claude --version to check your installed version.
Added
- Added support for turning on Code Mode while agent teams are enabled; if you already turned it on, sessions with teams on now start in Code Mode
- Added a startup warning when an
--allowedTools,--disallowedToolsor--toolsvalue has an entry that doesn’t look like a complete rule - Added a message explaining why MCP servers stop loading when an administrator adds
managed-mcp.jsonduring a session - Added a row to
/statusthat says when your provider or an environment variable such asDISABLE_TELEMETRYturns features off - Added a note in
/pluginandclaude plugin listwhen a malformedplugin.jsonfield such asauthororhomepageis ignored - Added Ctrl+O to show hidden startup notices
- Added a hint in the empty prompt that says what to do after a request fails
- Added an
esc to stopfooter hint in a subagent’s view where Esc would stop it - Added
OTEL_RESOURCE_INCLUDE_HOSTto includehost.nameandhost.idin OpenTelemetry resource attributes (off by default) - Added
overflowX="scroll"on a mod’sBox, andmaxLinesandwrap="clip"on itsText - [VSCode] Added a
claudeCode.promptSuggestionssetting that turns prompt suggestions off
- Changed edits in a session’s mods folder to ask more often, and sandboxed commands to write to fewer folders; if every edit asks, remove any link there to an unreadable or network path
- Changed terminal and VS Code sessions in organizations with the Claude Code + Cowork HIPAA configuration to start in auto mode by default;
disableAutoModeturns it off - Changed terminal and VS Code sessions with no mode set to start in auto mode, not Manual mode, when your organization’s policy hasn’t loaded
- Changed a
managed-mcp.jsonadded mid-session so it may apply at the next check; restart open sessions for it to apply in full - Changed Code Mode’s
$shell helper to refuse some values with an error; nothing runs - Changed auto mode to stop some shell commands that ran before
- Changed some Bash commands on macOS that ran before to ask or be refused
- Changed a hook’s
"decision": "block"to hold in more cases, so a hook that always prints one may start blocking calls - Changed sessions that start in auto by default to take a server-managed
defaultModeonce settings load, or Manual if settings fail with no stored copy and policy fails or requires it - Changed auto mode for organizations with the Claude Code + Cowork HIPAA configuration: on a direct Anthropic API connection, the server-side classifier now checks actions, as for other organizations
- Changed the Bash sandbox to keep its state in some cases during a session, until you run
/sandboxor/cd; Bash may ask first - Changed sandboxed commands on macOS, Linux and WSL to no longer write to some sensitive home folders by default; some can be allowed with
sandbox.filesystem.allowWritein user or managed settings - Changed the Bash sandbox to set aside unexpected entries in a repository’s
.gitfolder after a command - Changed the Bash sandbox on Linux and WSL to create some missing
--add-dirfolders before each sandboxed command, so commands that create them can fail, skip, or land inside them - Changed commands that a permission hook or SDK host rewrites while approving to stay in the sandbox if the original was checked as sandboxed; the approver can add
dangerouslyDisableSandbox: true - Changed
claude mcp serveto ignoresandbox.excludedCommands - Changed
/rewindon macOS, Linux and WSL to leave some files untouched and report them as skipped when the sandbox is on - Changed the Bash sandbox to ask before some lines that set environment variables when auto-allow is off; unattended runs can refuse them
- Changed Claude’s file tools to refuse writes to Claude Code’s sign-in files, such as
.credentials.json - Changed the Bash permission check to ask before more commands reach Claude Code’s sign-in files
- Changed some file tools to refuse some device and process files
- Changed some
cpandmvcommands to ask first - Changed some tool calls in bypass permissions mode to ask for approval, or be denied where no one can be asked
- Changed the Bash permission check to ask about some commands with very long values it can’t finish checking, or to refuse them where a deny rule may apply or no one can answer
- Changed subagents isolated in a worktree to refuse, in any repository, some git commands in Bash that can rewrite branches shared with your checkout; sessions are not affected
- Changed
ExitWorktreeto refuse removing a worktree while a background subagent or shell command may still be running in it, unless Claude passesdiscard_changes - Changed the computer use macOS permissions panel to ignore j and k; the arrow keys and Ctrl+P/Ctrl+N still move the selection
- Changed the PowerShell tool to ask before some more commands; unattended runs refuse them unless an allow rule, auto mode or bypass permissions mode allows them
- Changed the PowerShell tool to ask about some more commands, even in bypass permissions mode, and to refuse some of them
- Changed some malformed PowerShell commands to be denied
- Changed some PowerShell tool commands that define a function named like a built-in command to be refused, so Claude can rename it
- Changed auto mode to ignore some wildcard shell allow rules, such as ones that name no program or that start with
$HOME, so the safety check decides those commands - Changed auto mode to judge wildcard allow rules whose program is a Windows backslash path, so those calls can be blocked; exact rules keep their allow; we plan to restore rules that start with
.\ - Changed a managed
autoMode.environmentlist without"$defaults"to replace every built-in entry in Claude Desktop and Agent SDK apps too; add"$defaults"or copy the sensitive-target ones - Changed
--add-dirand/add-dirin local terminal sessions to show the trust prompt for a folder you have not trusted yet;claude --bgrefuses such a folder - Changed self-hosted runners under
--trust-workspace falseto ignore more repository settings; set such values in user or managed settings - Changed how permission answers over Remote Control and on self-hosted runners save “always allow” choices
- Changed enabled plugins that were refused for a malformed
plugin.jsondisplay field: they now load, with their hooks and MCP servers; administrators should check the plugins they deploy - Changed how
allowedHttpHookUrlsandhttpHookAllowedEnvVarsapply to hooks from managed settings - Changed HTTP hooks to ignore
Content-LengthandTransfer-Encodingin theirheaderssetting; the request’s length comes from its body - Changed plugin uninstall to also delete the files extracted from the plugin’s MCP bundles (
.mcpb,.dxt), unless another plugin uses the same bundle - Changed some worktree sessions to skip some plugin settings; a notice names what was skipped
- Changed
/pluginuninstall, remove and switch questions to lists answered with the arrow keys and Enter, or Esc to cancel;y,nanduno longer answer them - Changed automatic compaction in plan mode to use the model answering you, such as Opus with
opusplan, so it can cost more - Changed Cowork cloud sessions to hold an organization’s OpenTelemetry export until the first turn; a session that gets no message now sends nothing, so
claude_code.session.countcan fall - Changed Claude’s messages to a session on another machine to be refused when your session uses an API key, auth token, profile or workload identity beside a saved claude.ai login
- Changed
--baresessions on an unapproved API key: moving one to the background is now refused in most setups, andclaude --bare --bgexits with an error if the background service lacks the key - Changed “Continue automatically at usage limit” to stop with a notice, sending nothing, when your login changes or ends during the wait
- Remote Control: Changed how it connects through a proxy: requests now use a CONNECT tunnel, so a proxy that refuses CONNECT stops Remote Control with an error
- Remote Control: Changed the session debug logs written under
--verboseto go to~/.claude/child-logs/ - Self-hosted runner: Changed Linux and macOS sessions with no
ghto get a built-ingh apiby default, via Anthropic, which adds their GitHub sign-in, if any;--use-anthropic-gh falseturns it off - Changed agent view so that quitting with Ctrl+C twice, Ctrl+D twice or
/exitafter pressing←in an empty conversation ends that session, instead of leaving it running for 5 minutes - Changed
--await-initializeto be refused with--bgor--worktree --tmux, which never read the initialize request, instead of starting without the fields it carries - Changed worktree creation,
/teleportand IDE checkouts to run git with a stricterPATH; a git filter or helper may need its full path - Changed the card for approving a subagent’s plan to two choices, “Yes, and switch to default (ask each time) for this session” and “No”, chosen with the arrow keys and Enter (typed text moves to No)
- Changed the
/claude-apiskill to build agents on Managed Agents by default, unless you want to host the agent loop yourself - Changed
/usagein fullscreen mode to show plan limit bars before session cost for subscribers;/coststill starts with cost - Changed the wording of some messages that
claudesubcommands print; JSON output is unchanged - Changed first-run setup to preselect the “Auto (match terminal)” theme instead of Dark mode
- Changed fullscreen
/cdto refuse “Yes” while its trust prompt doesn’t fit, with a line saying to make the window taller - Removed the tmux notice that said the mouse wheel needs
set -g mouse onin fullscreen; the wheel scrolls without it - Removed the spinner tip that suggested
/clear30 minutes into a running turn - [VSCode] Changed a running command’s live output in the chat to look like the command card’s OUT row
- [VSCode] Changed the text of extension errors in event listeners, views and activation to stay out of the editor’s own error telemetry
Added
- Added a spinner tip in VS Code and similar integrated terminals about re-running a notebook cell, shown once Claude has read or edited a Jupyter notebook
