Anthropic Release Notes

• Added interactive Google Vertex AI setup wizard accessible from the login screen when selecting "3rd-party platform", guiding you through GCP authentication, project and region configuration, credential verification, and model pinning
• Added CLAUDE_CODE_PERFORCE_MODE env var: when set, Edit/Write/NotebookEdit fail on read-only files with a p4 edit hint instead of silently overwriting them
• Added Monitor tool for streaming events from background scripts
• Added subprocess sandboxing with PID namespace isolation on Linux when CLAUDE_CODE_SUBPROCESS_ENV_SCRUB is set, and CLAUDE_CODE_SCRIPT_CAPS env var to limit per-session script invocations
• Added --exclude-dynamic-system-prompt-sections flag to print mode for improved cross-user prompt caching
• Added workspace.git_worktree to the status line JSON input, set whenever the current directory is inside a linked git worktree
• Added W3C TRACEPARENT env var to Bash tool subprocesses when OTEL tracing is enabled, so child-process spans correctly parent to Claude Code's trace tree
• LSP: Claude Code now identifies itself to language servers via clientInfo in the initialize request
• Fixed a Bash tool permission bypass where a backslash-escaped flag could be auto-allowed as read-only and lead to arbitrary code execution
• Fixed compound Bash commands bypassing forced permission prompts for safety checks and explicit ask rules in auto and bypass-permissions modes …

Claude Cowork generally available

Claude Cowork is now generally available on macOS and Windows through the Claude Desktop app. With this, we’re expanding Claude Cowork with new capabilities:

• Claude Cowork in the Analytics API. For more information, see Analytics APIs.

• Usage analytics for Claude Cowork. For more information, see View usage analytics for Team and Enterprise plans.

• Claude Cowork OpenTelemetry support. For more information, see Monitor Claude Cowork activity with OpenTelemetry.

Role-based access controls for Enterprise plans

Admins can now organize users into groups, manually or via SCIM from your identity provider, and assign each group a custom role defining which Claude capabilities its members can use. Turn Claude Cowork on for specific teams, restrict features by department and adjust as adoption grows. Learn more about setting up role-based permissions, managing group spend limit[s](https://support.claude.com/en/articles/13799932-manage-grou…

• Added focus view toggle (Ctrl+O) in NO_FLICKER mode showing prompt, one-line tool summary with edit diffstats, and final response
• Added refreshInterval status line setting to re-run the status line command every N seconds
• Added workspace.git_worktree to the status line JSON input, set when the current directory is inside a linked git worktree
• Added ● N running indicator in /agents next to agent types with live subagent instances
• Added syntax highlighting for Cedar policy files (.cedar, .cedarpolicy)
• Fixed --dangerously-skip-permissions being silently downgraded to accept-edits mode after approving a write to a protected path
• Fixed and hardened Bash tool permissions, tightening checks around env-var prefixes and network redirects, and reducing false prompts on common commands
• Fixed permission rules with names matching JavaScript prototype properties (e.g. toString) causing settings.json to be silently ignored
• Fixed managed-settings allow rules remaining active after an admin removed them until process restart
• Fixed permissions.additionalDirectories changes in settings not applying mid-session
• Fixed removing a directory from settings.permissions.additionalDirectories revoking access to the same directory passed via --add-dir
• Fixed MCP HTTP/SSE connections accumulating ~50 MB/hr of unreleased buffers when servers reconnect …

• We've launched Claude Managed Agents in public beta, a fully managed agent harness for running Claude as an autonomous agent with secure sandboxing, built-in tools, and server-sent event streaming. Create agents, configure containers, and run sessions through the API. All endpoints require the managed-agents-2026-04-01 beta header. Learn more in Claude Managed Agents overview.
• We've launched the ant CLI, a command-line client for the Claude API that enables faster interaction with the Claude API, native integration with Claude Code, and versioning of API resources in YAML files. Learn more in the CLI quickstart.

