You can now use the Fable 5.2 and Mythos 5.2 models in Claude Code with the short names fable-5-2 and mythos-5-2. Fable 5.2 is now listed as the latest model, and fable52 works as an alias too. Plugins can use a new tool.check hook to make tool permissions stricter, but never looser for built-in tools. Plugins can also answer or rewrite outgoing web requests, and they are never lent your sign-in. Setting CLAUDE_CODE_ULTRAREVIEW_PR_AS_MERGED asks /ultrareview to review a pull request as it was merged. This release has 418 entries, and 270 of them are usable today.
Some features are in the build but are not switched on yet. A remote switch can make /privacy-settings open your privacy settings on claude.ai in your browser. Video attachments can carry a preview image, but only in hosted remote environments with a switch on. A skill's compatibility note can tell Claude when the skill needs your own computer or network, behind a switch that is off by default. Code editors may show a recommended extension, also behind a remote switch.
Stalled secure connections now stop with a clear error instead of hanging. The usual limit is 120 seconds, and the error says to check your network or proxy. The bridge command now loads the settings file you pass with --settings. Plugin sync now keeps plugins you already installed when your account's plugin list fails to load. A skills folder created after startup is now picked up and watched. The email address privacy prompt for shell and web actions no longer appears in the list of features.
Written by our agent from the shipped bundle, not by Anthropic.
Claude Code recognises the Fable 5.2 and Mythos 5.2 models, needs an explicit opt-in for fast mode on them, and keeps your model when fast mode supports it
Sessions & agentsclaude remote-control --settings <file> passes a settings file to spawned sessions, which refuse to start if that file or their start folder changed
A fable52 model alias is added, and the model picker gains a collapsible 'More models' row with Fable 5.2 highlighted
Claude Code can now check and merge a list of blocked MCP tools, called mcpDeniedTools, into its settings
ExtensionsPlugins can now hook outgoing web requests to answer or rewrite them, but are never lent your sign-in
ExtensionsWant the reasoning? Read walks the 54 entries that probably matter to you, each one opening to what changed and why.
Read this release → Every row →18 more of these are in What probably matters to you, on page 1.
The forceLoginMethod setting's description now says claudeai means any Claude subscription, including Team and Enterprise plans
Failed MCP servers are no longer retried when their setup cannot be read or a policy blocked them
(*) or () are matched differently#Rules written as a tool name followed by (*) or () are read as covering the whole tool, without the earlier allow-only check
Unclear It is unclear whether deny and ask rules written with (*) or () are now also treated as covering the whole tool.
In auto mode, a subagent is not started if the safety classifier cannot check its prompt, and Claude Code says why
Claude Code now detects NFS transcript folders, tolerates unreadable transcript lines, and limits retries on a full disk
A marketplace named in managed settings now removes a same-named install from another source, and admins can word the policy error
Unclear Where administrators set the custom lines for the unreadable-policy error is not stated.
Background agent summaries and auto prompt suggestions are now discarded if the model tries to call a tool instead of writing text
Skill listings gain an addsToEarlier flag that changes the reminder wording and lets skills be re-sent after a reset
Claude Code's list of shell commands that write or remove files, which held rm, rmdir and tee, now also includes cp and mv
Unclear It is not stated what this list controls, so how cp and mv commands are now treated differently is unknown.
When installing a plugin, a name beginning with a dash is now refused, as the repo and owner already were
Plugins from a marketplace named in managed settings are taken back, unless installed_plugins.json cannot be read, which stops it with an error
When Claude Code installs plugins without its interactive screen and no marketplaces are declared, it now runs an extra step instead of only logging
Unclear What the extra step does to plugin setup is not known.
When oversized tool output is cut short, Claude is now told if other large results in the same reply caused part of the shortening
Claude Code now handles a question timing out as a permission decision taken from config, plus a new sign-in-ended case and config-lock tracking
Unclear It is not clear what makes a question time out, so it is unclear when a timeout leads to an accept decision.
Claude Code now marks a shell test command with text after its closing ] as one it cannot check before it runs
Unclear It is not confirmed whether these commands now always ask for permission or how the note is shown to you.
When a plugin stops naming a hooks module, Claude Code reloads it and cancels any notification still pending for that plugin
An MCP server blocked by your organization's connector settings now says so and that an admin can change it on claude.ai
Claude Code now treats the git settings format.outputdirectory and format.suffix as dangerous, like core.worktree and core.bare
A stop request for an unknown shell-task-style ID now fails and asks for the agent's task ID instead of matching an agent by name
In remote sessions, rewinding no longer drops notices about background agents that are still running, when a switch allows it
Unclear The switch that decides whether the notices are kept, and its default, are not known.
