur Service-Steuerung, `send`/`agent`/`wake` für RPC, und `nodes canvas` für Canvas-Operationen. ### Fixes - **CLI/Gateway/Doctor:** Daemon-Laufzeit-Auswahl + verbesserte Logs/Status/Gesundheit/Fehler; Auth/Passwort-Behandlung für lokale CLI; reichhaltigere Close-/Timeout-Details; automatische Migration alter Configs/Sitzungszustände; Integritätsprüfungen + Reparaturhinweise; `--yes`/`--non-interactive`; `--deep` Gateway-Scans; bessere Neustart-/Service-Hinweise. - **Agenten-Schleife + Kompaktion:** Feineinstellungen bei Kompaktion/Bereinigen, Overflow-Handling, sicherer Bootstrap-Kontext, pro-Anbieter-Threading/Bestätigungen; Opt-in Tool-Ergebnis-Verkleinerung + kompakte Nachverfolgung. - **Sandbox + Tools:** Overrides pro Agenten-Sandbox, WorkspaceAccess-Zugriffe, Sitzungs-Tool-Sichtbarkeit, Tool-Policy-Overrides, Prozess-Isolation, Vereinheitlichung von Tool-Schema/Timeout/Reaktion. - **Provideren (Telegram/WhatsApp/Discord/Slack/Signal/iMessage):** Retry/Backoff, Threading, Reaktionen, Media-Gruppen/Anhänge, Erwähnungs-Gate, Tippverhalten, Stabilität von Fehlern/Logs; Langpolling + Foren-Themen-Isolation für Telegram. - **Gateway/CLI UX:** `clawdbot logs`, Cron-Listen-Farben/Aliase, Docs-Suche, Flows zum Listen/Hinzufügen/Löschen von Agenten, Status-Nutzungs-Schnappschüsse, Laufzeit-/Auth-Quellanzeige, Vereinheitlichung von `/status`/Befehls-Authentifizierung. - **Control UI/Web:** Logs-Tab, Fokus-Modus-Polish, Konfigurations-Formular-Robustheit, Streaming-Stabilität, Output-Limits von Tools, fensterbasierter Chat-Verlauf, und Neuanmeldungs-/Passwort-URL-Auth. - **macOS/Android/TUI/Build:** macOS-Gateway-Races, QR-Bundling, JSON5-Konfigurationssicherheit, Voice Wake-Härtung; Android EXIF-Drehung + APK-Namensgebung/-Versionierung; TUI-Tastenbehandlung; Fehler bei Tooling/Bundle. - **Packaging/Kompatibilität:** npm Dist-Ordner-Abdeckung, Node 25 qrcode-terminal-Importfehlerbehebungen, Bun/Playwright/WebSocket-Patches, und Docker Bun-Installationen. - **Docs:** neue FAQ/ClawdHub/Config-Beispiele/Showcase-Einträge und klärende Auth-, Sandbox- und systemd-Dokumentation. ### Maintenance - Skill-Erweiterungen (Himalaya E-Mail, CodexBar, 1Password). - Abhängigkeits-Refreshes (pi-* Stack, Slack SDK, discord-api-types, file-type, zod, Biome, Vite). - Refactors: Zentralisierte Gruppen-Allowlist/Mention-Policy; Lint-/Import-Aufräumarbeiten; Wechsel von tsx → bun für die TS-Ausführung. ## 2026.1.5

CursorvCLI

Neue CLI-Funktionen und verbesserte CLI-Leistung

8. Januar 2026

Großes CLI-Update mit Modellauswahl, Regelnverwaltung, MCP-Server-Steuerung und Leistungsverbesserungen.

Factory AI (Droid)vv0.45.0

Detaillierte Ansicht und Bugfixes

8. Januar 2026

Ctrl+O bietet eine detaillierte Ansicht der vollständigen akkumulierten Befehlsausgabe mit verschiedenen UI-Fixes.

Claude Codev2.1.1

v2.1.1

7. Januar 2026

Claude Codev2.0.76

v2.0.76

7. Januar 2026

## What's changed - Fixed issue with macOS code-sign warning when using Claude in Chrome integration

OpenCodev1.1.6

v1.1.6

7. Januar 2026

## Core - Encode non-ASCII directory paths in HTTP headers (@jwaterwater) - Add truncation for all tools - Make General SubAgent visible in agent list (@dbpolito) - Improve handling of unserializable errors - Added ability to hide subagents from primary agent's system prompt (@Sewer56) - Add --use-system-ca flag in execArgv to support system certificate authorities - Ensure commands listing subagent work ## TUI - Add theme functionality - Reload theme when configuration changes - Clear palette cache - Add agent color to inline task tool ASCII character and highlight agent name (@spoons-and-mirrors) - Escape backticks when passing commands to PowerShell (@NSPC911) - Restore attach session lookup behavior in TUI (@rmk40) ## Desktop - Fix pasting images when input is empty in desktop app (@dbpolito) - Bundle media framework in desktop application (@Brendonovich) - Fixed macOS application entitlements configuration (@Brendonovich) - Add audio input and camera permissions to entitlements (@saeedvaziry) - Bundle media framework in AppImage for desktop application (@Brendonovich) - Use Show instead of Suspense in desktop application (@Brendonovich) - Add UI package to desktop application (@Brendonovich) - Show window immediately during desktop startup (@usvimal) - Rework storage approach in app - Fix session navigation with keyboard shortcuts - Permission auto-accept should hide when not needed - Improve terminal text selection visibility with better color contrast **Thank you to 11 community contributors:** - @rmk40: - fix(tui): restore attach session lookup behavior (#7150) - @NSPC911: - fix: escape backticks when passing to powershell (#7157) - @Sewer56: - Added: Ability to hide subagents from primary agents system prompt. (#4773) - @Brendonovich: - use custom tauri-cli on linux - desktop: use regular tauri cli again - desktop: use correct tauri-cli branch - try using forked tauri-cli again - add ui package to desktop - desktop: use Show instead of Suspense - desktop: appimage bundleMediaFramework - fix entitlements - desktop: go back to regular tauri cli - desktop: bundleMediaFramework - @spoons-and-mirrors: - add agent color to inline task tool ascii char and highlight agent name (#7142) - @usvimal: - perf: show window immediately during desktop startup (#6734) - @dbpolito: - Make General SubAgent not hidden (#6752) - Desktop: Fix Paste image on empty input (#7130) - @saeedvaziry: - Add audio input and camera permissions to entitlements (#7117) - @shuv1337: - feat: add Night Owl theme (desktop) (#7151) - @jwaterwater: - fix: encode non-ASCII directory paths in HTTP headers (#7145) - @athal7: - fix(ui): prevent iOS Safari auto-zoom on input focus (#7214)