- Added a line under the error shown when your organization’s policy blocks a request, saying how to continue: run
/rewindand change the message, or run/clear - Added a note for Claude when a sandboxed
git checkout,git stashor autostash pull cannot update protected files such as those in.claude - [VSCode] Added prompt suggestions wherever the terminal shows them: Tab accepts the suggested next prompt;
promptSuggestionEnabled: falsein Claude Code’s settings.json turns them off - [VSCode] Added a Console sign-in that creates no API key
- [VSCode] Added a field to paste the redirect URL when signing in to an MCP server from the panel in SSH, WSL and container windows
- Changed sandboxed Bash commands to follow the normal permission flow by default; set
sandbox.autoAllowBashIfSandboxed: trueto keep running them without prompts - Changed managed settings that turn the sandbox on: to keep commands running without prompts, set
sandbox.autoAllowBashIfSandboxed: truein your highest-ranking managed source - Changed read-only git commands to ask first in a repository that isn’t the project’s or one you trusted; name it with
--add-dir, not theadditionalDirectoriessetting, to run them without prompts - Changed scheduled tasks and
/loopto ask before running a skill that declaresallowed-toolsor hooks; addSkill(name)topermissions.allowto run it unattended - Changed “Always allow” in Desktop, IDE and SDK hosts: a rule alone is offered only when saving it would allow that command again, so a scheduled task that relied on one can now wait for you
- Changed the sandbox to refuse sandboxed writes to
.claudesettings, skills, agents, commands and hooks in added directories, as it does in the launch folder - Changed worktree isolation: an isolated worktree session or agent is now refused
git bisect run,git maintenance register|startandscalar’s set-up commands, and is told what still works - Changed plugin MCP bundles (
.mcpb,.dxt) to unpack under the plugins directory, not.mcpb-cachein the plugin’s folder, so read-only plugin folders load; one fetched by URL downloads again once - Changed plugins of a remote marketplace that managed settings declare, enable and list in
prependPluginsorappendPluginsto count as the organization’s;pluginDigestsis deprecated and ignored - Changed plugin loading to skip, and
claude plugin validateto report, some unsafe paths and links in a plugin’s folder - Changed nested skill loading:
.claude/skillsfolders in--plugin-dirplugins no longer load from a session started above them; start Claude Code in the plugin folder instead - Changed EnterWorktree to refuse a
WorktreeCreatehook path that goes through a symlink inside the project; have the hook print the real path - Changed tmux and iTerm2 teammates, and new sessions from the agents view opened with ←, to start in the session’s folder instead of where the shell last ran
cd - Changed prompt suggestions in the terminal to refresh less often after 8 turns in a row started by background or scheduled tasks, saving usage until you press a key or return to the terminal
- Changed Esc on a background MCP task’s question to “answer later”: nothing is sent to the server, and the question stays in
/tasks - Changed
CLAUDE_CODE_MAX_CONTEXT_TOKENSto also set the context window of unrecognized model IDs that start withclaude-, such as a gateway’sclaude-sonnet-latest - Changed
/loginto ask for permission to change your privacy settings - Changed
/cost,$.session.usage(),--max-budget-usdand the cost and token metrics to count a mod’s$.model.completeand$.model.classifycalls - Changed the programs that Claude Code’s sandbox starts for its own use, such as its network bridge, to run without Claude Code’s sign-in tokens and API keys in their environment
- Changed ripgrep, local git commands and the sandbox runtime’s own helper programs to run without the mTLS client key’s passphrase in their environment
- Remote Control: Changed
/resumeand/teleportto disconnect first when the connected session has a share link, or when that can’t be checked; run/remote-controlto connect again - Changed Claude Code on the web, cloud scheduled tasks and Cowork cloud sessions to start despite
sandbox.failIfUnavailable: the hosted container has no command sandbox - Changed requests for settings, permission rules, hooks or status over a cloud or self-hosted session’s connection to return less or be refused; what applies in the session is unchanged
- Changed
claude -pwith no cached account data to wait for that data before its first request, so a first run in a clean environment can start a few seconds later - Changed Claude’s instructions to keep files you’ll want out of its temporary scratchpad folder, which can be cleared
- Changed marketplace
refto accept the full ID of any commit on a branch or tag; clients before this version can’t clone such a pin, so a marketplace pinned this way stops loading there - Changed plugin marketplace and plugin downloads to use a CONNECT tunnel when
ALL_PROXYis the only proxy set; a proxy that refuses CONNECT now fails them - Changed
-pand SDK runs in a worktree or submodule mounted into a container to apply.claude/settings.local.jsonallow rules only once the folder is trusted - Changed self-hosted runners to refuse a session whose session folder or repository path is a symbolic link, other than a
checkouthook’s checkout - Removed “ask your admin to raise it” from spend limit and shared budget errors in cloud sessions and under Remote Control
- [VSCode] Changed the install link’s marketplace confirmation button to “Trust marketplace and continue”
Added
- Added Code Mode, an experimental opt-in in which Claude writes JavaScript to call its tools; see https://code.claude.com/docs/en/code-mode to turn it on, or to turn it off for an organization
- Added
engines["claude-code"]toplugin.json: plugins outside their stated Claude Code version range are not installed, updated or loaded, and install and update pick the newest version that fits - Added type checking of a plugin’s TypeScript hooks module to
claude plugin validate; each finding is a warning, and--strictfails on it - Added to
claude plugin validatethe$nouns a mod’sengine.createhook adds, withholds or replaces - Added support for
CLAUDE_CODE_AUTO_MODE_MODELin managed settings for Claude Desktop sessions on third-party providers (takes effect once Claude Desktop includes this version) - Added an options menu (stop, switch to usage credits, upgrade) when a model needs usage credits that are turned off or used up
- Added a progress heartbeat every 30 seconds while hooks run long in
claude -pand SDK sessions, so a host such as Claude Desktop won’t end the turn as hung (once it includes this version) - Added
.docxsupport to Artifact tool asset uploads - Added a
/diff to closehint to the diff panel, and the keys that focus and close a panel to the?shortcut guide - Added a start-up warning for Bash allow rules with a
*in a variable’s value, such asBash(PORT=* npm run dev) - Added a warning when a plugin’s language server can’t start, naming the server and what to check
- Added a one-time startup notice naming
CLAUDE_CODE_PERFORCE_MODEwhen Perforce settings are found and the mode is off - Added
CLAUDE_CODE_DISABLE_AGENT_PROGRESS_SUMMARIES(available since 2.1.318): set it to1to turn off subagent progress summaries and the requests that write them (anthropics/claude-code#86879) - Added
CLAUDE_CODE_RESUME_HINT_COMMAND, which a program that starts Claude Code can set to have the exit hint name its own resume command - Added
unsetOptionCountto theclaude plugin install --jsonresult: how manyuserConfigoptions are not yet set - Added
isPromptCoveredto the props a mod’s pane is drawn with, telling it when a dialog or a field above the prompt covers the prompt - [VSCode] Added a Claude Code trust prompt before the first session in a folder you haven’t trusted; declining leaves the folder’s settings and MCP servers unused
- [VSCode] Added the live output of a running Bash or PowerShell command to the chat, under the command’s card
- [Claude Code on the web] Added a failure notification when a scheduled task’s run can’t start because its settings require a sandbox that isn’t available where it runs
- [Claude Code on the web] Added the No repository option to the repository picker for new sessions on a self-hosted environment
- [Claude Code on the web] Added a warning to a cloud session’s Delete dialog when deleting it will pause scheduled tasks that run in that session
- [Remote Control] Added a Try again button to the error shown when the computer running your session is signed out of Claude Code
- [Claude Tag] Added Claude Tag sessions started in Slack to the Compliance API on Enterprise plans, in beta: list the sessions and read their transcripts
- [Claude Tag] Added a confirmation step when an admin turns off Claude Tag for the organization, with a dialog that says what stops
- [Claude Tag] Added a Retry button to a Slack channel’s Configure page for when its settings fail to load
- [Claude Tag] Added a Scope filter to the Network events tab of Claude Tag’s Activity page, so admins can download one scope’s events instead of the whole organization’s
- [Code Review] Added a Download CSV button to Code Review admin settings: export review costs per repository by month, week or day, for one repository, one GitHub organization or all
- Self-hosted runner: Reverted the exit code of an on-demand runner whose work order was superseded to 1; 2.1.321 and 2.1.322 had changed it to 0
- Changed
/security-reviewto ask before Claude runs it, and skills’ inline git commands to stop in bare-repository worktrees and similar folders until you allow them, outside auto mode and bypass - Changed how many messages each of your sessions on the same machine can send the others without a prompt from you, from about 20 to 100,000
- Changed the number of messages a Claude Desktop Code tab session can send to your other sessions before it waits for you to type again from 20 to 100,000
- Changed organization settings and policy requests to no longer use a proxy set only by