Claude Code now tells the model why an artifact write was refused, not to retry it, and to publish under project/ instead
When text meant for the CLAUDE_JOB_DIR folder is over the length limit, Claude Code now skips writing it instead of saving a cut-off copy
Claude Code now refuses to attach files whose names contain a line break or other control character, and upload errors are reworded
Selection lists can show custom markers such as numbers, and Claude Code has a new inline "Enter y/n:" confirmation
/model says it is waiting when a cloud session is not ready yet#Switching models in a cloud session now shows a waiting message until the session is ready, then switches
After an outage, a remote session now tries one more time to renew its credentials when the host lacks a cloud sessions credential
Claude Code now checks where cp and mv really write, looks inside ${HOME} in rm targets, and lets deny rules override rm checks
Remote Control's settings check now counts every non-warning settings issue, including those that were only shown as status notes
Unclear It is not stated what Remote Control does when this check finds a problem, such as refusing to start.
When a skill cannot be found, Claude Code now tells Claude why, such as an older copy, a blocked skill or an unreadable file
A remote-control runner now exits with success when its session is no longer active, so it is not counted as a failure
On Windows, PowerShell arguments that could become network or device paths after character conversion now require your approval
When a transcript shared with feedback is too large, Claude Code now drops parts in up to eight steps instead of four
When an .xlsx upload to an artifact fails, the error now says extra stored copies block sharing and asks you to delete them
In remote sessions, behind a flag, a plain interrupt can cancel messages the server resent; --cloud now reports what was cancelled
Claude Code removes control, formatting and invisible characters from names and titles of files returned by MCP servers
Sending to a stopped helper agent now says it was stopped by the user only when it was, and simply "was stopped" otherwise
The warning shown when both a login token and an API key are set now reports the source of each
Unclear Whether the warning now appears in more or fewer situations than before is not stated.
Claude Code's stop message now says actions were refused and no work went ahead in between, so Claude stopped
A message for a subagent that was stopped in the meantime is now dropped with an error, and MCP tools wait longer before moving to the background
On Windows, startup keeps a Git for Windows launcher value in the global config, and MCP tools that publish artifacts are marked as doing so
Unclear What the stored Git for Windows launcher value is used for is not stated.
MCP OAuth sign-in now runs two checks before the consent step, and reports a cancellation if one fails after you cancel
Unclear What the two new checks look at, and so when they stop a sign-in, is not known.
Claude Code keeps the original text of MCP tool inputs so what it sends matches what was written
Runner help now says the process exits 0 when the runner is not needed and 1 for anything else
The notice about a model being retired now says Amazon Bedrock and Vertex AI set their own schedules, instead of naming Google Cloud
A server reply of HTTP 507 (Insufficient Storage) is now treated as worth retrying
The server errors Claude Code tries again now include 507, alongside 502, 503, 504, 520 and 524
When a connector sign-in is blocked, Claude Code now shows fixed text naming the connector and your organization's connector settings
The installer no longer shows set-up messages on their own screen; they still appear on the success screen
When usage cannot be shown, the message now starts "Plan usage isn't shown in this chat."
The shimmer on the Launching indicator now uses the normal text colour instead of a faint one
claude auth login as a command#The message telling you to sign in now formats claude auth login as a command
On Windows, Claude Code now saves the result of checking a git launcher and reuses it while the launcher file is unchanged
Inserting a notebook cell without an id now reports "a new cell" instead of a blank id
Multi-part file edits now fail with a clear error when the text to replace appears more than once, like single edits do
The download timeout message now explains that Claude Code allows one file at most that long, instead of suggesting a retry
Crash details for background sessions now drop repetitive worktree prefixes, add the exit code and mark sessions that will retry
The prompt footer now shows its tags, such as the fast mode tag, on the same row as the banner text when the window is wide enough
When a tool's output is too large and cannot be saved, Claude is now told only the first part was kept and how to fetch the rest
If the small model that reads a fetched page returns nothing, WebFetch now raises an error saying so and that retrying may work
If an MCP tool result arrives without its text, the error now warns that the tool may have run and tells Claude to check before rerunning
Claude Code now recognises cp and mv in shell commands regardless of letter case or a .exe ending, as it already did for tee
Unclear Which checks rely on this matching for cp and mv is not settled.
Global and project settings saved as Claude Code exits now take a lock, and a save that would lose data is dropped if another copy holds it
Unclear Whether this locking only runs when a remote switch is on is not settled.