Gemini CLIv0.23.0

Release v0.23.0

7. Januar 2026

## Was sich geändert hat * Code-Assist-Service-Metriken. von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15024 * chore/release: Version auf 0.21.0-nightly.20251216.bb0c0d8ee erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15121 * Docs von @Roaimkhan in https://github.com/google-gemini/gemini-cli/pull/15103 * Offizielles ACP SDK verwenden und HTTP/SSE-basierte MCP-Server unterstützen von @SteffenDE in https://github.com/google-gemini/gemini-cli/pull/13856 * Vordergrund für Themes außer Shades of Purple und Holiday entfernen von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/14606 * chore: Repo-spezifische Tipps entfernen von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15164 * chore: Benutzersuche in der Fußzeile im Debug-Modus entfernen von @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/15169 * Unnötiges Awaiting ablehnen. von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15172 * Einen Padding im Einstellungsdialog hinzufügen, um Flimmern zu vermeiden. von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15173 * feat(core): remote Agent-Infrastruktur einführen und lokalen Executor umbenennen von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15110 * feat(cli): Befehl `/auth logout` hinzufügen, um Anmeldeinformationen und Auth-Zustand zu löschen von @CN-Scars in https://github.com/google-gemini/gemini-cli/pull/13383 * (fix) Automatischer pr-Labeller von @DaanVersavel in https://github.com/google-gemini/gemini-cli/pull/14885 * feat: Gemini 3 Flash im Gemini CLI starten ⚡️⚡️⚡️ von @scidomino in https://github.com/google-gemini/gemini-cli/pull/15196 * Refactor: migrate console.error in ripGrep.ts zu debugLogger von @Adib234 in https://github.com/google-gemini/gemini-cli/pull/15201 * chore: a2a-js auf 0.3.7 aktualisieren von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15197 * chore(core): redundanten isModelAvailabilityServiceEnabled-Toggle entfernen und toten Code bereinigen von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15207 * feat(core): Späte Auflösung von `GenerateContentConfig`s und Mutationsreduzierung. von @joshualitt in https://github.com/google-gemini/gemini-cli/pull/14920 * Respect previewFeatures-Wert aus dem Remote-Flag, falls undefiniert von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15214 * feat(ui): Windows-Zwischenablage-Bildunterstützung und Alt+V-Einfüge-Umgehung von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15218 * chore(core): Legacy-Fallback-Flags entfernen und Schleifen-Erkennung migrieren von @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15213 * fix(ui): Verhindern, dass voreilige Slash-Befehls-Vervollständigung Geschwisterbefehle versteckt von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15224 * Docs: Changelog für 17. Dez 2025 aktualisieren von @jkcinouye in https://github.com/google-gemini/gemini-cli/pull/15204 * Code-Assist-Backend-Telemetrie für Akzeptieren/Ablehnen von Vorschlägen von @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15206 * fix(cli): Anfangs-Historienlänge bei Chat-Befehlen korrekt behandeln von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15223 * chore/release: Version auf 0.21.0-nightly.20251218.739c02bd6 erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15231 * Änderung detaillierter Modellstatistiken, um eine neue geteilte Table-Klasse zu verwenden, um Robustheit zu lösen. von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15208 * feat: Agent-TOML-Parser hinzufügen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15112 * Hinzufügen eines Core-Tools, das allen Kontext aus dem Core-Paket hinzufügt. von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15238 * (docs): Referenzabschnitt zu Hooks-Dokumentation hinzufügen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15159 * feat(hooks): Unterstützung für freundliche Namen und Beschreibungen hinzufügen von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15174 * feat: Hintergrundfarbe erkennen von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15132 * 3.0 zu zulässigen sensiblen Schlüsselwörtern hinzufügen von @scidomino in https://github.com/google-gemini/gemini-cli/pull/15276 * feat: Weitere Umgebungsvariablen an Shell-Ausführung übergeben von @galz10 in https://github.com/google-gemini/gemini-cli/pull/15160 * Entfernen ungenutzten Codes von @scidomino in https://github.com/google-gemini/gemini-cli/pull/15290 * Alle 429 als retryableQuotaError behandeln von @sehoon38 in https://github.com/google-gemini/gemini-cli/pull/15288 * Unnötige Abhängigkeiten entfernen von @scidomino in https://github.com/google-gemini/gemini-cli/pull/15291 * fix: Endlosschleife bei Fehlern in Prompt-Vervollständigung verhindern von @galz10 in https://github.com/google-gemini/gemini-cli/pull/14548 * fix(ui): Befehlsvorschläge auch bei perfekter Übereinstimmung anzeigen und sortieren von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15287 * feat(hooks): Protokollausgabe reduzieren und Fehlerbericht in der UI verbessern von @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15297 * feat: Bestätigungsbeschriftungen der Werkzeuge für bessere UX vereinfachen von @NTaylorMullen in https://github.com/google-gemini/gemini-cli/pull/15296 * chore/release: Version auf 0.21.0-nightly.20251219.70696e364 erhöhen von @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15301 * feat(core): JIT-Kontext-Speicherlade- und UI-Synchronisation implementieren von @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/14469 * feat(ui): "Allow for all future sessions" hinter eine Einstellung setzen (standardmäßig deaktiviert) von @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15322 * fix(cli): Platzhalter der Eingabe während des Shell-Modus ändern von @JayadityaGit in https://github.com/google-gemini/gemini-cli/pull/15135 * OAuth-Ressourcenparameter mit MCP-Server-URL validieren von @galz10 in https://github.com/google-gemini/gemini-cli/pull/15289 * docs(cli): System Prompt Override (GEMINI_SYSTEM_MD) hinzufügen von @ashmod in https://github.com/google-gemini/gemini-cli/pull/9515 * robustere Logs zur Befehlsparsing von Tools von @scidomino in https://github.com/google-gemini/gemini-cli/pull/15339 * Demoversion des Introspektions-Agenten von @scidomino in

Codex CLIvrust-v0.79.0

0.79.0

7. Januar 2026

## New Features - Add multi-conversation “agent control” so a session can spawn or message other conversations programmatically (#8783, #8788) - Add app-server `thread/rollback` so IDE clients can undo the last N turns without rewriting history (#8454) - Add `web_search_cached` to enable cached-only Web Search results as a safer alternative to live requests (#8795) - Allow global exec flags (model/json/sandbox toggles, etc.) to be passed after `codex exec resume` (#8440) - Show time/version-targeted announcement tips in the TUI, driven by a TOML file in the repo (#8752) - Add an `[analytics] enabled=...` config section to control analytics behavior (#8350) ## Bug Fixes - Fix TUI2 transcripts so streamed markdown reflows on resize and copy/paste preserves soft wraps (#8761) - Make `apply_patch` parsing tolerant of whitespace-padded `*** Begin Patch` / `*** End Patch` markers (#8746) - Render paths relative to the current working directory before checking git roots, improving output in non-git workspaces (#8771) - Prevent `CODEX_MANAGED_CONFIG_PATH` from overriding managed config in production, closing a policy bypass (#8762) - Ensure app-server conversations respect the config passed in by the client (#8765) - Reduce TUI UX glitches: suppress popups when browsing input history, fix copy “pill” rendering, and clear background terminals on interrupt (#8772, #8777, #8786) ## Documentation - Clarify onboarding/login guidance for headless/remote setups by steering users to `codex login --device-auth` (#8753) - Update model examples used in tooling/config to current `gpt-5.2` naming (#8566) ## Chores - Refactor skills discovery to use the config layer stack so all configured skill folders are considered (#8497) - Auto-populate GitHub release notes at release creation time (#8799) - Add more `apply_patch` test scenarios to harden coverage (#8230) - Tweak unified exec event emission to only send “begin” when a PTY is present (#8780) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.78.0...rust-v0.79.0 - #8753 [device-auth] Update login instruction for headless environments. @mzeng-openai - #8497 Use ConfigLayerStack for skills discovery. @xl-openai - #8440 Allow global exec flags after `resume` and fix CI codex build/timeout @fjord-oai - #8230 chore(apply-patch) additional scenarios @dylan-hurd-oai - #8746 fix: accept whitespace-padded patch markers @tibo-openai - #8761 tui2: stop baking streaming wraps; reflow agent markdown @joshka-oai - #8771 fix: render cwd-relative paths in tui @joshka-oai - #8770 fix: fix readiness subscribe token wrap-around @tibo-openai - #8780 chore: emit unified exec begin only when PTY exist @jif-oai - #8762 fix: stop honoring CODEX_MANAGED_CONFIG_PATH environment variable in production @bolinfest - #8566 fix: update model examples to gpt-5.2 @seuros - #8735 feat: add head-tail buffer for `unified_exec` @jif-oai - #8331 chore: add model/list call to app-server-test-client @owenlin0 - #8752 feat: forced tool tips @jif-oai - #8786 chore: clear background terminals on interrupt @jif-oai - #8350 feat: add analytics config setting @JaviSoto - #8783 feat: agent controller @jif-oai - #8772 suppress popups while browsing input history @xl-openai - #8777 Clear copy pill background and add snapshot test @joshka-oai - #8788 feat: drop agent bus and store the agent status in codex directly @jif-oai - #8454 feat(app-server): thread/rollback API @owenlin0 - #8765 [app-server] fix config loading for conversations @celia-oai - #8795 add web_search_cached flag @sayan-oai - #8799 fix: populate the release notes when the release is created @bolinfest

Factory AI (Droid)vv0.44.0

Statuszeile und MCP-Verwaltung

7. Januar 2026

Neuer Statusleisten-Befehl für benutzerdefinierte Konfiguration und Bulk-MCP-Toolverwaltung.