ALL_PROXY; setHTTPS_PROXYif your network needs a proxy for them - Changed the context window assumed for a model ID that Claude Code doesn’t recognize from 200K to 1M; use
/autocompactto compact earlier, orCLAUDE_CODE_DISABLE_1M_CONTEXT=1to keep 200K - Changed sessions on other vendors’ models with a smaller window: once full, they can fail every turn until
/clear, so setCLAUDE_CODE_DISABLE_1M_CONTEXT=1before they fill - Changed Claude Code’s own ripgrep and most internal git commands to run without most of its sign-in tokens and API keys, so a credential helper or hook under such a git command no longer finds them
- Changed
MCP_SDK_GENERATION=v1to run the v2 MCP client in a compatibility mode (earlier handshake, relaxed OAuth sign-in checks), because the v1 MCP client is removed - Changed the tool Claude uses to finish an MCP server sign-in so that it asks you to confirm, even in bypass and auto modes; where nobody can be asked, such as in
dontAskmode, use/mcp - Self-hosted runner: Changed a session at a connector’s approval prompt or sign-in inside a running tool to be released as paused at the idle or lifetime limit; a late answer resumes from a fresh clone
- Changed artifact updates to ask first, even under an allow rule, when people outside your organization can see the artifact or Claude can’t check who can
- Changed
glab config setwithout--globalor--host, andglab config edit --local, to fail read-only inside the Bash sandbox; run them outside it - Changed the Bash sandbox on macOS, Linux and WSL to ignore write access granted through a symlink in a writable folder that leads elsewhere or to a missing path; grant the real path instead
- Changed “don’t ask again” so an allow rule that looks like it holds a secret, or would be saved to a
settings.local.jsonthat git tracks, lasts for the session only and isn’t saved - Changed the sandbox on macOS and Linux:
git worktree add,moveandremovenow fail read-only, as do commits in a worktree the session wasn’t started in or given, even one inside the project - Changed start-up to stop when a
--settingsfile isn’t valid JSON and has asandboxkey, even one that turns the sandbox off - Changed what happens when a tool’s permission check fails with an error after a
PreToolUsehook returned “allow”: Claude Code now decides the call as if no hook had answered - Changed what happens when a tool’s permission check fails on input that a
PermissionRequesthook rewrote: Claude Code now asks again, or refuses the call where nobody can answer - Changed the GitLab merge request badge to show only for the repository a sandboxed session opened in
- Changed the
claude mcp add --helpexample to a local command - Changed mods: a
*, namespace or negated hook onengine.createnow only observes. To add or withhold a$noun, hookengine.createby name - Changed
claude plugin validateto fail a mod whoseengine.createhook isn’t written in the one accepted form; installed mods keep loading - Changed marketplace and plugin installs to fail when a git submodule isn’t on the commit its repository records
- Changed project (
.mcp.json) MCP server approvals to match the server’s exact name; an approval entry written in another spelling now asks again - Changed self-hosted runners under
--trust-workspace falseand VS Code sessions in an untrusted folder: the repository’s settings can no longer disable your hooks or turn plugins on or off - Changed
tool.listplugin hooks so the tools they add or remove apply to the model and every agent - Changed
claude plugin installand/pluginto show a warning instead of success when a plugin’s install command writes to stderr - Changed claude.ai chats and Cowork tasks to check a plugin file for secret keys before sending it, and to ask first if it may hold one
- Changed
claude rmand the agents view to say a session is removed from the list, not deleted;claude rmhelp says the conversation is kept andclaude --resumelists it - Changed
/stickersto say that the sticker pack is all claimed for now, instead of opening a page in the browser - Changed a managed settings
policyHelperto no longer take environment variables from user or--settingsfiles; set what it needs in managedenvor the launching shell - Changed a
/skillor@filetyped between two files picked in claude.ai or the desktop app, or after a picked file and a link, to no longer take effect - Changed split-pane agent teammates to check that requests come from their team lead: one that outlives its lead has its asks refused, and plan approvals must be re-sent after
/clearor/resume - Changed worktree creation to refuse checkouts whose
.gitfolders are links or hold hard links, such as git-new-workdir layouts andcp -alcopies; use a plain clone - Changed what Claude is told after safeguards stop a forked skill or a workflow agent: retry with a narrower prompt or tell you, instead of doing the work itself
- Changed browser tool calls (Claude in Chrome, the desktop browser pane) to refuse some escaped or JSON-text inputs and tell Claude how to respell them
- Changed workflows so that a prompt an app marks as not typed no longer reaches their agents as your request
- Changed computer use on macOS to refuse key chords that hold Space with Control or Option, and more system shortcuts that need your grant
- Changed first-run setup in Terminal.app to name the “Use Option as Meta key” setting, what Option stops typing, and how to undo it before you agree
- Changed the maximum of
CLAUDE_CODE_OVERLOADED_RETRY_MAX_DELAY_MSfrom 64000 to 300000 - Changed workflows in the desktop app’s Code tab to pause at a usage limit and continue after it resets
- Changed the spend limit and 1M context errors, and
/usage-credits, in cloud sessions and under Remote Control: they no longer give a web address - Changed which MCP servers are reported by name in OpenTelemetry exports to follow the current connector directory
- Changed
/logoutin a background session, which had no effect: it now signs out every session on the computer, deletes stored MCP logins and plugin secrets, and ends the session, as in a terminal - macOS: Changed the Bash sandbox so
lsofand similar tools may show less about processes a command did not start - Changed cloud scheduled tasks: a
PostToolUsehook from your own settings can no longer replace the plan inExitPlanMode’s result - Changed
CLAUDE_AUTOCOMPACT_PCT_OVERRIDEin Anthropic-hosted cloud sessions to stop applying to auto-memory extraction and, in some sessions, to/btw; the main conversation still follows it - Removed the
subagentStatusLinesetting: subagent rows always use the default look, and settings files that still hold the key load as before - Deprecated a plugin’s
agentsetting: it still applies, and now warns at startup, inclaude plugin listand inclaude plugin validate - [VSCode] Changed opening a Claude Code on the web session in a folder where Claude Code’s sandbox is on: the extension no longer switches to the session’s branch for you, and the dialog says why
- Added fast mode on Vertex in the global, us and eu regions, falling back to standard speed if Vertex refuses it
Added
- Added Claude Fable 5.2 (
claude-fable-5-2), now the default Fable model — 1M context, $8/$40 per Mtok with $0.20/Mtok cache reads - Added
--settings <file>toclaude remote-control, which applies a settings file to every session it starts; if the file changes, no new session starts until you restartclaude remote-control - [VSCode] Added a warning above the prompt box when both an auth token and an API key are set, matching the terminal’s startup warning
- Updated the session-feedback message and the
allowedProviderssetting description to name Ultra alongside Pro and Max, and theforceLoginMethoddescription to cover every paid Claude plan - Updated the bundled dataviz skill to suggest a list of meters, one row per item, for several ratios that each have their own limit
- Updated the “Upgrade to Max” options to say “Upgrade to Max or Ultra”
- Changed mods Claude writes to turn on without asking in auto and bypass modes, even with managed
allowUnsandboxedCommands: falseorallowManagedDomainsOnly, and not to load indontAskmode - Changed mods’ own tool calls (
$.tool.call,$.mcp.call) to raise no permission prompt, even under your own ask rules or indontAskand-pruns; deny rules and organization policy still apply - Changed saved Fable 5.1 model picks to move to
fable: on Pro and Max plans at once, on Team and Enterprise plans once your organization allows it; use/model claude-fable-5-1to switch back - Changed the fast mode labels in
/fast,/configand/modelto name your current model when it supports fast mode, and removed “(Opus only)” from the VS Code row - Changed nested skill loading:
.claude/skillsfolders inside~/.claude/pluginsno longer load when Claude works on files there - Self-hosted runner: Changed an on-demand runner that the server refuses because it is not needed (session ended or paused, work order superseded) to exit 0 instead of 1
- Windows: Changed how Claude Code finds programs it starts itself, such as git, to skip relative and UNC PATH entries and links onto a network share, and to pick up PATH changes from settings
- Changed OpenTelemetry hook events in non-interactive sessions to leave two built-in hooks out of
num_hooksandnum_successafter Bash and PowerShell calls - Changed the number of messages a Claude Desktop Code tab session can send to your other sessions before it waits for you to type again from 10 to 20
- Changed
AskUserQuestioninclaude -pand SDK sessions to deny a host’s permission answer that sets the undocumentedafkTimeoutMsinput field - Changed managed
extraKnownMarketplaces: a declared name can’t be added from another source, an entry from elsewhere is hidden, and plugins the declared marketplace lists are reinstalled from it - Changed Claude in Chrome tools that an app running Claude Code provides, such as the Claude desktop app’s Code tab, to follow your organization’s Claude in Chrome policy
- Changed
policy_mode: layeredin the Claude apps gateway: managed MCP servers now add up over every matching policy,match: {}included, so a gateway already on it serves more after the upgrade - Gateway: Changed Bedrock, Vertex and Foundry upstreams to ignore the