When an HTTP hook's address is a literal private IP, the error now says it is not a public address
Disabling or uninstalling a plugin now cancels its pending notifications, even if shutting the plugin down fails
1 more of these is in What probably matters to you, on page 1.
Opening a removed job in the fleet view now reports that the job was removed instead of a failed restart
If your account's plugin list fails or comes back incomplete during plugin sync, Claude Code now keeps plugins it had already installed this way
Unclear It is not clear which installed plugins this covers or what switches this sync path on.
If /cd cannot move the conversation record, Claude Code goes back to the old directory, or finishes the move leaving the record where it was
On Windows, a PowerShell found on PATH that is only a Microsoft Store alias is resolved to the real program or skipped
Claude Code now keeps checking for a skills folder that was deleted or missing, and starts watching it again when it appears
In remote sessions, an approval for a tool whose prompt is already open under a different request is now dropped instead of acted on
If a secure-connection error hits after a reply has started arriving, Claude Code now treats it the same as a reply that was cut short
Unclear Whether a reply cut off this way is now retried automatically is not known.
Prompts that ask a question in the terminal were broken and now work, including when input is piped in rather than typed
The error shown when deleting a scheduled loop fails now names the correct loop, and deleting a loop also updates its messages
Claude Code's check for whether a file was received now answers no on errors, symlinks or very large folders instead of yes
Unclear What a yes or no from this check changes for a reader is not clear.
Claude Code now falls back to a plain patch for a rejected change in one more case, and its error message names the right item
In remote sessions, Claude Code now waits for a file to be fully stored before reading it for a connector upload, and logs less about failures
If fetching the official MCP server list fails partway, Claude Code now keeps the addresses already collected instead of dropping them
If writing a file fails because the path is a symlink loop, Claude Code now removes the path and retries the write
An upload target with an invalid expiresAt date is now treated as having no expiry, instead of being rejected
9 more of these are in What probably matters to you, on page 1.
Behind a switch that is off unless turned on remotely, Claude Code can restore a session's task list at start even when nothing requested it
Claude Code can add an important reminder to Claude when you paste content, behind a switch that is off unless turned on remotely
Two new remote switches, off by default, decide whether certain tools are shown to Claude straight away instead of being held back
Behind a switch that is off by default, some files you mention with @ are no longer added to the list of files Claude is told about
Unclear It is not known which files are affected.
Claude Code now lists seven more server-controlled experiment switches, with no visible effect described yet
Behind a server-controlled switch, Claude Code can now add extra text worked out from the conversation's messages to a piece of its output
Unclear Which piece of text gets the extra content is not known.
Claude Code adds two remote switches to how it assembles messages and attachments; one also needs CLAUDE_CODE_REMOTE
Claude Code has a new remote switch that defaults to off, and what it controls is not known
Unclear What this switch controls is not known.
3 more of these are in What probably matters to you, on page 1.
The system prompt can now include new guidance sections on finishing work, memory and notes, and user privacy, and its research limit is no longer always empty
Unclear Which models or settings turn these sections on is not known.
A new code check marks a script as breaking the prompt cache when it contains an object with a breakPromptCache property
Unclear It is not clear which tool runs this check or whether its result is ever shown to users.
Claude Code adds text naming the model as Claude Fable 5.2, plus instructions on not stopping early and on what to save as memories
Unclear Whether these instructions are always sent or depend on a remote setting is not known.
Claude Code now rejects a kit's results unless they carry an advice field holding a list of text entries
Unclear Which kit this check applies to is not stated.
Submitting a non-empty prompt now records that a deferred privacy checkup notice was shown and dismisses another notice
Unclear Whether any reader is shown a privacy checkup notice, and what it says, is not clear.
One analysis-only background request now gets the session's MCP server connections as well as its tools
Unclear It is not clear which background request this is or what it decides for the reader.
When signed in directly with Anthropic, the billing details sent with requests can now say how a job was started
When Claude's reasoning is cut off in an SDK session, Claude Code can send the partial reasoning in the next request, behind a switch that is off by default
The cache Claude Code uses to reuse earlier message preparation now has an extra auditedRows setting beside its existing verify check
Saved state for background jobs can now hold an optional field recording which entry point launched the job
When a claude -p run exits with an error because it got no result, Claude Code now records a category and reason for the exit
Five AWS and Google credential environment variables were added to three lists of cloud provider settings
Unclear What the three lists control, for example limits in managed settings or removal before running other programs, is not clear.