OpenCodev1.1.4

v1.1.4

6. Januar 2026

## Core - Add timeout to prevent hanging operations (@spoons-and-mirrors) - Add support for URL-based instructions (@spoons-and-mirrors) - Load OPENCODE_CONFIG_DIR AGENTS.md into the system prompt (@ThomasGormley) - Fix title generation for subtask-only messages to extract actual user prompts instead of generic tool execution descriptions - Ensure 'name' isn't being sent in request body for custom agent - Restore skill filtering by agent permissions (@malhashemi) - Add missing await for available skills in skill tool (@mwolting) - Prevent Java Language Server path checks from throwing errors (@T1mn) - Add automatic jp. prefix assignment for Tokyo region (ap-northeast-1) (@ikeda-tomoya-swx) - Polish documentation markdown wording and capitalization (@T1mn) - Filter dead worktrees in core functionality - Handle cases where git is not available or initialized - Upgrade OpenTUI to v0.1.69 with text rendering performance improvements ## TUI - Add frecency-based file autocomplete to CLI (@galkatz373) - Add keyboard shortcut 'c' to copy device code in OAuth flow (@akinfelami) - Fix sidebar title padding to prevent scrollbar edge case (@ryanwyler) ## Desktop - Copy and paste in terminal was broken - Fix desktop documentation - Increase contrast in terminal text for better readability - Add view button to open files from review sidebar (@justrau) - Open review sidebar when selecting file from picker (@justrau) - Add middle-click to close tabs in review sidebar (@justrau) - Improved web load performance through code splitting - Improve scroll store performance in app - Update Cargo.lock (@Brendonovich) - Add single-instance plugin to prevent multiple desktop windows (@Junseo5) - Make subtasks clickable in the task list (@OpeOginni) **Thank you to 21 community contributors:** - @OpeOginni: - Feat/clickable subtask (#6846) - @T1mn: - docs: polish markdown wording and capitalization (#7019) - docs: add Simplified Chinese README (#7055) - fix: prevent jdtls path checks from throwing (#7052) - @shanebishop1: - docs: fix order of permissions in agents docs (permissions subsection) (#7041) - @xiliumz: - fix: use actual version in install script (#7044) - @Junseo5: - fix(desktop): add single-instance plugin to prevent multiple windows (#6966) - @Brendonovich: - update Cargo.lock - @Eric-Guo: - Fix(app): @pierre/diffs will crash when a diff has undefined text (#7059) - @justrau: - feat(app): add middle-click to close tabs in review sidebar (#7094) - fix(app): open review sidebar when selecting file from picker (#7096) - feat(app): add view button to open files from review sidebar (#7095) - fix(app): open links in new tab or browser (#7127) - @athal7: - docs(ecosystem): add opencode-devcontainers plugin (#7100) - @ryanwyler: - fix: sidebar title padding to prevent scrollbar edge case (#7089) - @ikeda-tomoya-swx: - fix(provider): add jp. prefix auto-assignment for Tokyo region (ap-northeast-1) (#7053) - @akinfelami: - feat(tui): add 'c' shortcut to copy device code in OAuth flow (#7020) - @mwolting: - fix: add missing await for available skills in `skill` tool (#7072) - @malhashemi: - fix: restore skill filtering by agent permissions (#7042) - @jerome-benoit: - fix(desktop): use current_binary() to support symlinked executables (#7102) - @DamianB-BitFlipper: - docs: Expand keybinds documentation (#7108) - @ThomasGormley: - load `OPENCODE_CONFIG_DIR` AGENTS.md into the system prompt (#7115) - @galkatz373: - feat(cli): frecency file autocomplete (#6603) - @spoons-and-mirrors: - feat: url based instructions (#7125) - adding timeout (#7128) - @mtymek: - docs: add OpenCode-Obsidian plugin (#7129) - @dbpolito: - Desktop: Fix Big Messages (#7133)

