Anthropic Release Notes

• Added --bare flag for scripted -p calls — skips hooks, LSP, plugin sync, and skill directory walks; requires ANTHROPIC_API_KEY or an apiKeyHelper via --settings (OAuth and keychain auth disabled); auto-memory fully disabled
• Added --channels permission relay — channel servers that declare the permission capability can forward tool approval prompts to your phone
• Fixed multiple concurrent Claude Code sessions requiring repeated re-authentication when one session refreshes its OAuth token
• Fixed voice mode silently swallowing retry failures and showing a misleading "check your network" message instead of the actual error
• Fixed voice mode audio not recovering when the server silently drops the WebSocket connection
• Fixed CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS not suppressing the structured-outputs beta header, causing 400 errors on proxy gateways forwarding to Vertex/Bedrock
• Fixed --channels bypass for Team/Enterprise orgs with no other managed settings configured
• Fixed a crash on Node.js 18
• Fixed unnecessary permission prompts for Bash commands containing dashes in strings
• Fixed plugin hooks blocking prompt submission when the plugin directory is deleted mid-session
• Fixed a race condition where background agent task output could hang indefinitely when the task completed between polling intervals
• Resuming a session that was in a worktree now switches back to that worktree …

• Added rate_limits field to statusline scripts for displaying Claude.ai rate limit usage (5-hour and 7-day windows with used_percentage and resets_at)
• Added source: 'settings' plugin marketplace source — declare plugin entries inline in settings.json
• Added CLI tool usage detection to plugin tips, in addition to file pattern matching
• Added effort frontmatter support for skills and slash commands to override the model effort level when invoked
• Added --channels (research preview) — allow MCP servers to push messages into your session
• Fixed --resume dropping parallel tool results — sessions with parallel tool calls now restore all tool_use/tool_result pairs instead of showing [Tool result missing] placeholders
• Fixed voice mode WebSocket failures caused by Cloudflare bot detection on non-browser TLS fingerprints
• Fixed 400 errors when using fine-grained tool streaming through API proxies, Bedrock, or Vertex
• Fixed /remote-control appearing for gateway and third-party provider deployments where it cannot function
• Fixed /sandbox tab switching not responding to Tab or arrow keys
• Improved responsiveness of @ file autocomplete in large git repositories
• Improved /effort to show what auto currently resolves to, matching the status bar indicator
• Improved /permissions — Tab and arrow keys now switch tabs from within a list
• Improved background tasks panel — left arrow now closes from the list view …

• Added --console flag to claude auth login for Anthropic Console (API billing) authentication
• Added "Show turn duration" toggle to the /config menu
• Fixed claude -p hanging when spawned as a subprocess without explicit stdin (e.g. Python subprocess.run)
• Fixed Ctrl+C not working in -p (print) mode
• Fixed /btw returning the main agent's output instead of answering the side question when triggered during streaming
• Fixed voice mode not activating correctly on startup when voiceEnabled: true is set
• Fixed left/right arrow tab navigation in /permissions
• Fixed CLAUDE_CODE_DISABLE_TERMINAL_TITLE not preventing terminal title from being set on startup
• Fixed custom status line showing nothing when workspace trust is blocking it
• Fixed enterprise users being unable to retry on rate limit (429) errors
• Fixed SessionEnd hooks not firing when using interactive /resume to switch sessions
• Improved startup memory usage by ~18MB across all scenarios
• Improved non-streaming API fallback with a 2-minute per-attempt timeout, preventing sessions from hanging indefinitely
• CLAUDE_CODE_PLUGIN_SEED_DIR now supports multiple seed directories separated by the platform path delimiter (: on Unix, ; on Windows)
• [VSCode] Added /remote-control — bridge your session to claude.ai/code to continue from a browser or phone
• [VSCode] Session tabs now get AI-generated titles based on your first message …