• Added support for Amazon Bedrock powered by Mantle, set CLAUDE_CODE_USE_MANTLE=1
• Changed default effort level from medium to high for API-key, Bedrock/Vertex/Foundry, Team, and Enterprise users (control this with /effort)
• Added compact Slacked #channel header with a clickable channel link for Slack MCP send-message tool calls
• Added keep-coding-instructions frontmatter field support for plugin output styles
• Added hookSpecificOutput.sessionTitle to UserPromptSubmit hooks for setting the session title
• Plugin skills declared via "skills": ["./"] now use the skill's frontmatter name for the invocation name instead of the directory basename, giving a stable name across install methods
• Fixed agents appearing stuck after a 429 rate-limit response with a long Retry-After header — the error now surfaces immediately instead of silently waiting
• Fixed Console login on macOS silently failing with "Not logged in" when the login keychain is locked or its password is out of sync — the error is now surfaced and claude doctor diagnoses the fix
• Fixed plugin skill hooks defined in YAML frontmatter being silently ignored
• Fixed plugin hooks failing with "No such file or directory" when CLAUDE_PLUGIN_ROOT was not set
• Fixed ${CLAUDE_PLUGIN_ROOT} resolving to the marketplace source directory instead of the installed cache for local-marketplace plugins on startup
• Fixed scrollback showing the same diff repeated and blank pages in long-running sessions …

• We announced Claude Mythos Preview is available as a gated research preview for defensive cybersecurity work as part of Project Glasswing. Access is invitation-only.
• The Messages API is now available on Amazon Bedrock as a research preview. The new Claude in Amazon Bedrock endpoint at /anthropic/v1/messages uses the same request shape as the first-party Claude API and runs on AWS-managed infrastructure with zero operator access. Available in us-east-1; contact your Anthropic account executive to request access. Learn more in Claude in Amazon Bedrock.

• Added forceRemoteSettingsRefresh policy setting: when set, the CLI blocks startup until remote managed settings are freshly fetched, and exits if the fetch fails (fail-closed)
• Added interactive Bedrock setup wizard accessible from the login screen when selecting "3rd-party platform" — guides you through AWS authentication, region configuration, credential verification, and model pinning
• Added per-model and cache-hit breakdown to /cost for subscription users
• /release-notes is now an interactive version picker
• Remote Control session names now use your hostname as the default prefix (e.g. myhost-graceful-unicorn), overridable with --remote-control-session-name-prefix
• Pro users now see a footer hint when returning to a session after the prompt cache has expired, showing roughly how many tokens the next turn will send uncached
• Fixed subagent spawning permanently failing with "Could not determine pane count" after tmux windows are killed or renumbered during a long-running session
• Fixed prompt-type Stop hooks incorrectly failing when the small fast model returns ok:false, and restored preventContinuation:true semantics for non-Stop prompt-type hooks
• Fixed tool input validation failures when streaming emits array/object fields as JSON-encoded strings
• Fixed an API 400 error that could occur when extended thinking produced a whitespace-only text block alongside real content …

• Added MCP tool result persistence override via _meta["anthropic/maxResultSizeChars"] annotation (up to 500K), allowing larger results like DB schemas to pass through without truncation
• Added disableSkillShellExecution setting to disable inline shell execution in skills, custom slash commands, and plugin commands
• Added support for multi-line prompts in claude-cli://open?q= deep links (encoded newlines %0A no longer rejected)
• Plugins can now ship executables under bin/ and invoke them as bare commands from the Bash tool
• Fixed transcript chain breaks on --resume that could lose conversation history when async transcript writes fail silently
• Fixed cmd+delete not deleting to start of line on iTerm2, kitty, WezTerm, Ghostty, and Windows Terminal
• Fixed plan mode in remote sessions losing track of the plan file after a container restart, which caused permission prompts on plan edits and an empty plan-approval modal
• Fixed JSON schema validation for permissions.defaultMode: "auto" in settings.json
• Fixed Windows version cleanup not protecting the active version's rollback copy
• /feedback now explains why it's unavailable instead of disappearing from the slash menu
• Improved /claude-api skill guidance for agent design patterns including tool surface decisions, context management, and caching strategy
• Improved performance: faster stripAnsi on Bun by routing through Bun.stripANSI …