ANTHROPIC_*_BASE_URLvariables in the gateway’s environment. Setbase_urlon the upstream to use another endpoint - Changed headless sessions to reload mods in
.claude/skillswhen their files change; setCLAUDE_CODE_PLUGIN_DIR_WATCH=0to turn this off - Changed the Cyber Verification Program sentence on a cyber safeguard block to use the same words as claude.ai
- Changed Cowork cloud sessions: a PostToolUse hook can no longer replace the answer to AskUserQuestion or the plan in ExitPlanMode’s result; Claude reads the originals
- Changed
cpandmvspelled by a path, such as/bin/cp, to get the same path checks ascpandmv - Changed background command notifications and TaskStop results to show Claude only the start and end of a command over 1,000 characters, which saves context
- Changed artifact design guidance to treat Inter and Space Grotesk as good default typefaces
- Changed the account email check to refuse only a shell command that holds your email and names a User-Agent, without asking, so other commands and scheduled runs no longer stop at its prompt
- Changed Remote Control sessions to keep the folder they were started in: a connected client can no longer move a session to another folder
- Changed managed
extraKnownMarketplacestake-overs from an undeclared source: saved data is set aside forcleanupPeriodDays, and options and secrets untilclaude plugin restore-set-aside
Added
- Added support for MCP tasks in the terminal: when an MCP server runs a tool call as a task, Claude Code moves it to the background right away and tells Claude the result when it finishes
- Added the
CLAUDE_CODE_DISABLE_MCP_TASK_BACKGROUNDenvironment variable to make Claude Code wait for an MCP task inside its tool call instead of moving it to the background right away - Added
permissions.denyInsteadOfPromptingsetting: a tool call that would wait on a permission prompt is denied and the turn continues, in terminal sessions, subagents and teammates - Added
/claude-api latency-optimizeto find where your Claude API code makes people wait and try faster settings against your eval - Added the
CLAUDE_CODE_DISABLE_FEATURE_FLAGSenvironment variable, which stops feature-flag fetching without turning off telemetry; every flag then uses its built-in default - Added spend for today, this week and this month to
/usagefor developers behind a Claude apps gateway who have no spend limit, when the gateway runs this version or later - Added
admin.pooled_conflict_ruleandadmin.pooled_priority_orderto the Claude apps gateway, to set which pooled group budget pays first for a developer in several groups - Added an optional
login_noticesetting to the Claude apps gateway that shows an operator-defined message on the gateway’s browser verification page and on Claude Code’s sign-in screen - Added
CLAUDE_GATEWAY_LOG_FORMAT=jsonto the Claude apps gateway: operational log lines become JSON withseverity,timeandmessage, for log collectors that read the level from a field - Added a Claude apps gateway boot warning for a case where another policy’s
availableModelsalso allows a custom model id that one policy lists by name - Added start-up warnings to the Claude apps gateway for policies with no
desktopHomethat rely oncoworkTabEnabled: falseorchatSessionRetentionDays, ahead of Claude Desktop’s November 10 change - Added a warning when your claude.ai login ends and Claude Code continues on the sign-in from
ant auth login, which bills usage to your Claude Console organization - Added a line at startup naming the Claude Console organization that is billed in a case where Claude Code uses the sign-in from
ant auth login - Added an “Earlier in this session” row to
/tasksthat lists finished subagents, so you can reopen one and message it even after it leaves the list or you restart - Added a note to
/loopconfirmations in cloud sessions that a timer stops with its container, with an offer of a loop that survives - Gateway: Added startup warnings for upstreams that mix AWS GovCloud with other upstreams, and for base URL variables in the gateway’s environment
- Gateway: Added a start-up log warning when managed settings name an Amazon Bedrock guardrail in a header, which the gateway doesn’t forward, and no
bedrockupstream sets aguardrail - [Claude Tag] Added a Claude Tag row to the Models admin page for Enterprise orgs with Claude Tag connected, linking to where its model defaults are managed
- Changed auto mode with telemetry or feature flags off, including behind LLM gateways: calls the server can’t check are asked about or denied, with no fallback to the built-in classifier; set
CLAUDE_CODE_AUTO_MODE_SERVER=0to keep using it (Bedrock, Vertex and Foundry are unchanged) - Changed auto mode behind an LLM gateway:
CLAUDE_CODE_AUTO_MODE_SERVER=1turns on the server-side check even withCLAUDE_CODE_DISABLE_EXPERIMENTAL_BETASset; if managed settings set the latter, they must set the1too - Changed MCP tasks to work in the terminal without feature-flag fetching (Bedrock, Vertex, Foundry, LLM gateways,
DISABLE_TELEMETRY); setCLAUDE_CODE_DISABLE_MCP_TASKS=1to turn them off - Changed auto mode so that a project’s settings files or a plugin can no longer switch its server-side check back on once you have switched it off outside the project
- Changed Monitor’s WebSocket watch to follow a rule on every Bash command: a deny refuses it, and under an ask rule an allow on Monitor, bypass mode or auto mode no longer skips the prompt
- Changed
CLAUDE_CODE_PROXY_RESOLVES_HOSTSso that settings files can no longer set it where the host manages the network; set it in the process environment - Changed a session’s exact Bash or PowerShell ask rule to be refused, not asked about, when a broader deny rule covers the command
- Changed the Bash tool to ask for permission before running
whichwith flags outside a lookup/display set - Changed
--add-dirdirectories to register plugin marketplaces only once that directory’s own trust dialog has been accepted - Changed nested skill loading: with the Bash sandbox on, skills in
.claude/skillsfolders below the starting folder load only if you start Claude Code there or/cdthere - Changed self-paced
/loopto end when a turn finishes without rescheduling, instead of scheduling a fallback wakeup; setCLAUDE_CODE_LOOP_KEEPALIVE=1to keep the old behavior - Changed PostToolUse hooks to run, with up to 10 seconds to finish, on tool calls that finish after a new message interrupts the turn
- Changed what Claude is told after safeguards stop a subagent: retry with a narrower prompt or tell you, instead of doing the work itself
- Changed cloud sessions to drop session-scoped allow rules at a move to a new worker when their deny and ask rules can’t all be kept, so you may be asked again
- Changed the automatic certificate repair behind a TLS-inspecting proxy: it no longer runs when
ALL_PROXYis the only proxy variable set. SetHTTPS_PROXYto the proxy’s http:// URL instead - Changed artifact database saves of rows others may have written: each file gets a hidden marker file, and existing files without one, earlier versions’ exports among them, are left alone and reported
- Updated
/claude-api managed-agents-onboardtemplates to match the Claude Console: removed Performance tuner, renamed two, and turned off web search and fetch in five whose prompts don’t use them - Removed
shortcuts_listandshortcuts_execute, which listed and started shortcuts saved in the extension, from Claude Code’s built-in Claude in Chrome tools - Self-hosted runner: Changed
CLAUDE_RUNNER_CHECKOUT_PATHso that each repository in a session gets its own folder: a folder name already taken in the session gets-2,-3, … at the end - Self-hosted runner: Removed the end-of-session upload of a session’s debug log; a failed session’s log stays on the runner host unless
--remove-session-stateis set - [VSCode] Changed links in Claude’s replies that point at Claude sign-in, connector or API addresses to show as text with the address, instead of opening in the browser
- [Code Review] Changed unlinking a GitHub organization to also turn off its All repositories setting in Code Review; it stays off after you relink until you turn it back on
Added
- Added a record of your answer when Claude Code asks to turn on the mods it writes in a session
- Added
assume_roleonmantleupstreams of the Claude apps gateway, for a role in another AWS account and for per-developer AWS cost attribution - Added an
effortoption to a mod’s$.model.fork, so a quick side job can run atlowwhatever effort the session uses; the result reports the effort sent - Added the date a stored claude.ai sign-in ends to
claude auth status(loginExpiresAt) and toclaude -pJSON results (credential_expires_at) - Added a one-line stderr warning to
claude -ptext output, starting 10 days before a stored claude.ai sign-in ends - Gateway: Added a log warning when the Google access token from a federation (
external_account) credentials file is about to expire - Self-hosted runner: Added
CLAUDE_RUNNER_SPAWN_REASONto the orchestrator’s spawn-runner hook environment: why Anthropic sent the request (e.g.runner_lost,resumed); unset when none was sent - Self-hosted runner: Added a startup warning when
--release-idle-session-minis the last argument with no value (still ignored) or has a value of only spaces (still read as0) - [Cloud sessions] Added a No repository option: a cloud session on an Anthropic-hosted environment can now start without picking a repository
- [Cloud sessions] Added the usual Chrome names to cloud sessions, so tools other than Playwright can find the built-in Chromium through google-chrome and CHROME_PATH
- [Cloud sessions] Added a New session button to the card shown when a cloud session’s environment was deleted, where the message box is hidden
- Changed Claude Code to no longer install or update the IDE extension when started in an IDE terminal; use
/idefrom an outside terminal or your editor’s extensions view - Changed subagents to stop instead of switching models when “Switch models when a message is flagged” is set to “Ask each time”