Codex CLIvrust-v0.78.0

0.78.0

6. Januar 2026

## New Features - Add `Ctrl+G` to open the current prompt in your configured external editor (`$VISUAL`/`$EDITOR`) and sync edits back into the TUI. (#7606) - Support project-aware config layering: load repo-local `.codex/config.toml`, honor configurable `project_root_markers`, and merge that with system config like `/etc/codex/config.toml`. (#8354, #8359, #8461) - Support enterprise-managed configuration requirements on macOS via an MDM-provided TOML payload. (#8743) - Improve `tui2` transcript navigation with multi-click selection, a copy shortcut/affordance, and a draggable auto-hiding scrollbar. (#8462, #8471, #8728) - Start Windows PowerShell sessions in UTF-8 mode to reduce encoding-related prompt/output issues. (#7902) - Exec policy rules can now include human-readable justifications, and policy loading follows the unified config-layer stack. (#8349, #8453, #8751) ## Bug Fixes - Fix failures when the model returns multiple tool calls in a single turn by emitting tool calls in the format the Chat Completions API expects. (#8556) - Make `/review` compute diffs from the session’s working directory so base-branch detection works correctly with runtime cwd overrides. (#8738) - Handle the legacy Chat Completions streaming terminator cleanly to avoid spurious SSE parse errors. (#8708) - Fix a set of `tui2` rendering/input edge cases (screen corruption, scroll stickiness, and selection/copy correctness). (#8463, #8695, #8449) - Improve diagnostics when `ripgrep` download fails during CLI packaging so failures are easier to debug. (#8486) - Avoid a panic when parsing alpha/stable version strings. (#8406) ## Documentation - Replace and de-duplicate user docs with links to the developer documentation site (including skills docs). (#8662, #8407) - Clarify configuration documentation, including the `developer_instructions` option. (#8376, #8701) - Fix broken README links. (#8682) ## Chores - Performance tuning for TUIs: cap redraw scheduling and reduce unnecessary rerenders during streaming to lower CPU usage. (#8499, #8681, #8693) - Update Rust dependencies across `codex-rs`. (#8414, #8596, #8597) ## Changelog Full Changelog: https://github.com/openai/codex/compare/rust-v0.77.0...rust-v0.78.0 - #8407 Point skills docs to developer documentation site @etraut-openai - #8406 fix: do not panic on alphas @aibrahim-oai - #8091 Update ghost_commit flag reference to undo @charlie-openai - #8423 test(tui2): re-enable ANSI for VT100 tests @joshka-oai - #8419 fix(tui2): constrain transcript mouse selection bounds @joshka-oai - #8424 chore: un-ship undo @jif-oai - #8376 docs: add `developer_instructions` config option and update descriptions @448523760 - #8431 chore: drop undo from the docs @jif-oai - #8410 chore(deps): bump peter-evans/create-pull-request from 7 to 8 @dependabot - #8412 chore(deps): bump test-log from 0.2.18 to 0.2.19 in /codex-rs @dependabot - #8413 chore(deps): bump landlock from 0.4.2 to 0.4.4 in /codex-rs @dependabot - #8414 chore(deps): bump clap from 4.5.47 to 4.5.53 in /codex-rs @dependabot - #8416 chore(deps): bump openssl-sys from 0.9.109 to 0.9.111 in /codex-rs @dependabot - #7902 feat(windows) start powershell in utf-8 mode @dylan-hurd-oai - #8349 Add ExecPolicyManager @pakrym-oai - #8354 feat: support in-repo .codex/config.toml entries as sources of config info @bolinfest - #8359 feat: add support for project_root_markers in config.toml @bolinfest - #8395 Handle concatenation nodes in bash command parser for exec policy @ivanmurashko - #8446 Update models.json @github-actions - #8444 use a junction for the cwd while read ACLs are being applied @iceweasel-oai - #8447 remove minimal client version @aibrahim-oai - #8448 chore: add ConfigLayerStack as a field of Config @bolinfest - #7606 feat: open prompt in configured external editor @sayan-oai - #8449 fix(tui2): copy transcript selection outside viewport @joshka-oai - #8418 chore(tui): include tracing targets in file logs @joshka-oai - #8456 chore: include User layer in ConfigLayerStack even if config.toml is empty @bolinfest - #8453 feat: load ExecPolicyManager from ConfigLayerStack @bolinfest - #8462 feat(tui2): add copy selection shortcut + UI affordance @joshka-oai - #8461 feat: honor /etc/codex/config.toml @bolinfest - #8463 fix(tui2): fix screen corruption @joshka-oai - #8458 chore: save more about turn context in rollout log file @apanasenko-oai - #8466 fix(tui2): start transcript selection on drag @joshka-oai - #8293 [tui] add optional details to TUI status header @sayan-oai - #8471 feat(tui2): add multi-click transcript selection @joshka-oai - #8486 fix(codex-cli): improve ripgrep download diagnostics @joshka-oai - #8460 fix: limit output size for exec command in unified exec @shijie-oai - #8493 fix: fix test that was writing temp file to cwd instead of TMPDIR @bolinfest - #8484 Remove reasoning format @aibrahim-oai - #8499 perf(tui): cap redraw scheduling to 60fps @joshka-oai - #8496 feat: introduce codex-utils-cargo-bin as an alternative to assert_cmd::Command @bolinfest - #8498 fix: declare test path relative to `$CARGO_MANIFEST_DIR` @bolinfest - #8307 [chore] add additional_details to StreamErrorEvent + wire through @sayan-oai - #8522 fix: restrict windows-sys to Windows target @RunjiLiu - #8595 chore(deps): bump toml_edit from 0.23.7 to 0.24.0+spec-1.1.0 in /codex-rs @dependabot - #8596 chore(deps): bump tracing-subscriber from 0.3.20 to 0.3.22 in /codex-rs @dependabot - #8597 chore(deps): bump tokio from 1.47.1 to 1.48.0 in /codex-rs @dependabot - #8598 chore(deps): bump regex-lite from 0.1.7 to 0.1.8 in /codex-rs @dependabot - #8491 Refresh on models etag mismatch @aibrahim-oai - #8483 use a SandboxUsers group for ACLs instead of granting to each sandbox user separately @iceweasel-oai - #8676 Log compaction request bodies @pakrym-oai - #8488 Remove model family from tui @aibrahim-oai - #8662 Replaced user documentation with links to developers docs site @etraut-openai - #8681 perf(tui2): reduce unnecessary redraws @joshka-oai - #8682 Fixed broken link in README @etraut-openai - #8677 Account for last token count on resume @aibrahim-oai - #8687 hard disable remote models refresh when feature is disabled @aibrahim-oai - #8415 chore(deps): bump tracing-opentelemetry from 0.31.0 to 0.32.0 in /codex-rs @dependabot - #8688 Attach more tags to feedback submissions @pakrym-oai - #8701 chore: update outdated docs @tibo-openai - #8693 perf(tui2): cache transcript view rendering @joshka-oai - #8697 fix: brighten transcript copy affordance @joshka-oai - #8629 ci: prevent workflows from running on forks @leezenn - #8695 fix(tui2): avoid scroll stickiness at cell boundaries @joshka-oai - #8716 fix(tui2): render copy pill at viewport bottom @joshka-oai - #8718 tui2: copy selection dismisses highlight @joshka-oai - #8721 chore(deps): bump clap_complete from 4.5.57 to 4.5.64 in /codex-rs @dependabot - #8723 chore(deps): bump tokio-stream from 0.1.17 to 0.1.18 in /codex-rs @dependabot - #8724 chore(deps): bump derive_more from 2.0.1 to 2.1.1 in /codex-rs @dependabot - #8725 chore(deps): bump insta from 1.44.3 to 1.46.0 in /codex-rs @dependabot - #8556 fix: chat multiple tool calls @jif-oai - #8738 fix: /review to respect session cwd @tibo-openai - #8694 [MCP] Sanitize MCP tool names to ensure they are compatible with the Responses APO @gpeal - #8728 feat(tui2): transcript scrollbar (auto-hide + drag) @joshka-oai - #8708 fix(codex-api): handle Chat Completions DONE sentinel @joshka-oai - #8743 (MacOS) Load config requirements from MDM @gt-oai - #8377 feat: expose outputSchema to user_turn/turn_start app_server API @apanasenko-oai - #8747 chore: GH pager @jif-oai - #8686 better idempotency for creating/updating firewall rules during setup. @iceweasel-oai - #8683 never let sandbox write to .codex/ or .codex/.sandbox/ @iceweasel-oai - #8459 chore: improve skills render section @gverma-openai - #7858 Use issuer URL in device auth prompt link @abrar71 - #8492 best effort to "hide" Sandbox users @iceweasel-oai - #8751 feat: add justification arg to prefix_rule() in *.rules @bolinfest

Factory AI (Droid)vv0.43.0

Terminal und UI Verbesserungen

6. Januar 2026

Terminalpanel-Anzeigefehler behoben und Chat-Eingabe verbessert.

OpenCodev1.1.3

v1.1.3

5. Januar 2026

## Core - Fixed terminal long-running processes in app - Preserve permission config key order to maintain user-defined permission precedence - Pass image parts to custom commands (@melihmucuk) - Add support for jp. prefix in Amazon Bedrock cross-region inference profiles (@ikeda-tomoya-swx) - Add configuration options and authentication precedence for Bedrock (@wnkz) - Transform agent improvements and optimizations - Filter out empty messages and reasoning for Anthropic API compatibility ## TUI - Expand directory on Tab, select on Enter in TUI autocomplete (@yuguorui) - Update GitHub Actions examples to use actions/checkout@v6 (@joshuadavidthomas) - Fix Lucent Orange background transparency for slash commands (@roerohan) - Add description for MCP command in documentation (@cravenceiling) - Add system theme resolution and event handling to TUI - Improve experience when operations are interrupted ## Desktop - Store image attachments in the app - Fix bottom padding in app interface - Fixed terminal flakiness issues in the app - Performance improvements through event batching in app - Fix prompt input covering content in app - Worktree selection now correctly navigates to the selected worktree - Avoid unintended submits during IME composition (@MogamiTsuchikawa) - Highlight collapsed active project in sidebar (@albingroen) - Fix flicker and navigation issues when collapsing or expanding projects in the app (@albingroen) - Add CLI installation to desktop application (@dbpolito) - Fix image dragging in project edit dialog (@albingroen) **Thank you to 10 community contributors:** - @albingroen: - fix(app): fix image dragging in project edit dialog (#6700) - fix(app): fix flicker and navigation when collapsing/expanding projects (#6658) - feat(app): highlight collapsed active project in sidebar (#6958) - @dbpolito: - Desktop: Install CLI (#6526) - @MogamiTsuchikawa: - fix(app): avoid unintended submits during IME composition (#6952) - @wnkz: - feat(bedrock): config options and authentication precedence (#6377) - @cravenceiling: - docs: add description for MCP command (#6944) - @roerohan: - fix: lucent-orng bg transparency for slash commands (#6938) - @joshuadavidthomas: - docs: update GHA examples to use `actions/checkout@v6` (#6969) - @ikeda-tomoya-swx: - fix(provider): add support for jp. prefix in Amazon Bedrock cross-region inference profiles (#6916) - @melihmucuk: - fix: pass image parts to custom commands (#6525) - @yuguorui: - tui: autocomplete: expand directory on Tab, select on Enter (#6975)