• Added /powerup — interactive lessons teaching Claude Code features with animated demos
• Added CLAUDE_CODE_PLUGIN_KEEP_MARKETPLACE_ON_FAILURE env var to keep the existing marketplace cache when git pull fails, useful in offline environments
• Added .husky to protected directories (acceptEdits mode)
• Fixed an infinite loop where the rate-limit options dialog would repeatedly auto-open after hitting your usage limit, eventually crashing the session
• Fixed --resume causing a full prompt-cache miss on the first request for users with deferred tools, MCP servers, or custom agents (regression since v2.1.69)
• Fixed Edit/Write failing with "File content has changed" when a PostToolUse format-on-save hook rewrites the file between consecutive edits
• Fixed PreToolUse hooks that emit JSON to stdout and exit with code 2 not correctly blocking the tool call
• Fixed collapsed search/read summary badge appearing multiple times in fullscreen scrollback when a CLAUDE.md file auto-loads during a tool call
• Fixed auto mode not respecting explicit user boundaries ("don't push", "wait for X before Y") even when the action would otherwise be allowed
• Fixed click-to-expand hover text being nearly invisible on light terminal themes
• Fixed UI crash when malformed tool input reached the permission dialog
• Fixed headers disappearing when scrolling /model, /config, and other selection screens …

• Added "defer" permission decision to PreToolUse hooks — headless sessions can pause at a tool call and resume with -p --resume to have the hook re-evaluate
• Added CLAUDE_CODE_NO_FLICKER=1 environment variable to opt into flicker-free alt-screen rendering with virtualized scrollback
• Added PermissionDenied hook that fires after auto mode classifier denials — return {retry: true} to tell the model it can retry
• Added named subagents to @ mention typeahead suggestions
• Added MCP_CONNECTION_NONBLOCKING=true for -p mode to skip the MCP connection wait entirely, and bounded --mcp-config server connections at 5s instead of blocking on the slowest server
• Auto mode: denied commands now show a notification and appear in /permissions → Recent tab where you can retry with r
• Fixed Edit(//path/**) and Read(//path/**) allow rules to check the resolved symlink target, not just the requested path
• Fixed voice push-to-talk not activating for some modifier-combo bindings, and voice mode on Windows failing with "WebSocket upgrade rejected with HTTP 101"
• Fixed Edit/Write tools doubling CRLF on Windows and stripping Markdown hard line breaks (two trailing spaces)
• Fixed StructuredOutput schema cache bug causing ~50% failure rate when using multiple schemas
• Fixed memory leak where large JSON inputs were retained as LRU cache keys in long-running sessions
• Fixed a crash when removing a message from very large session files (over 50MB) …

• We've raised the max_tokens cap to 300k on the Message Batches API for Claude Opus 4.6 and Sonnet 4.6. Include the output-300k-2026-03-24 beta header to generate longer single-turn outputs for long-form content, structured data, and large code generation tasks.
• We're retiring the 1M token context window beta for Claude Sonnet 4.5 and Claude Sonnet 4 on April 30, 2026. After that date, the context-1m-2025-08-07 beta header will have no effect on these models, and requests that exceed the standard 200k-token context window will return an error. To continue using 1M context windows, migrate to Claude Sonnet 4.6 or Claude Opus 4.6, which support the full 1M token context window at standard pricing with no beta header required.

• Added X-Claude-Code-Session-Id header to API requests so proxies can aggregate requests by session without parsing the body
• Added .jj and .sl to VCS directory exclusion lists so Grep and file autocomplete don't descend into Jujutsu or Sapling metadata
• Fixed --resume failing with "tool_use ids were found without tool_result blocks" on sessions created before v2.1.85
• Fixed Write/Edit/Read failing on files outside the project root (e.g., ~/.claude/CLAUDE.md) when conditional skills or rules are configured
• Fixed unnecessary config disk writes on every skill invocation that could cause performance issues and config corruption on Windows
• Fixed potential out-of-memory crash when using /feedback on very long sessions with large transcript files
• Fixed --bare mode dropping MCP tools in interactive sessions and silently discarding messages enqueued mid-turn
• Fixed the c shortcut copying only ~20 characters of the OAuth login URL instead of the full URL
• Fixed masked input (e.g., OAuth code paste) leaking the start of the token when wrapping across multiple lines on narrow terminals
• Fixed official marketplace plugin scripts failing with "Permission denied" on macOS/Linux since v2.1.83
• Fixed statusline showing another session's model when running multiple Claude Code instances and using /model in one of them
• Fixed scroll not following new messages after wheel scroll or click-to-select at the bottom of a long conversation …