- Changed background subagents to move a shell command to the background after 3 minutes on all providers, so the prompt cache doesn’t expire while it runs
- Changed subagent progress summaries on Bedrock, Vertex, Foundry, or with telemetry off: on a cold prompt cache, one can now wait up to 5 minutes for the subagent’s pending response
- Changed what a mod attaches at
prompt.submitto stay in the user’s turn rather than go out as a system message - Changed
deniedMcpServersURL entries to also block Anthropic-provided servers whose upstream URL matches, and connectors in cloud sessions and on self-hosted runners, where the session is given the connector’s own URL; otherwise deny a connector by name - Changed the
/permissionsrule details, built-in rules and delete confirmation views to use the standard dialog; pressing Ctrl+C twice there now goes back or cancels instead of quitting - Changed
/claude-api managed-agents-onboardto give a new agent’s built-in tools, and a quickstart template’s MCP tools, theautopermission policy (“Judge risky actions”) by default, matching the Claude Console; say “run without asking” to opt out - Removed the 2.1.288 fix that kept the model’s earlier thinking when resuming a conversation started on 2.1.286 or earlier; that thinking can now be dropped
- Self-hosted runner: Changed the orchestrator’s log lines:
ok session=…now prints at every log level on stdout (was debug only, stderr) withreason=…afterjti=…, andnack session=…ends its fields withreason=…afterretryable=…; update scripts that matchretryable=true:orretryable=false: - [Cloud sessions] Changed connector search results to tell Claude that the search covers only part of the connector directory
Added
- Added a warning to
/schedulewhen a new cloud scheduled task gets connectors you did not ask for, with how to remove them - Added a check that asks before a web or shell tool call sends your account email address to the network
- Added
onFailure: "block"for Agent SDK hook callbacks, so a callback that fails blocks the action instead of letting it run - Added a notice in Claude Desktop’s Code tab when your organization’s remote managed settings haven’t loaded for the session, and another once they do
- Added a notice when auto mode’s check has no answer for a subagent’s tool call, which used to leave nothing in the main view
- Added
anthropic.plan,anthropic.seat_type,anthropic.providerandanthropic.custom_base_urlto OTel metrics, events and spans - Added a cost report to the Claude apps gateway admin API,
GET /v1/organizations/analytics/cost, for reading past spend over a date range in total, per developer, per day or per IdP group - Added a daily record of each developer’s IdP groups on Claude apps gateways with
admin:set, kept foradmin.identity_retention_days, which the cost report groups by - Added
CLAUDE_CODE_DISABLE_NESTED_SESSION_ID: set it to1on a self-hosted runner to haveclaude -prun inside a session use that session’s ID and transcript, as in earlier versions - [Cloud sessions] Added a link card to cloud sessions, so you can open or decline a link an MCP server asks you to visit instead of the request going unseen
- Changed SendMessage so a message to a running background agent arrives earlier: its running command moves to the background, without being stopped
- Changed subagents and teammates on a different model from the main conversation to use that model’s saved effort level or its default, instead of the session’s effort level
- Changed exit to skip the “Background work is running” prompt for plugin monitors
- Changed the built-in
plugin-authoringskill to offer once, after a mod you built works, to help share the mod - Removed the Read tool’s “file unchanged since your last Read” stub: reading an unchanged file again returns its content
- [Cloud sessions] Changed the organization environment dialog in admin settings to the personal dialogs’ short layout: each section opens from a row, and unsaved input asks before closing
- [Claude Tag] Changed newly pasted artifact links in Slack to show a simpler card with an Open button that opens the artifact in Claude, instead of Slack’s side panel
Added
- Added
CLAUDE_CODE_DISABLE_ULTRATHINK, which turns theultrathinkkeyword off - Added pooled group budgets to the Claude apps gateway: one monthly budget that a group’s members share, on top of each member’s own limit, set with
pooled_budgeton the spend limits admin API - Added
CLAUDE_CODE_DISABLE_LOOPED_CROSS_CERT_DROPto keep system cross-certificates that certify each other in a loop, for servers whose private CA needs one - Added
CLAUDE_CODE_SUBAGENT_EFFORTto set the effort level subagents run at when nothing else names one, instead of inheriting the session’s - Added
managed.policy_mode: layeredto the Claude apps gateway: every policy that matches a developer applies, each merged over the ones listed below it, not only the first - Added
guardrail.input_taggingon Claude apps gateway Bedrock upstreams, which wraps prompts in the input tags that an Amazon Bedrock guardrail’s prompt attacks filter needs;all_messages_and_systemalso tags the system prompt - [VSCode] Added sign-in to a Claude apps gateway from the chat panel, with no terminal needed
- [Cloud sessions] Added a form card to cloud sessions, so you can fill in or decline a form an MCP server asks for instead of the request being canceled without you seeing it
- [Cloud sessions] Added a “Catching up…” line that shows while a cloud or Remote Control session you switch back to is still bringing its messages up to date
- [Code Review] Added a findings-by-severity card to the Code Review analytics page, showing the period’s total findings, a daily chart stacked by severity, and each severity’s count
- Deprecated fast mode for Opus 4.8 as of October 30, 2026; it will be removed on November 30, 2026. After removal, fast mode requests on Opus 4.8 return an error and do not fall back to standard Opus 4.8. Migrate to Opus 5.5 to keep the speedup
- Changed auto mode so that an
autoMode.allowlist without"$defaults"no longer removes the built-in rule on scheduled-task prompts - Changed server-managed settings for Team and Enterprise logins: settings fetched from Anthropic’s API now apply without an approval prompt, and a stored copy that doesn’t verify is refused
- [VSCode] Changed
ultrathinkto run its turn atxhigheffort and show in a rainbow in the chat box, as in the terminal - [Cloud sessions] Changed network secrets so Enterprise members can add and delete them on their own cloud environments, instead of only viewing them
- [Claude Tag] Changed a Slack channel’s Respond automatically switch to a “How Claude participates” choice on the channel’s Configure page and in Claude Tag admin settings
Added
- Added
permissions.blockWritesOutsideWorkingDirectories: Write, Edit and NotebookEdit refuse paths outside the working directories in every permission mode - Added an optional
groupslist to Claude apps gateway upstreams: only developers in the listed IdP groups are sent to that upstream, on failover too - Added
guardrail_traceto Claude apps gateway Bedrock upstreams, to log which Amazon Bedrock guardrail filter fired on a request, without the matched text - Added idle compaction to VS Code and Claude Desktop’s Code tab: a 200K+ token conversation left idle is summarized, and the summary is used on your next message if the prompt cache has expired
- [Cloud sessions] Added an Include completed button to the Scheduled tasks page when finished one-off tasks are the only ones hidden, so you can show them without opening the Sort menu
- [Claude Tag] Added a dismissible link to the Claude Tag API docs on the Bundles and Channels tabs in admin settings, for Enterprise organizations with Slack connected
- Updated the
/claude-apiskill: its Datadog setup for API tracing now sends thedd-llmobs-enabled: trueheader, as the Console does, and itsantinstall instructions drop the outdated macOSxattrstep - Updated the bundled dataviz skill to advise printing y-axis tick labels in full under 10,000, and horizontal ones in full under a million where they fit
- Updated the bundled dataviz skill to answer donut and pie chart requests with ranked bars or a segmented bar that prints its values
- Changed Claude’s file tools to ask before editing the local copy of your organization’s server-managed settings, as they do for your settings files
- Changed
/ultrareviewto check workspace trust before it runs git, asclaude ultrareviewdoes - Changed the Bash tool’s shell to clear any inherited
whichalias or function and to stop passing exportedwhich/unsetenvfunctions to commands, sowhichis the system command or zsh builtin - Changed
/permissionsRecently blocked to refuse approving a call you can’t read in full, including one the terminal is too short to show - Changed auto mode in Cowork cloud sessions: a hook’s allow of an MCP tool call now counts as no answer, so the classifier or a permission prompt decides
- Changed hook
ifconditions such asMonitor(tail *)to match the Monitor tool’s shell commands asBash(tail *)matches Bash’s - Changed the
claude-apiskill to ask to read a Managed Agents workflow run’s workflow before changing the agent to improve that run - Changed flagged requests to retry on another model, not end in a refusal, when
ANTHROPIC_DEFAULT_OPUS_MODEL,availableModelsordeniedModelsleaves only an older Opus model, or none - Changed idle compaction to also run for subscribers with telemetry turned off, and on Bedrock, Vertex, Foundry and LLM gateways once the one-hour prompt cache is on
- Changed
FORCE_HYPERLINK=false(alsonoandoff) to turn hyperlinks off instead of on - macOS: Changed
git grepin most commands that Claude runs to use the C locale, to avoid a reported memory leak in macOS’s regex library on files that aren’t valid UTF-8 - [VSCode] Changed
CLAUDE_CODE_DISABLE_CFC_PROMPTto also stop the extension adding the Claude in Chrome instructions to a session’s messages; the browser tools stay available - [Cloud sessions] Changed cloud sessions from the web, desktop and mobile apps to start even when a setup script exits with an error, instead of failing to start
