Changelog

Claude Agent SDK

Anthropics Agent SDK, hier mit TypeScript-Releases

Aktuell geprüft

Zuletzt aus offizieller Quelle geprüft: 7. Nov. 2026, 15:13 MEZ

Die Erfassung aus offiziellen Quellen ist begrenzt. Ältere Releases können fehlen. Diese Seite beansprucht keine vollständige Versionshistorie.

Versionshinweise ohne deutsche Übersetzung stehen im englischen Original.

Neueste Version:v0.3.319

Geladene Releases:95

Quelle:GitHub Releases

RSS

Neueste Releases (95)

v0.3.319Neu

v0.3.319

6. November 2026

•Added MCP task support in SDK sessions (not on Bedrock, Vertex or Foundry): with CLAUDE_AUTO_BACKGROUND_TASKS=1 and streaming input, an MCP task runs in the background with task_type "mcp_task"
•Fixed the OpenTelemetry claude_code.session.count metric counting prewarm() processes that were never claimed; a pre-started process now counts once, when SpareProcess.claim() succeeds
•Fixed session_writes_lost listing the wrong session in transcript_session_ids after /clear in a session resumed from a transcript file outside the projects folder
•Improved deleteSession() to reject with ENOTEMPTY if the session's folder could not be fully deleted
•Changed session_writes_lost to be sent only when the CLI exits 0; a run that ends on an error result now gets only the stderr line
•Updated to parity with Claude Code v2.1.319

v0.3.318Neu

v0.3.318

4. November 2026

•Added an optional credential_expires_at field to result messages: when the stored claude.ai sign-in in use ends
•Added the config_file_invalid startup failure reason, reported when a configuration file read at startup isn't valid JSON
•Fixed the interrupt response missing a message held behind a model change: it is now listed, cancelled with cancel_queued, and stopped at start-up as a queued message is
•Changed agentProgressSummaries: while the subagent waits for a response on a likely-cold prompt cache, a summary can wait up to 5 minutes, and is not sent if the subagent ends first
•Changed status in PushNotificationInput to optional
•Updated to parity with Claude Code v2.1.318

v0.3.317Neu

v0.3.317

3. November 2026

•Added onFailure to hook matchers: with "block", a hook that throws or returns invalid output blocks the action
•Added a warning message in the stream when a hook callback fails
•Added a session_writes_lost system message, sent after the last result when a full disk or disk quota kept the CLI from saving the session transcript or ~/.claude.json
•Added initialize_not_first to startup_failure_reason, for a --await-initialize launch whose first stdin line is not the initialize request
•Added task.description to the TaskCreate tool's tool_use_result: the description the task was created with
•Added lossy to readFile() utf-8 answers: true when decoding replaced bytes that are not valid UTF-8
•Removed the file_unchanged variant of FileReadOutput in sdk-tools.d.ts; a repeated Read now returns the file content
•Fixed setMcpServers closing a folder's .mcp.json and local-scope MCP servers in sessions started with prewarm() and bound with SpareProcess.claim()

Mehr anzeigen (1)

•Updated to parity with Claude Code v2.1.317

v0.3.316

31. Oktober 2026

•Added permissions.blockReadsExemptPaths to the Settings type: paths, read only from --settings and settings the host sends, that blockReadsOutsideWorkingDirectories lets through without making them working directories
•Updated to parity with Claude Code v2.1.316

v0.3.314

30. Oktober 2026

•Added sandbox_required_unavailable to startup_failure_reason, for the result Claude Code writes when a required sandbox cannot run
•Added permissions.blockWritesOutsideWorkingDirectories to the Settings type
•Added prewarm() (alpha) to the @anthropic-ai/claude-agent-sdk/core entry point
•Fixed a settings file left out of settingSources still supplying some settings, such as auto mode rules and saved permission consents
•Updated to parity with Claude Code v2.1.314

v0.3.312

27. Oktober 2026

•Added autoMode, skipAutoPermissionPrompt, useAutoModeDuringPlan and permissions.disableAutoMode to the Settings type
•Added unanswered (skipped, replied_in_chat or dismissed) to AskUserQuestion answers' per-question annotations; send it only when system/init lists the ask_user_question_unanswered_v1 capability
•Fixed resolveSettings() returning mistyped managed auto-mode settings, such as permissions.disableAutoMode: true, as written instead of reading them as the CLI does; a user or project settings file whose auto-mode keys hold an invalid value is now skipped, as in the CLI
•Updated to parity with Claude Code v2.1.312