OpenCodev1.1.2

v1.1.2

5. Januar 2026

## Core - Fix permission rule matching to use permission field instead of pattern field - Fix permission evaluation to use rule-based matching instead of wildcard patterns - Handle skill scan failures for .claude files gracefully - Handle case where big-pickle is unavailable as a fallback - Prevent main model thinking variant from applying to small model (@melihmucuk) - Increase default timeout to 30 seconds based on user feedback - Add session search functionality with debounced input and server-side filtering - Improve Rust formatter detection and add cargo fmt support ## TUI - Allow exit when viewing child session in TUI (@rmk40) - Fix symmetric padding in TUI input field - Prevent session list rows from wrapping to two lines (@rgodha24) - Fix optional session access to prevent runtime errors in TUI ## Desktop - Implement auto-scroll for active command in slash popover (@OpeOginni) - Fix responsive menu in desktop application (@dbpolito) - Optimize model dialog visibility lookups (@shkumbinhasani) - Add home icon to responsive menu in desktop app (@dbpolito) - Add SVG preview support in session viewer (@shuv1337) - Add AppStream MetaInfo file for desktop application (@NN708) - Fix mobile screen size issues (@carterworks) - Fix custom slash commands not showing on initial / (@Raviguntakala) - Improved prompt input handling in the application **Thank you to 11 community contributors:** - @melihmucuk: - fix: prevent main model thinking variant from applying to small model (#6839) - @rgodha24: - fix: prevent session list rows from wrapping to 2 lines (#6812) - @albingroen: - fix(ui): fix select chevron alignment (#6690) - @Raviguntakala: - fix(app): fix custom slash commands not showing on initial / (#6829) - @carterworks: - Fix a few mobile screen size issues (#6808) - @NN708: - feat(desktop): add AppStream MetaInfo file (#6030) - @shuv1337: - feat(app): add SVG preview support in session viewer (#6868) - @dbpolito: - Desktop: Improve Big Session Navigation - Scrollable (#6837) - Desktop: Adding Home Icon on Responsive Menu (#6794) - Desktop: Fix Responsive Menu (#6789) - @shkumbinhasani: - perf: optimize model dialog visibility lookups (#6791) - @OpeOginni: - feat(desktop): implement auto-scroll for active command in slash popover (#6797) - @rmk40: - fix(tui): allow exit when viewing child session (#6898)

OpenClawv2026.1.5-3

OpenClaw 2026.1.5-3

5. Januar 2026

## Fixes - NPM package: fehlende Runtime-dist-Ordner einschließen (slack/signal/imessage/tui/wizard/control-ui/daemon), um ERR_MODULE_NOT_FOUND bei Node 25 npx-Installationen zu vermeiden.

OpenClawv2026.1.5-1

clawdbot 2026.1.5-1

5. Januar 2026

### Fixes - NPM package: include `dist/sessions` so `clawdbot agent` resolves session helpers in npx installs. - Node 25: avoid unsupported directory import by targeting `qrcode-terminal/vendor/QRCode/index.js`.

OpenClawv2026.1.5

clawdbot 2026.1.5

5. Januar 2026

### Highlights - Models: added image-spezifische Modellkonfiguration (`agent.imageModel` + Fallbacks) und Scan-Unterstützung. - Agent Tools: neues `image` Tool, das zum image Modell geroutet wird (falls konfiguriert). - Config: standardisierte Modell-Abkürzungen (`opus`, `sonnet`, `gpt`, `gpt-mini`, `gemini`, `gemini-flash`). - Docs: dokumentieren integrierte Modell-Abkürzungen + Priorität (Benutzer-Konfiguration gewinnt). ### Fixes - Control UI: Markdown in Tool-Ergebnis-Karten rendern. - Control UI: Verhindern von überlappenden Aktions-Buttons in Discord-Serverregeln auf schmalen Layouts. - Android: Antippen der Foreground-Service-Benachrichtigung bringt die App nach vorn. (#179) — thanks @Syhids - Cron-Tool verwendet `id` für Update/Remove/Run/Runs (angleicht sich an Gateway-Parametern). (#180) — thanks @adamgall - Control UI: Chat-Ansicht verwendet Seiten-Scrollen mit sticky Header/Sidebar und festem Composer (kein inneres Scrollrahmen). - macOS: Behandle Standortberechtigung als immer-nur, um iOS-exklusive Enums zu vermeiden. (#165) — thanks @Nachx639 - macOS: Generierte Gateway-Protokoll-Modelle `Sendable` für Swift 6 strikte Concurrency. (#195) — thanks @andranik-sahakyan - macOS: bundle QR-Code-Renderer-Module, damit der DMG-Gateway-Boot bei fehlenden qrcode-terminal Vendor-Dateien nicht abstürzt. - WhatsApp: Tippindikator während Hintergrundaufgaben der Heartbeat-Sitzung unterdrücken. (#190) — thanks @mcinteerj - WhatsApp: Offline-Verlauf-Synchronisationsnachrichten als gelesen markieren ohne Auto-Antwort. (#193) — thanks @mcinteerj - Discord: Duplizierte Antworten vermeiden, wenn ein Provider späte Streaming-Events `text_end` emittiert (OpenAI/GPT). - CLI: Tailnet-IP für lokale Gateway-Anrufe verwenden, wenn Bind tailnet/auto (Behebt #176). - Env: globale `$CLAWDBOT_STATE_DIR/.env` laden (`~/.clawdbot/.env`) als Fallback nach CWD `.env`. - Env: optionale Login-Shell env-Fallback (Opt-in; importiert erwartete Keys ohne vorhandene Env zu überschreiben). - Agent Tools: OpenAI-kompatible Tool-JSON-Schemas (Fehler bei `browser` beheben, Unions-Schemas normalisieren). - Onboarding: beim Ausführen aus dem Quellcode automatisch fehlende Control-UI-Assets bauen (`pnpm ui:build`). - Discord/Slack: Reaktionen + Systembenachrichtigungen der richtigen Session zuordnen (kein Überschneiden mit Haupt-Session). - Agent Tools: Berücksichtigen der `agent.tools` Allow/Deny-Policy auch wenn Sandbox deaktiviert ist. - Discord: Duplizierte Antworten vermeiden, wenn OpenAI wiederholt `message_end` emittiert. - Commands: `/status` (Inline) und Befehls-Auth über Provider hinweg vereinheitlichen; Gruppen-Bypass für autorisierte Control-Befehle; Discord `/clawd` Slash-Handler entfernen. - Control UI: Markdown in Chat-Nachrichten rendern (gesäubert).

Factory AI (Droid)vv0.42.2

Model Updates

5. Januar 2026

GLM 4.7 Modellbereitstellung und Gemini SDK-Upgrade.