Added
- Added
oidc.user_id_claimto the Claude apps gateway, to identify people by an id_token claim other thansub, such as Microsoft Entra ID’soid - Added click to send: in fullscreen, clicking the “to send now” hint under a queued message sends it right away, as its key does
- [Cloud sessions] Added an effort level to the model picker when you create or edit a routine that runs in the cloud
- [Cloud sessions] Added a Bash editor with colors, line numbers and a Copy button to the setup script box in cloud environment settings
- Changed Bedrock, Vertex, Foundry and Mantle to use the shorter system prompt for model names Claude Code doesn’t recognize, such as custom deployment names and inference profile ARNs, even when they front Sonnet or Haiku; set
CLAUDE_CODE_SIMPLE_SYSTEM_PROMPT=0to keep the full prompt - Changed automatic model switches after a flagged message to also cover model names Claude Code doesn’t recognize, such as custom deployment names; set
CLAUDE_CODE_DISABLE_REFUSAL_FALLBACK=1to turn them off - Changed usage limit messages to call the 5-hour window the “5-hour limit” instead of the “session limit”
- Changed cloud sessions started from the CLI to get their names and branch names from the server
- Changed prompt and agent hooks to run at most 24 at a time per session, sub-agents included, plus at most 12 that have waited longer than their event’s timeout; a
timeoutunder 1 second now counts as 1 second - Changed a subagent’s
autoCompactWindowto lower the window set byCLAUDE_CODE_AUTO_COMPACT_WINDOW, where it had no effect before; it still never raises that window - Changed OpenTelemetry
mcp_server.nameto report the real name of fixed-URL remote MCP servers declared by plugins from the official Anthropic marketplace, instead ofcustom - Changed Claude Code to stop sending the
claude_code.safety_eventOpenTelemetry event - [VSCode] Changed usage limit resets to show a clock time (“Resets at 3:10 PM”) and to say “5-hour limit”
- Improved the prompt’s effort indicator: a change made mid-task now says it may wait for your next message, and the note clears when it no longer applies
- Improved WebFetch errors in cloud sessions: a host that does not resolve is now explained as possibly outside the allowed network domains, so Claude names the host to allow
- [Cloud sessions] Improved the step shown when Claude adds a repository to a cloud session: opening it now shows a short table of what happened instead of raw JSON
- [Claude Tag] Improved Claude Tag’s retry notice in Slack to name the model in use, or say rate limits were hit, instead of a generic API error
Added
- Added
--hold-idle-fileand--max-hold-mintoself-hosted-runner, to keep an idle session while your host reports processes it left running - Added a tip for the Explanatory output style, shown when you ask Claude for explanations
- [VSCode] Added syntax coloring to code blocks in Claude’s replies, in light, dark and high contrast dark themes
- [VSCode] Added a “Running UserPromptSubmit hooks” spinner status and screen reader announcement for slow UserPromptSubmit hooks
- [Claude Tag] Added a Skills view to Claude Tag analytics showing how often Claude runs each attached skill in Slack channels and which go unused, with Usage links in admin settings
- Changed
ultrathinkin the interactive terminal to run its turn at xhigh effort on Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.1 through Anthropic’s API; typed while Claude is working, it raises the rest of that turn - Changed the
ultrathinkrainbow to appear only where the keyword can raise effort, and not when the word is quoted or part of a file path - Changed Ctrl+D in agent view to match the prompt: on an empty input, pressing it twice exits
- Changed
--output-format jsonandstream-jsonoutput to write DEL, C1 control and Unicode format characters inside strings as\uXXXXescapes; parsed values are unchanged - Changed model names and IDs to drop the
[1m]suffix and its label on models where the suffix has no effect; the suffix is still accepted - Changed startup warnings about permission rules in settings files to show once per user until the rule changes
- Changed
/code-reviewto tell Claude to put the effort level first and use max only when asked - Changed
DISABLE_COMPACTin settings files and in a mod’s$.env.setso they can only turn compaction off, not back on - Changed Claude Code on Bedrock, Mantle, Vertex, and Foundry to load model settings from a signed file on downloads.claude.ai; set CLAUDE_CODE_MODEL_CATALOG=off to turn this off
- Changed sessions that set
CLAUDE_CODE_SYNC_SKILLSwith feature-flag fetching off to reuse a claude.ai skills list up to 10 minutes old that another session on the machine fetched - Changed clicks on scrollbar arrows and bars in fullscreen lists to act when the mouse button is released, so moving off the scrollbar before letting go cancels the click
- Changed
claude purgeto say it deletes data only on this computer, not anything stored by Anthropic - [VSCode] Changed the extension to start Claude Code sessions with
DISABLE_TELEMETRYset while VS Code’s telemetry setting is off, so they send no usage metrics or error reports and fetch no feature flags - [VSCode] Changed the session list side bar to load Web sessions when the Web tab is first opened, not at startup
- [Claude Tag] Changed Claude Tag admin settings to show a lock instead of # for private Slack channels that your own Slack account can see
Added
- Added the Monitor tool and plugin monitors on Bedrock, Vertex, and Foundry, and when telemetry or nonessential traffic is disabled; long foreground sleeps are now blocked there too
- [Cloud sessions] Added Claude Code cloud sessions and Remote Control permissions to Enterprise custom roles, so Owners can control access to each by role
- [Cloud sessions] Added a Choose file button to load a cloud environment’s setup script from a file
- Changed the error for an MCP tool call that times out in a non-interactive session while the server’s elicitation is unanswered: it now tells Claude that no answer came and not to repeat the call
- [Cloud sessions] Changed turning on Cloud sessions or Remote Control in admin settings to ask for confirmation and list the custom roles that have that permission
- [Cloud sessions] Changed the cloud environment dialog to open network access, setup script and environment variables from rows instead of listing every editor at once
Added
- [VSCode] Added a log warning that says what to do when dictation or the extension’s own requests fail certificate verification, such as behind a TLS-inspecting proxy
- Updated the bundled dataviz skill to accept APCA as a text-contrast measure (keeping WCAG 2 ratios where AA conformance is required), add an
apca(text, background)function to its palette validator scripts, and say whether a label inside a colored chart fill should be white or near-black - Updated the /claude-api skill: service account keys and Workload Identity Federation tokens can now manage OpenTelemetry destinations
- Changed
sed -ito get the standard Bash prompt and run as written, not a simulated file-edit preview; like other Bash edits, it is no longer restored by/rewind - Changed
claude ultrareviewto require workspace trust: it asks on a terminal, else refuses; scripts pass--trust-workspace, which takes no value (put a PR number or base branch before it), or setCLAUDE_CODE_DISABLE_ULTRAREVIEW_TRUST_CHECK=1 - Changed the dangerous
rmandrmdirprompt in bypass permissions mode in the Desktop app and VS Code to deny the command at once with a rewrite hint, so unattended sessions keep going - Changed commands the settings exclude from the sandbox to ask before writing to its write-list paths while the sandbox is enabled but cannot run; the unsandboxed fallback is refused there
- Changed worktree creation and reuse so git no longer fetches objects on demand, in ordinary clones and in partial clones with a Bash sandbox configured; such a partial clone fetches them first
- Changed the
/claude-apiskill to build Managed Agents subagent teams onmultiagent_20261001with dynamic workflows on by default, instead ofcoordinator
Added
- Added
/claude-api api-versionto move Claude API code to API version2026-10-21 - Added the
claude_code.safety_eventOpenTelemetry event for each tool call a permission rule, auto mode, a safety check, a hook or the sandbox stopped, prompted or sent back - [VSCode] Added a notice on the microphone button when voice dictation stops by itself (15 seconds without recognized speech, or the 2-minute limit), saying why and that the text so far is kept
- [Code Review] Added
@claude review stop(alsooffandpause) to Code Review, ending an@claude review alwayssubscription on a pull request without starting a new review
- Changed the notice Claude gets when the WebSearch budget is used up to also tell it not to query search engines in a browser or through computer use
- Changed the
axScreenReadersetting andCLAUDE_AX_SCREEN_READERin a settingsenvblock to apply only from user, managed or--settingssettings (since 2.1.305) - Self-hosted runner: Changed checkout hooks to be stopped after 30 minutes by default, failing the session’s setup; if yours needs longer, raise
--checkout-hook-timeout-minor set it to0 - Self-hosted runner: Changed the certificate file built for Anthropic-managed git to filter its trusted certificates as the CLI does; a
[runner:warn]line gives the number left out - [VSCode] Changed the “still open somewhere else” notice to also cover a Claude process left over from an extension restart, which usually ends on its own within about 10 minutes
- [VSCode] Changed voice dictation to filter its trusted certificates as the CLI does
- [VSCode] Changed the extension’s own requests to filter their trusted certificates as the CLI does, and to trust the
NODE_EXTRA_CA_CERTSfile named in the environment or in user settings
- Fixed