• Added CLAUDE_CODE_MCP_SERVER_NAME and CLAUDE_CODE_MCP_SERVER_URL environment variables to MCP headersHelper scripts, allowing one helper to serve multiple servers
• Added conditional if field for hooks using permission rule syntax (e.g., Bash(git *)) to filter when they run, reducing process spawning overhead
• Added timestamp markers in transcripts when scheduled tasks (/loop, CronCreate) fire
• Added trailing space after [Image #N] placeholder when pasting images
• Deep link queries (claude-cli://open?q=…) now support up to 5,000 characters, with a "scroll to review" warning for long pre-filled prompts
• MCP OAuth now follows RFC 9728 Protected Resource Metadata discovery to find the authorization server
• Plugins blocked by organization policy (managed-settings.json) can no longer be installed or enabled, and are hidden from marketplace views
• PreToolUse hooks can now satisfy AskUserQuestion by returning updatedInput alongside permissionDecision: "allow", enabling headless integrations that collect answers via their own UI
• tool_parameters in OpenTelemetry tool_result events are now gated behind OTEL_LOG_TOOL_DETAILS=1
• Fixed /compact failing with "context exceeded" when the conversation has grown too large for the compact request itself to fit
• Fixed /plugin enable and /plugin disable failing when a plugin's install location differs from where it's declared in settings …

• Added PowerShell tool for Windows as an opt-in preview. Learn more at https://code.claude.com/docs/en/tools-reference#powershell-tool
• Added ANTHROPIC_DEFAULT_{OPUS,SONNET,HAIKU}_MODEL_SUPPORTS env vars to override effort/thinking capability detection for pinned default models for 3p (Bedrock, Vertex, Foundry), and _MODEL_NAME/_DESCRIPTION to customize the /model picker label
• Added CLAUDE_STREAM_IDLE_TIMEOUT_MS env var to configure the streaming idle watchdog threshold (default 90s)
• Added TaskCreated hook that fires when a task is created via TaskCreate
• Added WorktreeCreate hook support for type: "http" — return the created worktree path via hookSpecificOutput.worktreePath in the response JSON
• Added allowedChannelPlugins managed setting for team/enterprise admins to define a channel plugin allowlist
• Added x-client-request-id header to API requests for debugging timeouts
• Added idle-return prompt that nudges users returning after 75+ minutes to /clear, reducing unnecessary token re-caching on stale sessions
• Deep links (claude-cli://) now open in your preferred terminal instead of whichever terminal happens to be first in the detection list
• Rules and skills paths: frontmatter now accepts a YAML list of globs
• MCP tool descriptions and server instructions are now capped at 2KB to prevent OpenAPI-generated servers from bloating context …

• Added managed-settings.d/ drop-in directory alongside managed-settings.json, letting separate teams deploy independent policy fragments that merge alphabetically
• Added CwdChanged and FileChanged hook events for reactive environment management (e.g., direnv)
• Added sandbox.failIfUnavailable setting to exit with an error when sandbox is enabled but cannot start, instead of running unsandboxed
• Added disableDeepLinkRegistration setting to prevent claude-cli:// protocol handler registration
• Added CLAUDE_CODE_SUBPROCESS_ENV_SCRUB=1 to strip Anthropic and cloud provider credentials from subprocess environments (Bash tool, hooks, MCP stdio servers)
• Added transcript search — press / in transcript mode (Ctrl+O) to search, n/N to step through matches
• Added Ctrl+X Ctrl+E as an alias for opening the external editor (readline-native binding; Ctrl+G still works)
• Pasted images now insert an [Image #N] chip at the cursor so you can reference them positionally in your prompt
• Agents can now declare initialPrompt in frontmatter to auto-submit a first turn
• chat:killAgents and chat:fastMode are now rebindable via ~/.claude/keybindings.json
• Fixed mouse tracking escape sequences leaking to shell prompt after exit
• Fixed Claude Code hanging on exit on macOS
• Fixed screen flashing blank after being idle for a few seconds …

Computer use research preview in Cowork and Claude Code + Dispatch improvements

Users on Pro and Max plans can give Claude access to computer use. Claude can open files, run dev tools, point, click, and navigate to what’s on your screen to perform tasks itself—with no setup required. Claude’s new computer use capability makes Dispatch even more helpful by allowing it to use your computer on your behalf while you’re away. For more information, see Let Claude use your computer in Cowork or our blog post: Put Claude to work on your computer.