OpenCodev1.1.1Breaking Change

v1.1.1

4. Januar 2026

This is a **major change** that overhauls the permissions system in OpenCode. ### Tools Merged into Permission The `tools` configuration has been deprecated and merged into the `permission` field. Previously, you could enable/disable tools like this: ```json { "tools": { "bash": true, "edit": false } } ``` Now, this should be configured using `permission`: ```json { "permission": { "bash": "allow", "edit": "deny" } } ``` The old `tools` config is still supported for backwards compatibility and will be automatically migrated to the permission system. ### Granular Permissions with Object Syntax Permissions now support granular control using an object syntax with pattern matching. When you specify a permission as an object, you can set different rules for different patterns: ```json { "permission": { "bash": { "npm *": "allow", "git *": "allow", "rm *": "deny", "*": "ask" }, "edit": { "*.md": "allow", "*.ts": "ask", "*": "deny" } } } ``` Each key in the object is a glob pattern that matches against the tool's input, and the value is the action to take: - `"allow"` - automatically approve - `"deny"` - automatically reject - `"ask"` - prompt the user for approval You can also set a blanket permission using a simple string: ```json { "permission": { "bash": "allow", "edit": "ask" } } ``` Or set all permissions at once: ```json { "permission": "allow" } ``` ### Breaking Changes for SDK Users The permission events have changed significantly. The new `PermissionNext` module (`permission/next.ts`) has a different event structure compared to the old `Permission` module (`permission/index.ts`): **Old Event Structure (`Permission.Event`):** - `Updated`: `{ id, type, pattern, sessionID, messageID, callID, message, metadata, time }` - `Replied`: `{ sessionID, permissionID, response }` **New Event Structure (`PermissionNext.Event`):** - `Asked`: `{ id, sessionID, permission, patterns, metadata, always, tool: { messageID, callID } }` - `Replied`: `{ sessionID, requestID, reply }` Key differences: - Event name changed from `permission.updated` to `permission.asked` - `type` renamed to `permission` - `pattern` is now `patterns` (array of strings) - `message` field removed - `response` renamed to `reply` - `permissionID` renamed to `requestID` - New `always` field contains patterns that would be approved for future requests if user selects "always" The reply values are the same: `"once"`, `"always"`, or `"reject"` ### Server Changes - New endpoint: `POST /permission/:requestID/reply` for responding to permission requests - Old endpoint `POST /session/:sessionID/permissions/:permissionID` is now deprecated - `GET /permission` now returns `PermissionNext.Request[]` instead of `Permission.Info[]` - Added CORS whitelist support via `server.cors` config option - Added Content-Type headers for proxied static assets - mDNS service name now includes port for uniqueness ### Other Changes - Agent `tools` field is deprecated - use `permission` instead - Agent `maxSteps` is deprecated - use `steps` instead - Unknown agent properties are now collected into `options` - Mode and plugin globs no longer search subdirectories (simplified to top-level only) ## Core - Add development scripts for better debugging workflow in TUI - Ensure @opencode-ai/plugin exists only on first run - Add MCP resources support (@paoloricciuti) - Initialize config in worktree - Prioritize free GPT-5-mini for small model in GitHub Copilot - Remove hardcoded .env read block and use new permissions model instead - Prevent duplicate plugin function initialization (@ShpetimA) - Update GitHub Copilot model priority list - Add variant field to chat message input - Update AGENTS.md documentation (@aryasaatvik) - Fixed dependency installation and git worktree branch creation - Fixed an issue in the codebase - Add sandbox support for git worktrees to allow working in multiple directories per project - Add timeout to MCP client connection calls (@RhysSullivan) - Add per-project MCP config overrides (@jknlsn) - Improve root detection for Gradle multi-project builds in Kotlin language server (@JBou) - Improve plugin loading to handle builtin plugin failures gracefully - Add reject message support to permission dialogs for better user feedback - Wrap queued user messages with reminder to stay on track - Add managed git worktrees - Upgrade OpenTUI to v0.1.68 using GPA - Add escape key handling to permission dialogs for better keyboard navigation - Add TUI session selection API endpoint for navigation (@code-yeongyu) - Add Kotlin LSP integration (@tjg184) - Merge instructions arrays across config files instead of overriding them (@elithrar) - Add variant support for minimal mode - Fix Cloudflare AI Gateway SDK chat undefined error (@englishm) - Handle NotFoundError for non-shared sessions in sync (@Hona) - Improve plugin loading to handle builtin plugin failures gracefully - Update CI configuration - Add reload functionality and improve lazy utility with reset capability - Remove memory leak fixes documentation after implementation - Fix memory leaks in session management and improve permission error handling - Fix TUI permission tests for new evaluate function signature - Improve permission error handling and evaluation logic in TUI - Ignore - Add assistant metadata to session export (@dmmulroy) - Ensure new permissions changes work for special case bash commands like rm, cd, etc - Rework permission system - Remove outdated Haiku filter for GitHub Copilot (@alcpereira) - Make install dependencies non-blocking - Use --no-cache flag when behind proxy to prevent connection hangs - Check for context overflow mid-turn in finish-step (@aryasaatvik) - Display error if invalid agent is used in a command (@Leka74) - Add Content-Type headers for proxied static assets (@monotykamary) ## TUI - Add missing theme list keybind in TUI (@aspiers) - Handle duplicate PR creation when agent creates PR (@elithrar) - Fix system theme diff highlighting in TUI - Add --variant flag to run command (@shuv1337) - Remove OpenRouter provider from priority list - Make LSP status icon muted when no LSPs are active (@itsrainingmani) - Make MCP status icon muted when no MCP servers are enabled (@itsrainingmani) - Pass attach directory to SDK client in TUI (@shuv1337) - Handle actions/checkout v6 credential storage change (@elithrar) - Use Bun.sleep instead of Promise with setTimeout (@edlsh) - Add Osaka Jade theme (@st-eez) - Add heap snapshot option to system menu for debugging memory usage - Fixed Windows fallback for "less" command in session list (@itsrainingmani) - Fix import command regex to properly handle file paths - Fix stats command day calculation and time filtering ## Desktop - Fix scroll position restoration in app - Don't override Ctrl+A on Windows - Improve auto-scroll behaviors in the application - Fixed editing projects in desktop application (@dbpolito) - Update server URL normalization to retain path (@OpeOginni) - Add image preview support in session viewer (@shuv1337) - Add /compact session command to compress conversation history - Adjust window drag region layout for desktop application - Relax request timeouts - Properly decode session ID for permission context in desktop app (@OpeOginni) - Work in progress on application features - Improve application startup time - Add work in progress desktop functionality - Add file context feature to app - Improve desktop window resize handle (@dbpolito) **Thank you to 30 community contributors:** - @monotykamary: - fix(server): add Content-Type headers for proxied static assets (#6587) - @Leka74: - fix: display error if invalid agent is used in a command (#6578) - @aryasaatvik: - fix(session): check for context overflow mid-turn in finish-step (#6480) - docs: update AGENTS.md (#6800) - @alcpereira: - fix: remove outdated Haiku filter for GitHub Copilot (#6593) - @dbpolito: - Desktop: Improve Resize Handle (#6608) - Desktop: Edit Project Fix (#6757) - @itsrainingmani: - fix: windows fallback for "less" cmd in `session list` (#6515) - fix(tui): make mcp status icon muted when no mcp servers are enabled (#6745) - fix(tui): make lsp status icon muted when no lsps are active (#6773) - @dmmulroy: - feat: add assistant metadata to session export (#6611) - @Hona: - fix(share): handle NotFoundError for non-shared sessions in sync (#6634) - @OpeOginni: - fix(desktop): Properly decode session id for permission context (#6580) - fix(server): update server URL normalization to retain path (#6647) - @albingroen: - fix(ui): fix slight vertical overflow in project selector (#6589) - @st-eez: - feat(theme): add Osaka Jade theme (#6609) - @edlsh: - refactor: use Bun.sleep instead of Promise setTimeout (#6620) - @englishm: - fix: cloudflare-ai-gateway sdk.chat undefined error (#6407) - @elithrar: - fix: handle actions/checkout v6 credential storage change (#6667) - fix: merge instructions arrays across config files (#6663) - github: handle duplicate PR creation when agent creates PR (#6777) - docs: add logging best practices for plugin authors (#6833) - @spoons-and-mirrors: - docs: add subtask2 to ecosystem page (#6704) - @tjg184: - feat: Add kotlin lsp integration (#6601) - @code-yeongyu: - feat(plugin): add tui.session.select API endpoint for TUI navigation (#6565) - @shuv1337: - fix(tui): pass attach directory to sdk client (#6715) - feat(app): add image preview support in session viewer (#6678) - feat: add --variant flag to run command (#6805) - @JBou: - fix(kotlin-ls): improve root detection for Gradle multi-project builds (#6717) - @jknlsn: - feat: add per-project MCP config overrides (#5406) - @sin4ch: - docs: enhance MCP servers documentation with a tip (#6713) - @RhysSullivan: - fix(mcp): add timeout to client.connect() calls (#6760) - @jerilynzheng: - docs: Add Vercel AI Gateway to provider docs (#6790) - @benjaminshafii: - docs: Add opencode-scheduler plugin to ecosystem (#6804) - @ShpetimA: - fix(plugin): prevent duplicate plugin function initialization (#6787) - @johnconnor-sec: - docs: typo in subtask documentation (#6821) - @felipeorlando: - Remove opencode-skills entry from ecosystem.mdx (#6817) - @jerome-benoit: - feat(nix): preliminary desktop app flake integration (#6135) - @paoloricciuti: - feat: mcp resources (#6542) - @aspiers: - fix(tui): add missing `theme_list` keybind (#6779)