claude -pinside cloud sessions reporting “Not logged in” when started without the session’s environment variables - Fixed an API error being replaced by “prompt is too long”, and an unneeded compaction, in sessions with many MCP tools
Added
- Added
sandbox.allowNestedUserNamespacesso sandboxed commands on Linux can create user namespaces (Chromium, rootless podman); this lets them get around the sandbox’s file protections - Added
CLAUDE_CODE_RETRY_WATCHDOG_USAGE_LIMIT_RECHECK_MSso unattended retry mode (CLAUDE_CODE_RETRY_WATCHDOG) notices a usage limit that was lifted before its reset - [VSCode] Added a “Running start-up hooks” spinner status, with a screen reader announcement, while SessionStart or Setup hooks run
- [Claude Tag] Added public Slack channels Claude hasn’t joined to the Claude Tag admin Add place search, so admins can set one up from there
- Changed
allowedMcpServersto stop filteringmanaged-mcp.jsonstdio servers that use${VAR}only inenv: one your allowlist left out now loads on upgrade; usedeniedMcpServersto keep it off - Changed plugins that Claude Desktop mounts by path to no longer keep their
allowed-toolswhen your organization’s managed settings setallowManagedPermissionRulesOnly - Changed the
/artifactsdelete confirmation to take 1 or Enter to keep, 2 to delete, and Esc to cancel;yandnno longer answer it - Changed a mod’s docked pane to stay within 45% of the terminal’s width, including a width kept from a wider terminal
- [VSCode] Changed the message shown when a session does not start in time and no hook is running: it now leads with the output logs and lists causes instead of pointing to sign-in and network first
- [Code Review] Changed Code Review re-reviews so they no longer open with LGTM while earlier blocking findings are still open
Added
- Added a transcript line when a plugin that runs first skips one of your mod’s hooks, naming the event and the plugin
- Added
decider: "person"to a mod’stool.checkanswer: only the permission dialog or a host’s callback can approve the ask, and it is refused with the mod’s reason where nobody can be asked - Added
e.isReadOnlyto a mod’stool.checkevent, set when the tool call is read-only - Added
/auto-mode-setup: a guided setup that reads your project and recent sessions to teach auto mode about your environment, now also available on Bedrock, Vertex, Foundry, Claude apps gateway and telemetry-disabled sessions - Added
/claude-api managed-agents-secure, which compares a Managed Agents agent’s settings with Anthropic’s published security guide, without changing anything - Added the names of the tools called earlier to the compaction summary in the transcript view (Ctrl+O)
- Added a warning on Pro and Max plans when resuming an inactive conversation would use a noticeable share of your 5-hour usage limit: resume anyway, or start a new conversation and Claude can look up the previous one if you refer to it
- [VSCode] Added mod drawing in the chat panel for sessions on Amazon Bedrock, Google Vertex AI and Microsoft Foundry, behind an LLM gateway, or with telemetry turned off
- [Code Review] Added review levels to Code Review: Owners choose Quick, Standard (the default), or Deep for the organization, and can give individual repositories their own level
- [Code Review] Added an option in Code Review admin settings to review every repository in a GitHub organization, including ones added later, without selecting them one by one
- Changed mods’
settings.readhooks to receive each settings layer and rewrite the merged settings, so a managed mod can make an organization’s setting a ceiling, including thecleanupPeriodDaysthe retention sweep uses - Changed the
/artifactsdelete confirmation to show numbered choices: 1 or Enter keeps the artifact, 2 deletes it, and y, n and Esc still work - Changed the question that turns on the mods Claude writes in a session to plain text with no link or session id; it stays up when a new turn starts, and Claude tells you what loaded when you turn them on
- Changed tool search to be on by default behind a custom
ANTHROPIC_BASE_URL; a 400 or 422 from the gateway is retried without it - Changed compaction to send Claude the skill list again afterwards;
/reload-skillsnow sends only new or changed skills - Changed skill hooks: the skills of one source (a repo, your own folder, or a plugin) can register at most 1,024 hooks and 16,384 matcher characters per session
- Changed skills under a policy that allows hooks from plugins only: a skill used in the main conversation whose hooks are kept out no longer gets its
allowed-tools - Updated the /claude-api skill to use the
antCLI on Claude Platform on AWS instead of falling back to the SDK or curl - Removed the notes from
claude plugin validatethat predicted where a mod’s hooks would run - Removed the “(research preview)” label from the
/fastdialog title on Opus 5.5 - [VSCode] Changed background sub-agents to show as one status line under their Agent card instead of streaming every step into the chat; the full transcript stays in /tasks
Added
- Added
CLAUDE_CODE_AUTO_MODE_MODELto choose the model the client-side auto mode classifier uses; it does not change the model for server-side auto mode - Added an
Idle compactionrow to/configfor turning idle compaction off or back on; the notices shown after an idle compaction now say how to turn it off - Added a notice to
/design-syncand/design-loginthat standalone Claude Design (claude.ai/design) closes by December 14, 2026 - [Cloud sessions] Added a banner on a routine’s page that says why its last run didn’t start, such as your organization reaching its limit on concurrent cloud sessions
- Changed
claude -pand Agent SDK sessions with no permission mode to start indefaultmode on third-party providers, with telemetry off or on a fresh install;--permission-mode autokeeps auto - Changed fork subagents (
subagent_type: "fork") to be on by default inclaude -p, the Agent SDK, the desktop app and VS Code, as they already are in the terminal;CLAUDE_CODE_FORK_SUBAGENT=0turns them off, andclaude mcp servenever offers them - Changed the desktop app and VS Code to always run the main conversation’s subagents in the background, as the terminal does, rather than let Claude choose; a subagent’s own Agent calls in
claude -p, the Agent SDK, the desktop app and VS Code now run in the foreground by default - Changed the
← for agentsfooter hint to hide 10 seconds after a session opens; the count of agents waiting on you still shows - Changed where the Claude in Chrome instructions sit in the system prompt so that new conversations can read them from the shared prompt cache
- Changed prompt and agent hooks so that at most 12 of one event run at a time; the rest wait their turn
- Changed the message shown when your organization’s policy turns off
/feedback: it now points to GitHub issues instead of an organization admin - [Claude Tag] Changed Claude Tag admin settings to organize Claude’s access into four tabs: Bundles, Connectors, Skills and plugins, and Channels
- [Claude Tag] Changed the Roles page in organization settings to group the roles created for Claude Tag channel managers in a collapsed section below your organization’s own roles
Added
- Added
CLAUDE_CODE_RETRY_WATCHDOG_MAX_USAGE_LIMIT_WAIT_MSso unattended retry mode (CLAUDE_CODE_RETRY_WATCHDOG) ends a request at a usage limit that resets too far away instead of waiting - Added OpenTelemetry tracing guidance to the
claude-apiskill, including setting up trace destinations with theantCLI - Added the
%motion to vim mode: jump to the matching bracket, alone or with d/c/y - Added
rowsto a mod’sInput: a text field of several lines in the terminal, where Enter submits and Shift+Enter or a backslash before Enter starts a new line - Added
classifier_verdictandpermission_modeto OpenTelemetrytool_decisionevents, so auto mode classifier decisions can be told apart from permission rules - Added
classifier_categoryto OpenTelemetrytool_decisionevents, naming the built-in rule a classifier block matched, whenOTEL_LOG_TOOL_DETAILS=1 - [VSCode] Added LaTeX math rendering in Claude’s responses, including
$$and multi-line\[ … \]blocks and inline$…$and\(…\)math - [VSCode] Added a “Get more usage” link to the usage-limit banner for Pro and Max plans in the VS Code and JetBrains panels
- Changed a mod hook’s
.catchhandler to run only when the hook throws or times out; events raised beneath the hook’s own calls now skip that hook without asking its.catch - Changed a mod’s
$.env.setto resolve with{}once the variable is set or unset, or with{ deny }saying why when the write is not made
- Improved switching to a model with a smaller context window: a conversation too long for the new model is now summarized by the previous model first
- Improved rendering performance while a long-running subagent works: its progress trail now updates incrementally instead of being reprocessed in full on every render
- Improved artifacts: pages Claude writes now paint the band above a sticky header, so scrolled text no longer shows there in the mobile apps
- Improved mods: in VS Code and the desktop and mobile apps, a tree drawn above the question dialog can now hold Markdown and up to 48 rows; the terminal keeps 12 plain rows
Added
- Added
virtualTextto a mod’s prompt decorations: text drawn in the prompt box that never enters the draft or the submitted prompt - Added
cache_creation_5m_tokensandcache_creation_1h_tokensto the OpenTelemetryapi_requestevent, splitting cache writes by lifetime - Added
$.ui.view({ agentId })for mods: shows an agent’s transcript in the main view when you ask, and$.ui.view({})goes back - Added a
dirprop on a mod’sTextfor right-to-left text: on Windows and in VS Code’s terminal its rows read in that direction and sit flush right - Added
/advisoron Claude Platform on AWS; no advisor runs until you pick one with/advisor,advisorModelor--advisor, andCLAUDE_CODE_DISABLE_ADVISOR_TOOL=1turns it off - Added OpenTelemetry GenAI semantic-convention attributes to trace spans, such as
gen_ai.provider.name,gen_ai.conversation.id,gen_ai.tool.nameanderror.type - Added support for rewriting a workflow agent’s