• Added StopFailure hook event that fires when the turn ends due to an API error (rate limit, auth failure, etc.)
• Added ${CLAUDE_PLUGIN_DATA} variable for plugin persistent state that survives plugin updates; /plugin uninstall prompts before deleting it
• Added effort, maxTurns, and disallowedTools frontmatter support for plugin-shipped agents
• Terminal notifications (iTerm2/Kitty/Ghostty popups, progress bar) now reach the outer terminal when running inside tmux with set -g allow-passthrough on
• Response text now streams line-by-line as it's generated
• Fixed git log HEAD failing with "ambiguous argument" inside sandboxed Bash on Linux, and stub files polluting git status in the working directory
• Fixed cc log and --resume silently truncating conversation history on large sessions (>5 MB) that used subagents
• Fixed infinite loop when API errors triggered stop hooks that re-fed blocking errors to the model
• Fixed deny: ["mcp__servername"] permission rules not removing MCP server tools before sending to the model, allowing it to see and attempt blocked tools
• Fixed sandbox.filesystem.allowWrite not working with absolute paths (previously required // prefix)
• Fixed /sandbox Dependencies tab showing Linux prerequisites on macOS instead of macOS-specific info
• Security: Fixed silent sandbox disable when sandbox.enabled: true is set but dependencies are missing — now shows a visible startup warning …

• Increased default maximum output token limits for Claude Opus 4.6 to 64k tokens, and the upper bound for Opus 4.6 and Sonnet 4.6 models to 128k tokens
• Added allowRead sandbox filesystem setting to re-allow read access within denyRead regions
• /copy now accepts an optional index: /copy N copies the Nth-latest assistant response
• Fixed "Always Allow" on compound bash commands (e.g. cd src && npm test) saving a single rule for the full string instead of per-subcommand, leading to dead rules and repeated permission prompts
• Fixed auto-updater starting overlapping binary downloads when the slash-command overlay repeatedly opened and closed, accumulating tens of gigabytes of memory
• Fixed --resume silently truncating recent conversation history due to a race between memory-extraction writes and the main transcript
• Fixed PreToolUse hooks returning "allow" bypassing deny permission rules, including enterprise managed settings
• Fixed Write tool silently converting line endings when overwriting CRLF files or creating files in CRLF directories
• Fixed memory growth in long-running sessions from progress messages surviving compaction
• Fixed cost and token usage not being tracked when the API falls back to non-streaming mode
• Fixed CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS not stripping beta tool-schema fields, causing proxy gateways to reject requests …

Control Cowork from your phone with a persistent thread (research preview for Pro/Max)

Users on Pro and Max plans can access a persistent agent thread via Claude Desktop or Claude for iOS/Android to manage tasks in Cowork. We're rolling this out to Max plans starting today, and Pro plans over the next two days. For more information, see Assign tasks to Claude from anywhere in Cowork.

https://www.youtube.com/embed/fVIV-L49eBs?rel=0

• Added MCP elicitation support — MCP servers can now request structured input mid-task via an interactive dialog (form fields or browser URL)
• Added new Elicitation and ElicitationResult hooks to intercept and override responses before they're sent back
• Added -n / --name <name> CLI flag to set a display name for the session at startup
• Added worktree.sparsePaths setting for claude --worktree in large monorepos to check out only the directories you need via git sparse-checkout
• Added PostCompact hook that fires after compaction completes
• Added /effort slash command to set model effort level
• Added session quality survey — enterprise admins can configure the sample rate via the feedbackSurveyRate setting
• Fixed deferred tools (loaded via ToolSearch) losing their input schemas after conversation compaction, causing array and number parameters to be rejected with type errors
• Fixed slash commands showing "Unknown skill"
• Fixed plan mode asking for re-approval after the plan was already accepted
• Fixed voice mode swallowing keypresses while a permission dialog or plan editor was open
• Fixed /voice not working on Windows when installed via npm
• Fixed spurious "Context limit reached" when invoking a skill with model: frontmatter on a 1M-context session
• Fixed "adaptive thinking is not supported on this model" error when using non-standard model strings
• Fixed Bash(cmd:*) permission rules not matching when a quoted argument contains # …

• Added 1M context window for Opus 4.6 by default for Max, Team, and Enterprise plans (previously required extra usage)
• Added /color command for all users to set a prompt-bar color for your session
• Added session name display on the prompt bar when using /rename
• Added last-modified timestamps to memory files, helping Claude reason about which memories are fresh vs. stale
• Added hook source display (settings/plugin/skill) in permission prompts when a hook requires confirmation
• Fixed voice mode not activating correctly on fresh installs without toggling /voice twice
• Fixed the Claude Code header not updating the displayed model name after switching models with /model or Option+P
• Fixed session crash when an attachment message computation returns undefined values
• Fixed Bash tool mangling ! in piped commands (e.g., jq 'select(.x != .y)' now works correctly)
• Fixed managed-disabled plugins showing up in the /plugin Installed tab — plugins force-disabled by your organization are now hidden
• Fixed token estimation over-counting for thinking and tool_use blocks, preventing premature context compaction
• Fixed corrupted marketplace config path handling
• Fixed /resume losing session names after resuming a forked or continued session
• Fixed Esc not closing the /status dialog after visiting the Config tab
• Fixed input handling when accepting or rejecting a plan …