OpenClawv2.0.0-beta5

clawdis 2.0.0-beta5

3. Januar 2026

### Fixed - Media: beibehalten GIF-Animation beim Hochladen zu Discord/anderen Anbietern (JPEG-Optimierung für image/gif überspringen). - Agent runtime: pi-mono Abhängigkeiten auf 0.31.1 aktualisieren (Agent-Core-Split). - Dependencies: auf neueste kompatible Versionen erhöhen (TypeBox, grammY, Zod, Rolldown, oxlint-tsgolint). - Tests: Abdeckung von Read Tool Image-Metadaten + Textausgabe. - Tests: Queue-Modus-Abdeckung hinzufügen (collect/followup + directive parsing). ### Breaking - Skills Config-Schema verschoben unter `skills.*`: - `skillsLoad.extraDirs` → `skills.load.extraDirs` - `skillsInstall.*` → `skills.install.*` - Mapping der einzelnen Skill-Konfigurationen wurde zu `skills.entries` verschoben (z. B. `skills.peekaboo.enabled` → `skills.entries.peekaboo.enabled`) - Neues optional gebundene Allowlist: `skills.allowBundled` (beeinflusst nur gebundene Skills) - Sessions: Gruppen-Schlüssel verwenden jetzt `surface:group:<id>` / `surface:channel:<id>`; Legacy-Keys `group:*` migrieren beim nächsten Message; `groupdm`-Keys werden nicht mehr erkannt. - Discord: veraltete `discord.allowFrom`, `discord.guildAllowFrom` und `discord.requireMention` entfernen; verwenden Sie `discord.dm` + `discord.guilds`. - Providers: Discord/Telegram starten nicht mehr automatisch aus Token-Umgebungsvariablen; fügen Sie `discord: { enabled: true }` / `telegram: { enabled: true }` zu Ihrer Konfiguration hinzu, wenn Sie `DISCORD_BOT_TOKEN` / `TELEGRAM_BOT_TOKEN` verwenden. - Config: entfernen Sie `routing.allowFrom`; verwenden Sie stattdessen `whatsapp.allowFrom` (führen Sie `clawdis doctor` zur Migration aus). - Config: entfernen Sie `routing.groupChat.requireMention` + `telegram.requireMention`; verwenden Sie stattdessen Standardwerte von `whatsapp.groups`, `imessage.groups` und `telegram.groups` (führen Sie `clawdis doctor` zur Migration aus). ### Features - Discord: erweitern Sie Aktionen des Tools `discord` (Reaktionen, Sticker, Umfragen, Threads, Suche, Moderations-Gates) (#115) — Dank an @thewilloftheshadow. - Discord/Telegram: Antworten-Tags hinzufügen (`[[reply_to_current]]`, `[[reply_to:<id>]]`) mit provider-spezifischem `replyToMode` (off|first|all) für native Threaded Replies. - Talk-Modus: kontinuierliche Sprachgespräche (macOS/iOS/Android) mit ElevenLabs TTS, Antwort-Direktiven und optionalem Unterbrechungsauslöser bei Sprache. - Auto-Antwort: erweiterte Queue-Modi (steer/followup/collect/steer-backlog) mit Debounce-/Limit-/Drop-Optionen und Followup-Backlog-Verarbeitung. - UI: Optionales `ui.seamColor` Akzent zur Tönung der Talk Mode Seitenblase (macOS/iOS/Android). - Nix-Modus: deklarative Konfiguration + schreibgeschützte Einstellungs-UI, wenn `CLAWDIS_NIX_MODE=1` (danke an @joshp123 für die Persistenz — vertraute mir; ich integriere das künftig). - CLI: Google Antigravity OAuth-Auth-Option für Claude Opus 4.5/Gemini 3 (#88) — Dank an @mukhtharcm. - Agent Runtime: akzeptieren Sie legacy `Z_AI_API_KEY` für Z.AI Provider-Auth ( mapped zu `ZAI_API_KEY` ). - Gruppen: pro-Gruppe Mentions-Gating-Defaults/Overrides für Telegram/WhatsApp/iMessage via `*.groups` mit `"*"` Defaults; Discord unterstützt nun `discord.guilds."*"` als Default. - Discord: Benutzer-installierte Slash-Command-Handhabung mit per-User Sessions und automatischer Registrierung (#94) — Danke an @thewilloftheshadow. - Discord: DM-Enable/Allowlist plus Guild Channel/User/Guild Allowlists mit ID/Name Matching. - Signal: JSON-RPC-Support via `signal-cli` zum Senden/Empfangen über den Signal Provider. - iMessage: iMsg JSON-RPC-Integration (stdio), Chat-ID-Routing und Group-Chat-Unterstützung. - Chat-UI: Recent-Session-Dropdown-Switcher (erstes Main) in macOS/iOS/Android + Control UI. - UI: Discord/Signal/iMessage Verbindungs-Panels in macOS + Control UI (Danke an @thewilloftheshadow). - Discord: erlauben Sie agenten-getriggerte Reaktionen via `clawdis_discord`, wenn aktiviert, und surface Message-IDs im Kontext. - Discord: überarbeitet Guild Routing-Konfiguration mit pro-Guild/Channel-Regeln und slugged Display-Namen; optionale Group-DM-Unterstützung (Standard aus). - Discord: entfernen veralteter Guild/Channel Ignore-Listen zugunsten von pro-Guild Allowlists (und vorgeschlagene pro-Guild Ignore-Listen). - Skills: Trello-Skill für Boards/Listen/Karten-Management (Danke an @clawd). - Docker: containerisierte Gateway/CLI Setup via Dockerfile, Compose und Setup-Skript (Danke an @dan-dr). - Tests: Live-Z.AI Test-Gate für Smoke-Validierung bei vorhandenem Key. - macOS Debug: App-Log-Verbose-Modus und Rolling-File-Log-Umschaltung für swift-log-basierte App-Logs. - CLI: Onboarding-Assistent (Gateway + Workspace + Skills) mit Daemon-Installern und Anthropic/Minimax Setup-Pfaden. - CLI: ASCII-Banner-Header zu Wizard-Entry-Punkten hinzufügen. - CLI: Wizards für `configure`, `doctor` und `update` für fortlaufende Einrichtung, Health-Checks und Modernisierung. - CLI: Signal CLI Auto-Install aus GitHub-Releases im Wizard und Persistierung der Run-Metadaten im Config. - CLI: Remote Gateway Client Config `gateway.remote.*` mit Bonjour-unterstützter Erkennung. - CLI: `clawdis tui` gateway-verbundene Terminal-UI (lokal oder remote). - Skills: Erlauben Sie `bun` als Node-Manager für Skill-Installationen. - Skills: Trello-like Things-Mac (Things 3 CLI) für Read/Search plus Add/Update via URL-Schema. - Skills: Apple Notes + Reminders Skills via Memo-CLI (Danke an @tylerwince). - Tests: Docker-basierte Onboarding-E2E-Harness hinzufügen. - Tests: Wizard-E2E-Flows für Reset, Provider, Skills und Remote non-interaktive Runs härten. - Browser-Tools: Remote CDP URL-Unterstützung, Linux-Launcher-Optionen (`executablePath`, `noSandbox`) und Surface `cdpUrl` im Status. - Skills: tmux-basierter Coding-Agent-Skill + `requires.anyBins` Gate für Multi-CLI-Setup (Danke an @sreekaransrinath). ### Fixes - Gog-Kalender: Datumsbereiche RFC 3339 mit Zeitzone formatieren, um Google Calendar API zu erfüllen (Danke an @jayhickey). - macOS-Onboarding: Scrollbarer Seitenrand für überlaufenden Inhalt (#105) — Danke an @thewilloftheshadow. - Chat-UI: Chat-Ansicht nach Wechseln der Sessions bis zum neuesten Nachricht beibehalten. - Chat-UI: Anzeigen von Rich Session Display-Namen in Web Chat + SwiftUI + Android. - Auto-Antwort: Streaming von fertigen Antwort-Blöcken, sobald sie abgeschlossen sind (konfigurierbare Default-Einstellung + Break); leere Tool-Only-Blocks überspringen, außer verbose. - Discord: Duplizierte Senden vermeiden, wenn Block-Streaming aktiv ist (Race mit Typing-Hook). - Providers: Outbound-Text-Chunk-Limits konfigurierbar machen via `*.textChunkLimit` (Standards bleiben 4000/Discord 2000). - CLI-Onboarding: Gateway-Token in Config persistieren, damit lokales CLI-Auth funktioniert; Auth Off empfehlen, sofern kein Multi-Machine Access benötigt wird. - Control-UI: Akzeptieren Sie einen `?token=` Query-Parameter zum Auto-Ausfüllen der Gateway-Auth; Onboarding öffnet nun das Dashboard mit Token-Auth, wenn konfiguriert. - Agent-Prompt: Entfernen Sie hartcodierten Benutzernamen im System-Prompt-Beispiel. - Chat-UI: Extra oberes Padding vor dem ersten Nachrichten-Bubble in Web Chat (macOS/iOS/Android). - Control-UI: Verfeinern Sie das Web Chat Session-Selector-Layout (Chevron-Abstand + Hintergrund). - WebChat: Streamen Sie Live-Updates für Sessions, auch wenn Runs außerhalb der Chat-UI starten. - Gateway CLI: `CLAWDIS_GATEWAY_PASSWORD` aus der Umgebung in `callGateway()` lesen — ermöglicht Auth mit `doctor`/`