v0.3.311

26. Oktober 2026

•Added cacheCreationEphemeral5mInputTokens and cacheCreationEphemeral1hInputTokens to ModelUsage, splitting cacheCreationInputTokens by cache lifetime
•Changed background start-up work to wait until the first model request is sent, for about 10 seconds at most, when the first message arrives during start-up
•Changed the minimum supported Node.js version from 18.0.0 to 18.14.0
•Updated to parity with Claude Code v2.1.311

v0.3.310

24. Oktober 2026

•Updated to parity with Claude Code v2.1.310

v0.3.309

23. Oktober 2026

•Fixed host-supplied managedSettings dropping disableAllHooks, disableSkillShellExecution and empty allowedHttpHookUrls / httpHookAllowedEnvVars lists
•Changed extractFromBunfs() to extract the embedded CLI into its own per-user temp directory (claude-agent-sdk-<uid>, or claude-agent-sdk on Windows) instead of Claude Code's temp root
•Changed extractFromBunfs() to reuse an already-extracted file only when its bytes match the embedded binary, replacing it otherwise
•Updated to parity with Claude Code v2.1.309

v0.3.308

22. Oktober 2026

•Updated to parity with Claude Code v2.1.308

v0.3.307

22. Oktober 2026

•Added wakeup_source to the command_lifecycle started frame of a scheduled wakeup or task, and scheduled_for for a wakeup (capability lifecycle_wakeup_source_v1), so a host can tell which one fired
•Added supportedDialogKinds to browser SDK query(), sent on initialize; a non-empty list requires onUserDialog
•Updated to parity with Claude Code v2.1.307

v0.3.306

21. Oktober 2026

•Updated to parity with Claude Code v2.1.306

v0.3.305

20. Oktober 2026

•Added recommendations and fix to the get_usage response's behavior windows
•Fixed servers set with setMcpServers being replaced later in the session by a same-named server from the user's own settings
•Fixed startup() with initializeTimeoutMs: null counting time spent in SessionStart or Setup hooks against its 60-second limit; null now works like leaving the option out
•Changed the startup() and prewarm() timeout error for a start with no open SessionStart or Setup hook: it lists possible causes and no longer says to check authentication and network
•Updated to parity with Claude Code v2.1.305

v0.3.304

19. Oktober 2026

•Added suspendSseStream(query) and resumeSseStream(query) to the browser SDK: close a query's event stream and reopen it later without ending the query
•Added stable row ids to getContextUsage(), and, when the CLI answers on its own stdio, plugin ids, agent and skill file paths, and per-server MCP instruction token shares
•Fixed startup() timing out while a SessionStart or Setup command hook was running when no initializeTimeoutMs is passed: up to 10 minutes of that time no longer counts toward the 60-second limit, and the timeout error now says what the process reported
•Fixed startup() keeping every hook_progress message of a running SessionStart or Setup hook in memory until it resolved; it now keeps only the newest one for each hook
•Updated to parity with Claude Code v2.1.304

v0.3.303

19. Oktober 2026

•Updated to parity with Claude Code v2.1.303

v0.3.302

16. Oktober 2026

•Fixed structured output results delivering a response's output as structured_output when the API refused that response; the result is now an error with no payload
•Changed an omitted permissionMode to start in default mode on third-party providers, with telemetry off or on a fresh install; pass permissionMode: 'auto' to keep auto mode
•Changed fork subagents (subagent_type: "fork") to be on by default in SDK sessions, as in the terminal; run_in_background is still offered, and CLAUDE_CODE_FORK_SUBAGENT=0 turns them off
•Updated to parity with Claude Code v2.1.302

v0.3.301

15. Oktober 2026

•Added per-tool permission policies to sdk-type MCP server configs: tools entries with permission_policy always_ask or always_deny, applied on initialize and setMcpServers()
•Added BUILTIN_TOOL_NAMES and LEGACY_TOOL_NAME_ALIASES exports: the public built-in tool names and their earlier names, for checking permission rules and tool settings
•Added addPlugins to Query.reloadPlugins() so a host can load more plugin folders into a running session
•Fixed setModel() with a model ID failing while the account is at its five-hour or weekly usage limit; the model now applies and the next message reports the limit
•Changed the error when a full disk keeps the login from being refreshed: error is now invalid_request, not authentication_failed, and api_error_status is null, not 401
•Changed mcpServerStatus() to still report a tool's _meta.ui.visibility when the rest of its _meta.ui fails validation
•Updated to parity with Claude Code v2.1.301