• Added actionable suggestions to /context command — identifies context-heavy tools, memory bloat, and capacity warnings with specific optimization tips
• Added autoMemoryDirectory setting to configure a custom directory for auto-memory storage
• Fixed memory leak where streaming API response buffers were not released when the generator was terminated early, causing unbounded RSS growth on the Node.js/npm code path
• Fixed managed policy ask rules being bypassed by user allow rules or skill allowed-tools
• Fixed full model IDs (e.g., claude-opus-4-5) being silently ignored in agent frontmatter model: field and --agents JSON config — agents now accept the same model values as --model
• Fixed MCP OAuth authentication hanging when the callback port is already in use
• Fixed MCP OAuth refresh never prompting for re-auth after the refresh token expires, for OAuth servers that return errors with HTTP 200 (e.g. Slack)
• Fixed voice mode silently failing on the macOS native binary for users whose terminal had never been granted microphone permission — the binary now includes the audio-input entitlement so macOS prompts correctly
• Fixed SessionEnd hooks being killed after 1.5 s on exit regardless of hook.timeout — now configurable via CLAUDE_CODE_SESSIONEND_HOOKS_TIMEOUT_MS
• Fixed /plugin install failing inside the REPL for marketplace plugins with local sources …

• Added modelOverrides setting to map model picker entries to custom provider model IDs (e.g. Bedrock inference profile ARNs)
• Added actionable guidance when OAuth login or connectivity checks fail due to SSL certificate errors (corporate proxies, NODE_EXTRA_CA_CERTS)
• Fixed freezes and 100% CPU loops triggered by permission prompts for complex bash commands
• Fixed a deadlock that could freeze Claude Code when many skill files changed at once (e.g. during git pull in a repo with a large .claude/skills/ directory)
• Fixed Bash tool output being lost when running multiple Claude Code sessions in the same project directory
• Fixed subagents with model: opus/sonnet/haiku being silently downgraded to older model versions on Bedrock, Vertex, and Microsoft Foundry
• Fixed background bash processes spawned by subagents not being cleaned up when the agent exits
• Fixed /resume showing the current session in the picker
• Fixed /ide crashing with onInstall is not defined when auto-installing the extension
• Fixed /loop not being available on Bedrock/Vertex/Foundry and when telemetry was disabled
• Fixed SessionStart hooks firing twice when resuming a session via --resume or --continue
• Fixed JSON-output hooks injecting no-op system-reminder messages into the model's context on every turn
• Fixed voice mode session corruption when a slow connection overlaps a new recording
• Fixed Linux sandbox failing to start with "ripgrep (rg) not found" on native builds…

Work across Excel and PowerPoint

We’ve improved our Claude for Excel and Claude for PowerPoint add-ins. They can now share the full context of your conversation, so every action Claude takes in one application is informed by everything that’s happened in the other. We also added support for skills in the add-ins, and the ability for Amazon Bedrock, Google Cloud’s Vertex AI, or Microsoft Foundry users to connect to them via an LLM gateway. For more information, refer to the following articles:

• Work across Excel and PowerPoint

• Use Claude for Excel and PowerPoint with an LLM gateway

• Fixed tool search to activate even with ANTHROPIC_BASE_URL as long as ENABLE_TOOL_SEARCH is set.
• Added w key in /copy to write the focused selection directly to a file, bypassing the clipboard (useful over SSH)
• Added optional description argument to /plan (e.g., /plan fix the auth bug) that enters plan mode and immediately starts
• Added ExitWorktree tool to leave an EnterWorktree session
• Added CLAUDE_CODE_DISABLE_CRON environment variable to immediately stop scheduled cron jobs mid-session
• Added lsof, pgrep, tput, ss, fd, and fdfind to the bash auto-approval allowlist, reducing permission prompts for common read-only operations
• Restored the model parameter on the Agent tool for per-invocation model overrides
• Simplified effort levels to low/medium/high (removed max) with new symbols (○ ◐ ●) and a brief notification instead of a persistent icon. Use /effort auto to reset to default
• Improved /config — Escape now cancels changes, Enter saves and closes, Space toggles settings
• Improved up-arrow history to show current session's messages first when running multiple concurrent sessions
• Improved voice input transcription accuracy for repo names and common dev terms (regex, OAuth, JSON)
• Improved bash command parsing by switching to a native module — faster initialization and no memory leak
• Reduced bundle size by ~510 KB …