OpenCodev1.0.223

v1.0.223

1. Januar 2026

## TUI - Don't show 'Agent not found' toast for subagents ## Desktop - Add context window configuration to app

OpenCodev1.0.222

v1.0.222

1. Januar 2026

## Core - Add Windows support for PTY and cross-platform build scripts in desktop application (@Hona) - Prevent TimeoutOverflowWarning by capping setTimeout delay to maximum 32-bit signed integer value ## Desktop - Add in-app update toasts to desktop application - Add subagent mention support to desktop application (@dbpolito) **Thank you to 2 community contributors:** - @dbpolito: - Desktop: Add Subagents Mention Support (#6540) - @Hona: - fix(desktop): Windows support for PTY and cross-platform build scripts (#6508)

OpenCodev1.0.221

v1.0.221

31. Dezember 2025

## Core - Default to ai-sdk/openai-compatible when no npm package is provided - Add support for LSP workspace/didChangeWatchedFiles notifications (@marceldev89) ## Desktop - Improved model selector in app - Auto-accept colors in app - Fix tooltip colors in desktop application - Better affordance for auto-accept in desktop app - Add input changes functionality - Don't show notifications when auto-accepting tasks in desktop app - More defensive access in desktop application - Better variant toggle in app - Add keybind tooltip component to app - Fix desktop prompt input cleanup behavior **Thank you to 2 community contributors:** - @marceldev89: - Add support for LSP workspace/didChangeWatchedFiles (#6524) - @vtemian: - fix(tui): don't show 'Agent not found' toast for subagents (#6528)

OpenCodev1.0.220

v1.0.220

31. Dezember 2025

## Core - Fix plugin and mode glob patterns - Add debug agent subcommand to test agents in isolation (@Eric-Guo) - Display MCP tag for prompts in autocomplete but not in prompt (@paoloricciuti) ## TUI - Clarify agent-name placeholder in tips (@anzulaqeel-anunzio) ## Desktop - Remove status bar and add new elements in header - Better new session button in desktop app - Add unified diff toggle to desktop application - Fix desktop not rendering large sessions **Thank you to 4 community contributors:** - @paoloricciuti: - fix: display MCP tag for prompts in autocomplete but not in prompt (#6531) - @anzulaqeel-anunzio: - fix: Clarify agent-name placeholder in tips (#6520) - @Eric-Guo: - feat(cli): New debug agent <name> subcommand (#6529) - @backnotprop: - docs: Add plannotator plugin (#6510)

OpenCodev1.0.219

v1.0.219

31. Dezember 2025

## Core - MDNS no longer fails when service is already registered - Allow configurable CORS hosts - Add MCP prompts as slash commands (@paoloricciuti) - Add Gemini 3 Flash to fast models list (@junmediatek) - Bundle @ai-sdk/vercel version 1.0.31 for AI SDK v5 support (@OpeOginni) - Cleaner view for subagents hint text - Add configurable timeout for MCP tool calls to prevent hanging requests - Read plural resource types and stop erroring on them - Fix OpenAI variants for Codex models - Add model usage statistics with input/output token breakdown to stats command (@processtrader) - Add rustfmt formatter for Rust files (@frhd) - Ensure variants work for completely custom models ## TUI - Adjust keys for uniqueness calculations to use provider/model ## Desktop - Don't open native folder select when connected to remote server - Add model variants support to app - Better notification icon in desktop app - Upgrade to latest version on error page in desktop app **Thank you to 6 community contributors:** - @frhd: - feat(format): add rustfmt formatter for Rust files (#6482) - @processtrader: - feat: add model usage statistics with input/output token breakdown to stats command (#6296) - @brianclinkenbeard: - Update mise install command (#6504) - @OpeOginni: - feat: bundle in @ai-sdk/vercel version 1.0.31 for aisdk v5 support (#6512) - @junmediatek: - feat: add gemini-3-flash to fast models list (#6497) - @paoloricciuti: - feat: mcp prompts as slash commands (alternative) (#5767)

Gemini CLIv0.22.5

Release v0.22.5

30. Dezember 2025

## Was sich geändert hat * fix(patch): cherry-pick 07e597d to release/v0.22.4-pr-15684 [CONFLICTS] by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15733 **Vollständiger Changelog**: https://github.com/google-gemini/gemini-cli/compare/v0.22.4...v0.22.5

OpenCodev1.0.218

v1.0.218

30. Dezember 2025

## Core - Add CORS exception for Tauri desktop application - Add Prisma language server (@ravshansbox) - Add extract reasoning middleware (@snipeship) **Thank you to 2 community contributors:** - @snipeship: - feat: add extract reasoning middleware (#6463) - @ravshansbox: - Add prisma language server (#6462)