Claude Code's bundled default model is now Opus 5.5 at medium effort, replacing Opus 5 at high effort. You can set CLAUDE_CODE_SKIP_PLUGIN_LOCAL_MCP_SERVERS to keep only the remote MCP servers that plugins bring. A program that runs claude mcp serve can pick which plugins load with CLAUDE_CODE_MCP_SERVE_ENABLED_PLUGINS. A new allowNestedUserNamespaces sandbox setting lets browsers or rootless podman run inside the sandbox on Linux. It is off by default because it weakens the sandbox. Plugin modules can now read and override the settings in effect through a new settings.read event.
Some work in this build is not switched on yet. In-app surveys for enterprise organizations stay off unless turned on remotely. Setting CLAUDE_CODE_DISABLE_FEEDBACK_SURVEY skips them. Groundwork for showing skills as plugins has landed but cannot be reached. Leaving a worktree now checks whether to keep it, but that check always says no in this build. A launch option to turn off plan or auto mode is partly built, with no visible setting yet.
Creating a worktree no longer resets an existing branch with the same name. Claude Code now stops with an error instead. Your cached remote settings are no longer cleared when a proxy answers with a 204 or 404 reply. Very long timeouts are now capped instead of firing immediately. Removing one skill's hooks no longer removes matching hooks added by another skill. Claude Code no longer fetches skills from MCP servers when it connects to them.
Written by our agent from the shipped bundle, not by Anthropic.
New environment variables add low disk and memory warnings, limit plugin MCP servers, and pass startup and project details to child sessions
ExtensionsA new settings.read event lets plugin hooks see and override your effective settings, re-checked whenever settings or plugins change
ExtensionsA new carry_across_restarts option lets post-turn memory extraction save and restore its progress when its worker restarts
Sessions & agentsprompt.fill and suggest-prompt requests now carry answersPress, plugins get press_drawn and press_fill hook events, plus two new defaults
Extensions
CLAUDE_CODE_ATTRIBUTION_PROJECT setting checked against claude.ai Project IDsClaude Code now recognises CLAUDE_CODE_ATTRIBUTION_PROJECT and drops a value that is not a valid claude.ai Project ID from saved jobs
Want the reasoning? Read walks the 43 entries that probably matter to you, each one opening to what changed and why.
Read this release → Every row →13 more of these are in What probably matters to you, on page 1.
The Monitor tool's start message now says when it may stop and that a late event is not the user's reply
Sandbox block lists are now built in one shared place, and credential paths stay in their own list instead of joining denyRead
In some sessions Claude Code now reports that the sandbox is enabled but unavailable in this session
Unclear Which sessions this condition applies to is not known.
When the sandbox proxy blocks a request, the reply can now use a status other than 403 and a different error header
Unclear Which blocked requests now get a status other than 403 or a different header is not stated.
Scheduled wakeups and queued messages can now carry a plugin name, and messages from people are taken from the queue first
Missed and due scheduled tasks can now be kept back or removed by whatever handles them, instead of being deleted as soon as they are reported
Session scheduled tasks now store the plugin that created them, and on resume a task with an invalid plugin name is skipped
The run skill's Playwright MCP helper can now pass --preview to its script, and new live-page files ship with it
Unclear What turns the preview option on is not stated.
When skill files change on disk, Claude Code no longer compares the skill list first and passes the change on every time
On resume, Claude Code no longer restores its record of which skills it already listed to Claude, and replayed skills now note whether they came from an MCP server
Unclear It is not clear whether the record of already-listed skills is rebuilt somewhere else, so the list may or may not be sent again after a resume.
Several startup and cleanup jobs now stop early in a certain mode, and --branches is no longer on a list of allowed git options
Unclear Which mode makes these jobs stop early is not stated.
Claude Code's startup fetch now records a hash of the login token and whether its address moved, and is hardened in one mode
Unclear It is not stated which mode turns off proxies and redirects for this fetch.
A startup check under CLAUDE_CODE_RESTRICTED gained an extra condition that cannot currently be true, so it behaves as before
Claude Code can now track several limit-reset offers with resets left for each, and caches that status for claude.ai accounts for 3 days
Claude Code now contains text for picking a usage-limit reset, and the helper consent dialog accepts two fewer options
Unclear It is not clear whether anything in Claude Code shows this reset text yet.
Claude Code's usage-limit interface now keeps state for an offer about unused resets, usable when not paused and resets remain
Unclear Whether any of this is shown to users yet is not stated.