promptandmodelin theagent.spawnmod hook, which could only refuse workflow agents before - Added the
prompt.changemod hook, raised when the prompt box’s text changes without a typed edit: a recalled history entry, a clear, an undo, or a plugin’s fill - [VSCode] Added the
claudeCode.attachSelectionsetting to stop text selected in the editor from being added to your messages - [Claude Tag] Added support for Slack Connect channels: an Owner or Claude Tag admin can let Claude in Slack answer in channels shared with another organization; it’s off by default
- [Claude Tag] Added a private Allow once / Deny card in Slack threads for connector actions that need your approval, so Claude asks instead of refusing; others only see that it’s waiting on you
- [Claude Tag] Added support for Slack voice clips in Claude in Slack: Claude reads Slack’s transcript of a clip when Slack has one
- Changed
CLAUDE_CODE_PROVIDER_MANAGED_BY_HOSTsessions to ignore CA-bundle, tool-proxy, git TLS and git config variables likeSSL_CERT_FILEin settings; under Claude Desktop, only in project settings - Changed exit to force-kill background commands still running 1.5 seconds after being asked to stop; in fullscreen mode they could previously be left running
- Changed Claude Desktop’s Code tab to apply managed settings when
forceLoginGatewayUrlnames an Anthropic-operated gateway that serves the session’s hostname; that gateway must be reachable at start - Changed
claude gatewayto stop at start, naming the key, on adesktop:policy key that Claude Desktop cannot use from a gateway, where it earlier started with a warning - Changed large requests to api.anthropic.com to use gzip compression even when telemetry is disabled; set
CLAUDE_CODE_GZIP_REQUEST_BODIES=0to turn it off - Changed the
claude-apiskill to leave Managed Agents sessions in place unless you ask to archive or delete them - Changed background sessions to resume a carried workflow through the Workflow tool’s approval instead of running it from the handoff file
- Changed local session transcripts to also record each API request’s system prompt, tools and parameters, so a transcript you choose to share can reproduce a request exactly
- Changed the
/sandboxmode options to read “Sandbox commands” instead of “Sandbox BashTool” - Changed
PermissionRequesthooks in Cowork cloud sessions: asetModeupdate in a hook’s allow is now ignored; the rest of the approval still applies - Changed hooks:
updatedInputfrom PreToolUse and PermissionRequest hooks is now validated like Claude’s own input: unknown keys are dropped, and input the tool rejects denies the call - Removed the
--sandboxand--no-sandboxflags fromclaude remote-control, which had no effect; passing either now exits with an error. Setsandbox.enabledin settings to sandbox sessions - [Claude Tag] Changed access bundles an admin attached to a Slack channel to keep applying once the channel becomes a Slack Connect channel, with no need to remove and re-attach them
- Improved WebFetch in Cowork cloud sessions to tell Claude that a page failed moments ago instead of fetching it again
- Improved mod sharing in the built-in
plugin-authoringskill: Claude scans a mod’s files for keys before opening them, and asks before scanning a folder outside the project - Improved the note Claude 3 models get about an @-mentioned PDF, so they less often answer without reading it
- Improved artifact updates: Claude now applies a small change and republishes the artifact in one call, and makes larger edits and republishes within a single response
Added
- Added
CLAUDE_CODE_DISABLE_FILE_MENTIONSenvironment variable: Claude gets the paths of@file mentions to read with its file tools, not the files’ contents - Added idle compaction in the terminal: a 200K+ token conversation left idle for about 54 minutes is compacted before its one-hour prompt cache expires, so your next message doesn’t reprocess it all
- Added
CLAUDE_CODE_IDLE_COMPACT_MIN_TOKENSto change idle compaction’s 200K-token minimum;"idleCompaction": falsein settings.json turns it off. Both need v2.1.290 or later - Added
$.prompt.mediaso aprompt.submithook of a mod installed by managed settings can read the bytes of an image or document attached to the prompt - Added
$.session.identity()for mods installed by managed settings: how the session is signed in, and its organization and email when managedforceLoginOrgUUIDlists that organization - Added
size,weight,monoandtabularto a mod’sTextand named spacing steps (xstoxl) to itsBox;boldis deprecated in favor ofweight - Added
--shutdown-drain-sectoclaude self-hosted-runner orchestrator: on SIGTERM it stops polling and gives the poll in flight and running spawn hooks up to that many seconds to finish - Added the skill’s frontmatter, source, plugin name and file path to what a mod’s
skill.prompthook receives - Added
/effortand/fastin agent view to set what new sessions start with; a bare/model,/effortor/fastthere now opens a picker - Added a “safe mode” line to the agents view peek panel for background sessions started with
--safe-mode - Added a one-time note when plan mode won’t ask before most edits or commands because bypass permissions is available
- Added
--disallow-toolsand adisallowed_toolscase field toclaude plugin evalto keep specific built-in tools, such as Agent, out of eval runs - Added support for turning off the attribution block in Claude Desktop’s Code tab: set
CLAUDE_CODE_ATTRIBUTION_HEADER=0in theenvof a Claude apps gateway’scodepolicy - Added a field to the attribution block in Claude Code sessions behind a Claude apps gateway; remove the block with
CLAUDE_CODE_ATTRIBUTION_HEADER=0 - Added a
queuedoor to thesession.appendmod hook, so a mod can make the transcript’s queue records differ from the prompt as sent - Added a note in
/pluginandclaude plugin validatenaming a mod’s hooks that are skipped where an organization manages Claude Code - Added
/usagewarnings for max effort, workflows, always-on context and large tool results - [Claude Tag] Added the ability for Claude in Slack to delete a canvas it created in a channel when someone asks it to
- Changed plugin hooks and workflow scripts to refuse values they hand back that nest deeper than 28,000 levels, with an error naming the limit (workflow scripts allowed 100,000)
- Changed actions a plugin hook can block to be refused, not let through, when the hook cannot be handed them or cannot hand back its answer; a tool that already ran has its result withheld
- Changed requests to claude.ai’s connector proxy and to Anthropic-hosted MCP servers to no longer follow HTTP redirects; redirects by a connector’s own server are unaffected
- Changed MCP requests in cloud and self-hosted runner sessions to no longer follow HTTP redirects from the session’s MCP proxy
- Changed skill and command loading: one whose file or folder name has hidden, direction-changing or blank-like characters,
<or>is no longer offered until renamed; such aname:is ignored - Changed
ui.rendertrees inclaude plugin test: each mod’s Box or Text root now carriespart: { plugin }naming the mod, so exact-tree assertions must include it - Changed WebFetch page reads to pick their effort as web searches do: the caller’s level on the caller’s own model, otherwise the level settings give the small model
- Changed
claude plugin testto fail a call to a mod’s served$method that has not answered within ten seconds, as a session does - Changed MCP allow rules to reach only the server they name:
mcp__ano longer allows a server nameda__bora_; writemcp__<server>__*with the server’s whole normalized name - Changed agent
tools:lists to take only the named MCP server’s tools; list a server whose name contains__asmcp__<whole name>__* - Changed Claude in Chrome in Cowork cloud sessions outside auto mode: an approved-site action the safety classifier flags now asks you instead of being refused, unless managed settings apply
- Changed cloud session and Remote Control MCP connectors to negotiate MCP protocol version 2026-07-28 by default on installs that fetch no flags;
MCP_PROTOCOL_NEGOTIATION=legacyopts out - Changed MCP tool descriptions and server instructions from Anthropic’s first-party MCP servers at
api.anthropic.comto reach Claude in full instead of being cut at 4,096 characters - Changed auto mode in Cowork cloud sessions: a
PermissionRequesthook can no longer approve shell commands or subagents; those prompts wait or are denied - Changed the resume command printed at exit: it names the session by its ID and no longer by its name, so that nothing in a session name can run when the line is pasted
- Changed the resume command printed at exit of a worktree session: it uses
-wand-rif the long flags do not fit the terminal. If neither fits a kept worktree, rows above it say where to run it - Changed
/btwside questions in non-interactive sessions to return a timeout error when still unanswered after 11 minutes - Changed
UserPromptSubmithooks to run before a skill or custom slash command expands, so a hook that blocks the prompt now also stops its embedded shell commands - Changed a skill’s own front-matter
UserPromptSubmithook to first run on the prompt after the one that invokes the skill - Changed the
turn.startmod hook to also hear subagents’ and teammates’ turns, not only the main loop’s; tell them apart bye.agentId(absent on the main loop) - Changed the errors shown when
/ultrareviewcan’t upload a repository to stop suggestinggit add ., which could stage or commit untracked files such as.env - Changed the cyber safeguard notice to no longer ask people to apply to the Cyber Verification Program
- [Cloud sessions] Changed deleting a self-hosted environment to turn off its routines right away, with the reason shown on each routine, instead of leaving their runs to fail
- [Claude Tag] Changed how Claude in Slack corrects a message it got wrong: it edits the message in place and adds a short “Edited” note instead of striking through the old text
- [Claude Tag] Changed Claude in Slack to join channels with more than 100 members quietly, with no intro message