Published verbatim by Anthropic for v2.1.321. 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 118 bullets, 23 name something an entry on this page also names, 30 name something no entry here does, and 65 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.
claude-fable-5-2), now the default Fable model — 1M context, $8/$40 per Mtok with $0.20/Mtok cache reads
Probably fast-mode-now-needs-an-explicit-opt-in-on-fable-52-mythos, fable-52-and-mythos-52-model-recognition-added --settings <file> to claude remote-control, which applies a settings file to every session it starts; if the file changes, no new session starts until you restart claude remote-control
Probably cloud-sessions-refuse-an-automode-block-instead-of-dropping, remote-control-refuses-settings-before-the-verb-with-a-hi, remote-control-settings-forwarded-to-child-sessions-dire, remote-control-child-sessions-refuse-a-settings-file-that-ch, bridge-fast-path-now-honours-settings, remote-control-cli-adds-settings-and-debug-file-rejects, remote-control-gets-a-settings-flag-for-spawned-sessions cp or mv into a directory writes
Nothing to match on claude mcp login printing a SEP-2352 issuer warning each time you sign in again to a server that Claude Code registered with
No entry names this #108158[BUG] MCP OAuth: SEP-2352 "no 'issuer' stamp" warning on every claude mcp login — provider doesn't round-trip issuer Open
NotebookEdit deleting or overwriting a notebook's first cell when the call named no cell, on network paths (UNC shares, WSL folders, /net mounts)
No entry names this old_string, or editing a notebook as text, on network paths (UNC shares, WSL folders, /net mounts)
No entry names this /config showing Manual as the default permission mode when new sessions start in auto mode; pressing Space on "Auto (default)" now saves Don't Ask, not Plan
No entry names this FileChanged hooks stopping for paths set through watchPaths after the first shell command, when the session's working directory had a lowercase drive letter
No entry names this --plugin-dir . and --plugin-dir ./builtin writing generated type files for Claude Code's built-in plugins into a builtin folder
No entry names this /compact showing twice in Remote Control on web and mobile
No entry names this #100427[BUG] Remote Control: one /compact shows twice in the phone app (typed prompt and command-name record both rendered) Open
/plugin marketplace updates downloading a plugin again for each worktree or folder of a repository
No entry names this git stash, including after entering a worktree
No entry names this claude -p sometimes telling Claude that background commands get 30 minutes when they are stopped after 10
Probably headless-p-error-exit-carries-a-reason #103473[BUG] Print mode: Bash schema tells the model background commands get 30 min, but they are killed at 10 min (and the stated default flips between launches, breaking the prompt cache) Open
pkill, pgrep and pidwait in cloud sessions matching the session's own parent processes
Nothing to match on ask reason instead of another hook's when both asked
Nothing to match on ask being lost when another hook's defer finished first and could not be honored
Nothing to match on /desktop on macOS saying Claude Desktop is not installed when it is in ~/Applications
No entry names this #107903/desktop does not detect Claude Desktop installed in ~/Applications (still reproduces on 2.1.319, same as #41899) Open
#98476[Bug] Desktop command fails when claude.app installed in ~/Applications instead of /Applications Open
vcs.*) missing from Claude Desktop sessions that start on a pre-started process
No entry names this #108072[BUG] Desktop app sessions never report vcs.repository.* OTel attributes (spare process starts outside the repo) Open
claude agents ignoring --permission-mode, --dangerously-skip-permissions, --model, --effort or --agent typed before agents when a flag such as --verbose was also on the line
Probably remote-control-refuses-settings-before-the-verb-with-a-hi, remote-control-gets-a-settings-flag-for-spawned-sessions claude -p, SDK and VS Code sessions hanging, when memory ran out while the transcript was being read
Probably headless-p-error-exit-carries-a-reason grep -r foo "$PWD/src" asking for permission when the same command with the path typed out runs without asking
No entry names this desktop setting holds a field newer than the gateway, such as a managed MCP server's oauth.omitResource; it now warns and serves it
No entry names this .zshrc sets KSH_OPTION_PRINT
No entry names this #96333[BUG] Bash tool produces no output when KSH_OPTION_PRINT is set in ~/.zshrc Open
UserPromptSubmit hook blocked a replayed permission answer; these hooks no longer run on that replay
No entry names this claude remote-control accepting a y typed without Enter at its first-run question when the terminal's input ended
Probably remote-control-refuses-settings-before-the-verb-with-a-hi, remote-control-settings-forwarded-to-child-sessions-dire, remote-control-cli-adds-settings-and-debug-file-rejects code: rules