When a workflow agent stalls and is retried, Claude Code now reports the restarts, the outcome and how much work was lost
After a workflow is saved, Claude Code no longer resets its record of which skills it has already listed to the model
When a workflow fails to compile because of one particular kind of error, Claude Code now reports a separate reason instead of a general syntax error
When a diskless session has no storage for teammate messages, Claude Code now logs that nothing was read or written and lists the likely causes
Claude Code can now say "Teammates are not available in this session." and records a new link for running tools remotely
Unclear It is not clear when the teammates message appears or what the remote tool link does for the user.
Auto mode can add a Chrome-MCP rule for Claude in Chrome tools when CLAUDE_CODE_AUTO_MODE_REMOTE_BROWSER is set or a remote setting allows it
Claude Code now includes the session's mode in the details sent with side requests and auto mode checks
The file upload path check is no longer tied to Chrome only, and invalid browser_batch action names are cleaned before an error shows them
When remote browser support is set, tools from a remote browser server are now grouped with web tools
When fetching a cached configuration fails with a temporary error, Claude Code now marks its saved copy as fresh again
A check tied to the message about uploads being off under CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC now gives a yes-or-no answer from a saved reply
Unclear It is not clear whether file uploads are now blocked under different conditions than before.
A permission decision whose decider is not one of a fixed set of values is now rejected instead of accepted as any text
Unclear Which decider values are allowed is not known.
A permission answer that arrives late is refused if the input it supplies is too large or too deeply nested
In plan mode, Claude is now told when the plan goes to the tool for leaving plan mode, and the plan file is sometimes not read
Unclear When the plan file is skipped, and whether that is on by default, is not stated.
The full plan mode reminder is now built with the list of tools available in the session
When Claude Code fetches Grove settings, an answer whose grove_enabled value is not true, false or empty is now treated as a failure
When fetching the Grove notice settings fails, Claude Code no longer keeps the failure and reports an error instead
When a session process exits, the app running Claude Code can now replace the default suggestion to try again with its own text
Claude Code adds several small internal fields and counters around cloud sessions, the SDK and the plugin cache
Rate-limit messages sent to VS Code extension sessions can now include a usage credit offer, such as a percentage off or extra credit
Buying a custom amount of prepaid credit now sends the expected credit and price along with the request
When Claude Code removes old agent worktrees, it now passes an option about worktrees kept inside another folder
When reading git files for a linked worktree, Claude Code now checks the opened file is the one it inspected and reads nothing if not
Three pieces of text asking upfront for consent to let auto-mode cloud sessions run commands without asking each time were removed
Unclear It is not clear whether the upfront consent step was removed or moved somewhere else.
A setting that used to be included when attaching to a remote session is switched off at build time in this release
Unclear It is not known what the removed option did, so it is unclear whether attaching to a remote session behaves any differently.
The error message about clean-up failing after a sandboxed git run has been removed
Published verbatim by Anthropic for v2.1.304. Text is unmodified from the upstream changelog. Everything else on this page came out of the bundle instead, which is why the two lists don't match.
Of these 76 bullets, 13 name something an entry on this page also names, 24 name something no entry here does, and 39 name nothing specific enough to line up either way. The pairings are made on names both sides wrote down, a flag or a setting or a slash command, so read one as probably the same thing rather than as a fact, and read the middle number as candidates rather than as a miss count.