v0.3.300

14. Oktober 2026

•Fixed importSessionToStore skipping transcript rows that contain a Unicode line or paragraph separator (U+2028, U+2029) on Node 24 and Bun
•Fixed a background subagent's final report in auto mode being joined into one turn with another message from the same subagent; the report now always runs as its own turn
•Fixed rewind_conversation with interrupt_if_running stopping the running turn before refusing with "unseen later turn"
•Reverted canUseTool options carrying suppressAlwaysAllowRule: true for connector tools that an organization set to require approval (added in 0.3.292)
•Improved the browser SDK bundle (@anthropic-ai/claude-agent-sdk/browser): 40% smaller to download, by no longer bundling a Node crypto polyfill
•Changed thinking: { type: 'disabled' } and maxThinkingTokens: 0 to run as adaptive thinking on models that can't turn thinking off
•Changed updatedInput from PreToolUse and PermissionRequest hook callbacks: it is now validated like Claude's own input and unknown keys are dropped; canUseTool is unchanged
•Updated to parity with Claude Code v2.1.300

v0.3.299

14. Oktober 2026

•Added an optional nextFireAt (epoch ms of the next run) to the CronCreate and CronList tool outputs
•Added property_order to the elicitation control request: a form's field names in the order the MCP server listed them, for hosts whose transport keeps no JSON object key order
•Fixed rewind_conversation answering "target not found" for a message stopped during its prompt hooks or plugins: until a reply or input follows, it answers rewound: true with a null prefillText
•Fixed in-process SDK MCP servers named constructor or toString being silently dropped when a prewarm() process is claimed with SpareProcess.claim()
•Fixed queued user messages with content-block arrays running together without a line break when merged into one turn
•Fixed origin.body on a user message that merges several queued peer messages (from other sessions or agents) carrying only the first message's text; it is now omitted
•Changed initialize, mcp_set_servers and claim_session to refuse the reserved MCP server names __proto__, __defineGetter__, __defineSetter__, __lookupGetter__, __lookupSetter__
•Updated to parity with Claude Code v2.1.299

v0.3.296

9. Oktober 2026

•Added claude_code_version to the initialize control response so clients know the CLI version before the first turn
•Added autoCompactWindow to AgentDefinition, so a subagent can auto-compact earlier than the main conversation's window
•Fixed the sandbox option discarding an inline settings.sandbox block: the two now merge, the option's values win, and deny and credential lists from both sides are combined
•Changed the default limit on MCP tool descriptions sent up front and on MCP server instructions, in-process SDK servers included, from 2,048 to 4,096 characters
•Changed permission answers that arrive after a restart to be parsed like live ones: over 4,096 updatedPermissions entries count as a denial, and a malformed list is ignored as a whole
•Updated to parity with Claude Code v2.1.296

v0.3.295

8. Oktober 2026

•Added overageEnabled to SDKRateLimitInfo: on usage-limit warnings, whether the account has extra usage turned on
•Fixed an MCP server added with mcp_set_servers sometimes keeping an outdated tool list when another server in the same request connected later
•Fixed assistant text blocks losing their citations in streamed responses
•Fixed Claude being told "disabled by the user" about an in-process MCP server that was switched off with toggleMcpServer()
•Fixed rate_limit_event with status allowed_warning omitting overageStatus, overageResetsAt and overageDisabledReason, which allowed and rejected events already carry
•Changed MCP tool descriptions the model loads through tool search, including those from in-process SDK servers, to be cut at 16,384 characters instead of 2,048
•Changed a permission answer whose updatedPermissions holds over 4,096 updates, rules and directories to count as a denial
•Changed tool_result_meta[].non_execution_kind to be absent when a mod withheld the result of a tool that ran without error; it was permission-rule

Mehr anzeigen (6)

•Changed pasted_content on user messages to be ignored when it holds over 1,000 entries and blocks in all
•Changed how the SDK passes option values to Claude Code: the values of its named options are now sent in the same argument as their flag (--flag=value)
•Changed tool_use_result for MCP Apps tools: structuredContent or _meta over 8,388,608 JSON characters is left off, and _meta.ui with no ui:// resource no longer lifts the ordinary caps
•Changed tool_use_result for MCP tools: when a long result is saved to a file, or cut because it could not be, the text kept past 500,000 characters is dropped, with a note of how much
•Changed tool_use_result for MCP tools: a result _meta over 1,048,576 JSON characters is left off, except for SDK-server tools; so is text blocks' _meta that totals over 65,536
•Updated to parity with Claude Code v2.1.295

v0.3.294

8. Oktober 2026

•Updated to parity with Claude Code v2.1.294

v0.3.293

7. Oktober 2026

•Added an optional subagent_type to background_tasks_changed task entries, so hosts can name each subagent's type without pairing with task_started
•Updated to parity with Claude Code v2.1.293