Nothing to match on PreToolUse mcp_tool hooks on stdio and HTTP MCP servers being skipped when their answer was over 16 MiB; the tool call is now refused
Probably remote-mcp-hook-calls-get-clearer-errors-on-oversize-results claude -p on Linux exiting 0 in silence when a full NFS export or a used-up NFS quota lost transcript lines: it now prints an error and, in text and json output, exits 1
Probably headless-p-error-exit-carries-a-reason claude mcp list output, when a stored MCP sign-in record was null
No entry names this /model picker: "More models" is now an unnumbered row below every other, and older models open under it without renumbering the rest
Probably remote-control-slash-commands-unavailable-here-get-specific, model-switch-notice-while-the-cloud-session-is-not-ready, model-picker-gets-a-collapsible-more-models-row-and-fable /device page: they no longer name gateway settings, which the gateway's own log still names
No entry names this claude install and the install scripts: they finish 2 to 4 seconds sooner
No entry names this claude --bg --worktree session that couldn't set up its worktree: it now shows the error from its beginning, not an internal tag or its last 200 characters
Probably worktree-startup-errors-routed-through-a-common-exit-helper /login is offered only if it's one of yours, and Claude now says where a copy may be after an account move
Probably remote-control-other-org-message-reworded, artifact-error-for-other-org-now-suggests-login, artifact-watch-error-messages-reworded, artifact-lookup-messages-cover-organizations-the-user-has-le claude remote-control's messages for an unanswered first-run question: each now says what to do next
Probably remote-control-refuses-settings-before-the-verb-with-a-hi, remote-control-settings-forwarded-to-child-sessions-dire, remote-control-cli-adds-settings-and-debug-file-rejects allowedProviders setting description to name Ultra alongside Pro and Max, and the forceLoginMethod description to cover every paid Claude plan
Probably managed-settings-read-failures-now-re-read-and-report-reason, login-method-setting-description-reworded allowUnsandboxedCommands: false or allowManagedDomainsOnly, and not to load in dontAsk mode
Probably hooksmods-refused-in-dontask-permission-mode $.tool.call, $.mcp.call) to raise no permission prompt, even under your own ask rules or in dontAsk and -p runs; deny rules and organization policy still apply
Probably hooksmods-refused-in-dontask-permission-mode fable: on Pro and Max plans at once, on Team and Enterprise plans once your organization allows it; use /model claude-fable-5-1 to switch back
Probably remote-control-slash-commands-unavailable-here-get-specific, model-switch-notice-while-the-cloud-session-is-not-ready, model-picker-gets-a-collapsible-more-models-row-and-fable /fast, /config and /model to name your current model when it supports fast mode, and removed "(Opus only)" from the VS Code row
Probably remote-control-slash-commands-unavailable-here-get-specific, model-switch-notice-while-the-cloud-session-is-not-ready, model-picker-gets-a-collapsible-more-models-row-and-fable, fast-mode-prompt-text-changed-no-longer-says-it-never-downg, fast-mode-off-status-line-hints-fast .claude/skills folders inside ~/.claude/plugins no longer load when Claude works on files there
No entry names this num_hooks and num_success after Bash and PowerShell calls
No entry names this AskUserQuestion in claude -p and SDK sessions to deny a host's permission answer that sets the undocumented afkTimeoutMs input field
Probably headless-p-error-exit-carries-a-reason, askuserquestion-timeout-can-be-admitted-as-nobody-answered, server-supplied-askuserquestion-prompt-text-removed-tengu-c 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
No entry names this policy_mode: layered in 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
No entry names this ANTHROPIC_*_BASE_URL variables in the gateway's environment. Set base_url on the upstream to use another endpoint
Probably gateway-warning-for-base-url-env-vars-changed-and-now-covers, bedrock-requests-always-get-an-explicit-regional-base-url .claude/skills when their files change; set CLAUDE_CODE_PLUGIN_DIR_WATCH=0 to turn this off
No entry names this cp and mv spelled by a path, such as /bin/cp, to get the same path checks as cp and mv
Probably telemetry-and-config-additions extraKnownMarketplaces take-overs from an undeclared source: saved data is set aside for cleanupPeriodDays, and options and secrets until claude plugin restore-set-aside
Probably new-claude-plugin-restore-set-aside-command-with-disca A model matched these bullets to the GitHub issues they fix, so a link can be wrong.
2 added and 2 removed, of 218 lines, about 38 words, in the prompt 14 of 27 arms receive. 4 other prompts also changed. The appended system-reminder blocks moved: 2 lines added, 2 lines removed.
Claude Code, interactive mode
5 prompt changes in this release could not be quoted from the build, so no entry on this page describes them.
834 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 57 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?