decider: "person" to a mod's tool.check answer: 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
No entry names this e.isReadOnly to a mod's tool.check event, set when the tool call is read-only
No entry names this /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
No entry names this /claude-api managed-agents-secure, which compares a Managed Agents agent's settings with Anthropic's published security guide, without changing anything
Probably claude-api-managed-agents-secure-guard-on-ant-commands ANTHROPIC_API_KEY at startup or switching "Use custom API key" in /config
Probably policy-limits-re-fetch-after-an-api-refusal-anthropic-api-k, auto-compact-window-dialog-shows-an-extra-warning-line sandbox.filesystem.denyRead glob contains brackets that are not a character class, such as a [user-id] route folder
No entry names this rm -r of a mounted drive or a folder at its top level (/Volumes, /media, WSL's /mnt/c) not getting the safety prompt that rm -rf /Applications gets
Probably dangerous-rm-detection-for-mounted-drive-paths CLAUDE_CODE_TRANSCRIPT_LOCAL_GC dropping messages from the saved transcript when the disk filled up while the file was being trimmed
No entry names this --max-turns, Stop, or ended input by spawning subagents or re-running a finished one
No entry names this key were redrawn unchanged
Nothing to match on plugin-authoring skill stopping at the secret scan on machines with managed settings and on Team and Enterprise plans
No entry names this ../../…; they now show ~/… or the full path
No entry names this context_management: Extra inputs are not permitted errors behind LLM gateways; Claude Code now resends the request without that field
No entry names this /privacy-settings offering choices that may not apply to you after a network or server error
No entry names this claude --help text for --tools, --resume, --debug, --json-schema, --disable-slash-commands, --system-prompt-snapshot and claude update to match actual behavior
Probably usage-rate-limit-message-shows-hold-time, tools-help-text-default-now-means-the-default-set, json-schema-help-says-it-only-works-with-print, sandbox-runtime-probes-bwrap-style-binary-for-disable-user, misc-ui-custom-api-key-prompt-disable-slash-commands-tex, claude-update-description-mentions-package-manager-installs, await-claim-now-rejected-with-sdk-url, resume-help-text-now-says-session-id-or-name, system-prompt-snapshot-now-defaults-off-under-bare-outside claude --worktree removing the worktree and its branch without asking at exit when HEAD was detached or on another branch, which could lose commits
No entry names this /effort status, /effort ultracode and the prompt's effort indicator showing a guessed effort level when CLAUDE_CODE_EFFORT_LEVEL is auto or unset and no level is sent
No entry names this /loop wakeups scheduled at the 1-hour maximum firing just after the 1-hour prompt cache expired, so those wakeups no longer re-read the conversation uncached
Probably dynamic-loop-wakeups-carry-an-owner-and-plugin, schedulewakeup-and-loop-plugin-aware-scheduling-and-missed once hook also removing another skill's identical hook after it ran
Nothing to match on { and } landing the prompt on an inconsistent row, or off screen, instead of the transcript's first row
Nothing to match on CLAUDE_CODE_AUTO_COMPACT_WINDOW=200k silently setting a 100K window: the variable now accepts k and M suffixes like --autocompact, and warns when a value is out of range or ignored
Probably auto-compact-window-setting-shows-a-note-about-the-env-var, claude-code-auto-compact-window-parsing-with-km-shorthand-a /usage to say when a rate limit on usage data ends instead of "try again in a moment"
Probably usage-rate-limit-message-shows-hold-time, usage-credit-purchase-flow-gains-offers-discount-tiers-and, usage-credit-promo-offer-starry-waterfall-read alwaysLoad: true
No entry names this claude -p invalid-settings warning: a file that did not load lists only what stopped it, and long lists are capped per file
Probably claude-test-plugin-checks-agents-and-renders-a-pane settings.read hooks to receive each settings layer and rewrite the merged settings, so a managed mod can make an organization's setting a ceiling, including the cleanupPeriodDays the retention sweep uses
Probably remote-settings-unknowable-pause-for-retention-cleanup-disk, hooks-module-settingsread-merge-is-re-raised-when-plugins-o, new-settingsread-plugin-event-mods-runtime, queue-and-plugin-input-hold-refactor-with-a-plugin-settings /artifacts delete confirmation to show numbered choices: 1 or Enter keeps the artifact, 2 deletes it, and y, n and Esc still work
No entry names this ANTHROPIC_BASE_URL; a 400 or 422 from the gateway is retried without it
Probably tool-search-refusal-by-a-custom-anthropic-base-url-endpoint /reload-skills now sends only new or changed skills
No entry names this allowed-tools
No entry names this ant CLI on Claude Platform on AWS instead of falling back to the SDK or curl
Probably claude-api-managed-agents-secure-guard-on-ant-commands claude plugin validate that predicted where a mod's hooks would run
No entry names this /fast dialog title on Opus 5.5
Probably fast-mode-label-drops-research-preview-conditionally claudeCode.claudeProcessWrapper is set
No entry names this @Claude !fork typed outside a thread getting no visible reply; Claude now shows you a private note saying to run it from inside a thread
No entry names this A model matched these bullets to the GitHub issues they fix, so a link can be wrong.
4 of 29 tool descriptions changed. 2 of 27 tool schemas changed. The appended system-reminder blocks moved: 14 lines added, 1 line removed.
Claude Code, interactive mode
One prompt change in this release could not be quoted from the build, so no entry on this page describes it.
622 documentation changes were recorded within 24 hours either side of this release, nearest first. The closest 12 are below. They're here because they happened near this release in time. That's not a claim that this release caused the edit, or that the page documents anything in it.
The 47 literal strings found in the bundle, with the number of entries that name each one. Picking one searches for it. A name is here because this build's code mentions it, which is not the same as it working or being finished.
What's wrong with this entry?