• Added /loop command to run a prompt or slash command on a recurring interval (e.g. /loop 5m check the deploy)
• Added cron scheduling tools for recurring prompts within a session
• Added voice:pushToTalk keybinding to make the voice activation key rebindable in keybindings.json (default: space) — modifier+letter combos like meta+k have zero typing interference
• Added fmt, comm, cmp, numfmt, expr, test, printf, getconf, seq, tsort, and pr to the bash auto-approval allowlist
• Fixed stdin freeze in long-running sessions where keystrokes stop being processed but the process stays alive
• Fixed a 5–8 second startup freeze for users with voice mode enabled, caused by CoreAudio initialization blocking the main thread after system wake
• Fixed startup UI freeze when many claude.ai proxy connectors refresh an expired OAuth token simultaneously
• Fixed forked conversations (/fork) sharing the same plan file, which caused plan edits in one fork to overwrite the other
• Fixed the Read tool putting oversized images into context when image processing failed, breaking subsequent turns in long image-heavy sessions
• Fixed false-positive permission prompts for compound bash commands containing heredoc commit messages
• Fixed plugin installations being lost when running multiple Claude Code instances
• Fixed claude.ai connectors failing to reconnect after OAuth token refresh …

• Fixed API 400 errors when using ANTHROPIC_BASE_URL with a third-party gateway — tool search now correctly detects proxy endpoints and disables tool_reference blocks
• Fixed API Error: 400 This model does not support the effort parameter when using custom Bedrock inference profiles or other model identifiers not matching standard Claude naming patterns
• Fixed empty model responses immediately after ToolSearch — the server renders tool schemas with system-prompt-style tags at the prompt tail, which could confuse models into stopping early
• Fixed prompt-cache bust when an MCP server with instructions connects after the first turn
• Fixed Enter inserting a newline instead of submitting when typing over a slow SSH connection
• Fixed clipboard corrupting non-ASCII text (CJK, emoji) on Windows/WSL by using PowerShell Set-Clipboard
• Fixed extra VS Code windows opening at startup on Windows when running from the VS Code integrated terminal
• Fixed voice mode failing on Windows native binary with "native audio module could not be loaded"
• Fixed push-to-talk not activating on session start when voiceEnabled: true was set in settings
• Fixed markdown links containing #NNN references incorrectly pointing to the current repository instead of the linked URL
• Fixed repeated "Model updated to Opus 4.6" notification when a project's .claude/settings.json has a legacy Opus model string pinned
• Fixed plugins showing as inaccurately installed in /plugin …

• Added the /claude-api skill for building applications with the Claude API and Anthropic SDK
• Added Ctrl+U on an empty bash prompt (!) to exit bash mode, matching escape and backspace
• Added numeric keypad support for selecting options in Claude's interview questions (previously only the number row above QWERTY worked)
• Added optional name argument to /remote-control and claude remote-control (/remote-control My Project or --name "My Project") to set a custom session title visible in claude.ai/code
• Added Voice STT support for 10 new languages (20 total) — Russian, Polish, Turkish, Dutch, Ukrainian, Greek, Czech, Danish, Swedish, Norwegian
• Added effort level display (e.g., "with low effort") to the logo and spinner, making it easier to see which effort setting is active
• Added agent name display in terminal title when using claude --agent
• Added sandbox.enableWeakerNetworkIsolation setting (macOS only) to allow Go programs like gh, gcloud, and terraform to verify TLS certificates when using a custom MITM proxy with httpProxyPort
• Added includeGitInstructions setting (and CLAUDE_CODE_DISABLE_GIT_INSTRUCTIONS env var) to remove built-in commit and PR workflow instructions from Claude's system prompt
• Added /reload-plugins command to activate pending plugin changes without restarting
• Added a one-time startup prompt suggesting Claude Code Desktop on macOS and Windows (max 3 showings, dismissible) …