v0.3.292

6. Oktober 2026

•Added agent_id to the assistant and user messages a subagent produces; it equals the task_id on that subagent's task events and stays the same when the subagent is resumed
•Added parent_task_id to task_started events and background_tasks_changed entries, naming the subagent task that launched a task
•Added run_id to background task events and origin.runId to task notifications, so a host can tell a resumed task's runs apart
•Added typed sections and notes to the ListAgents tool's tool_use_result, so hosts can list agents without parsing its text
•Added recipient_kind to SendMessage tool_use inputs on assistant messages, and their top-level approve is now always a boolean, so hosts no longer infer them from the raw input
•Added two optional fields to the ReadNotifications tool output: read_at, when the tool read the session's queued notifications, and arrived_at on each notification, when it reached the session
•Fixed subagents with a declared auto permission mode having their tool calls decided by the auto-mode classifier instead of the canUseTool callback when auto mode is unavailable
•Changed background_tasks_changed for a finishing background task to arrive after that task's task_updated and task_notification

Mehr anzeigen (2)

•Changed canUseTool options to carry suppressAlwaysAllowRule: true for connector tools that an organization set to require approval, so hosts can hide their "always allow" choice
•Updated to parity with Claude Code v2.1.292

v0.3.291

6. Oktober 2026

•Updated to parity with Claude Code v2.1.291

v0.3.290

6. Oktober 2026

•Added an optional offset field to the WebFetch tool input for reading on through long pages
•Fixed failed Claude in Chrome tool calls dropping the result's _meta: when the result has one, tool_use_result is now { content, _meta }, as for other MCP tools
•Fixed deny and ask rules missing toolAliases tools when written with a wildcard (Bash*), a match-all pattern (Bash(**), also on the mapped name), or an input field also in disallowedTools
•Fixed a user message sent again under the same uuid being replayed (--replay-user-messages) while the first copy still waited for its turn; its replay now comes when a turn takes it
•Fixed includePartialMessages streams leaving a message without message_stop when the stream was cut, interrupted or fell back to non-streaming
•Changed user_message_uuid, user_message_uuids and resume_reason to also be set on a turn a restarted worker resumes from a permission answer or from tool calls still in flight
•Updated to parity with Claude Code v2.1.290

v0.3.289

4. Oktober 2026

•Updated to parity with Claude Code v2.1.289

v0.3.288

2. Oktober 2026

•Updated to parity with Claude Code v2.1.288

v0.3.287

1. Oktober 2026

•Added optional remote-session latency fields (first_text_post_queue_wait_ms, first_text_post_queued_behind) to the success result message
•Fixed includePartialMessages streams sending a cut-short reply's message_stop late or never, so apps could show the reply as still in progress
•Fixed the error for a revoked claude.ai login, which now reads "Failed to authenticate: OAuth token revoked" instead of a generic or "does not have access" message
•Fixed a tool call to an in-process MCP server being left waiting after toggleMcpServer() disabled the server or setMcpServers() removed it
•Fixed commands_changed arriving before init, or twice, at session start: the initialize response now includes commands registered at startup
•Changed tool_use_result for a WebFetch or WebSearch call that steps aside for a priority "now" message to { detachedToolCall: true }; the result follows in a later turn
•Changed tool_use_result for MCP tools: structuredContent over 1,048,576 JSON characters is left off and structuredContentOmitted: true set, except for SDK-server and MCP Apps tools
•Updated to parity with Claude Code v2.1.287

v0.3.286

30. September 2026

•Added the initialize response field sdk_mcp_manifests_parked and the system/init capabilities sdk_mcp_manifests and sdk_mcp_tools_list_changed
•Fixed foreground subagents sometimes not receiving the task-tracking tools listed in tools or allowedTools
•Fixed an SDK MCP server listing no tools when one tool's schema cannot be converted to JSON Schema; that tool is now left out with a warning naming it
•Fixed toggleMcpServer() not disconnecting, and failing to re-enable, an in-process MCP server created with createSdkMcpServer(), except one named claude-in-chrome
•Changed a person's priority now message to move running shell commands, agents and MCP calls to the background and join the running turn instead of stopping it
•Changed the TypeScript Agent SDK to leave an omitted permissionMode to Claude Code, so a settings defaultMode now applies and, on third-party providers or with telemetry off, the session starts in auto mode like claude -p; pass permissionMode: 'default' for manual approvals
•Updated to parity with Claude Code v2.1.286

Zeige 30 von 95