Skip to content

Claude Code 2.1.321

118changes
  • Add Claude Fable 5.2 as the new default Fable model with a 1M-token context window.
  • Apply a settings file to every session claude remote-control starts with a new --settings flag.
  • Answer background subagent reports that finish together in one model call instead of several.
  • Fix a read-only Bash command that could run outside the sandbox without a permission prompt.
  • Fix deny rules and protected-file prompts not applying to the actual file a cp or mv writes.

New features

3
More changes (1) Fewer changes

Improvements

41
More changes (35) Fewer changes

Reorganize the /model picker's More models row

Improved the /model picker: "More models" is now an unnumbered row below every other, and older models open under it without renumbering the rest

Gateway: Improve spend limits on busy replicas

Gateway: Improved spend limits on a busy replica, which now skips fewer checks and records spend sooner

Hide gateway setting names in sign-in errors

Improved the Claude apps gateway's sign-in limit messages in Claude Code and on the /device page: they no longer name gateway settings, which the gateway's own log still names

Speed up claude install and install scripts

Improved claude install and the install scripts: they finish 2 to 4 seconds sooner

Resume crashed Remote Control sessions faster

Improved how soon a Remote Control session whose process crashed can pick up a new message

Show full worktree setup error text

Improved the agent view's row for a 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

Clarify other-organization artifact messages

Improved the messages shown when an artifact is in another organization: /login is offered only if it's one of yours, and Claude now says where a copy may be after an account move

Reduce repeat reads of dropped images

Improved sessions with more images than fit in one request: Claude is told whether it has already seen a dropped image, and repeat reads of an unchanged image are limited

Clarify remote-control's unanswered prompts

Improved claude remote-control's messages for an unanswered first-run question: each now says what to do next

Hide credentials in gateway error messages

Improved Claude apps gateway error messages to show the gateway's address without any user name or password

Show spend limit reset in local time

Improved the Claude apps gateway's spend limit error: the reset now shows in your local time, such as "resets 4pm (America/Los_Angeles)"

Update plan names in messages to add Ultra

Updated the session-feedback message and the allowedProviders setting description to name Ultra alongside Pro and Max, and the forceLoginMethod description to cover every paid Claude plan

Suggest meter lists for multiple ratios

Updated the bundled dataviz skill to suggest a list of meters, one row per item, for several ratios that each have their own limit

Mention Ultra in upgrade-to-Max options

Updated the "Upgrade to Max" options to say "Upgrade to Max or Ultra"

Let mods turn on without asking by default

Changed mods Claude writes to turn on without asking in auto and bypass modes, even with managed allowUnsandboxedCommands: false or allowManagedDomainsOnly, and not to load in dontAsk mode

Let mods' own tool calls skip prompts

Changed mods' own tool calls ($.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

Move saved Fable 5.1 picks to fable alias

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-5-1 to switch back

Show current model name in fast mode labels

Changed the fast mode labels in /fast, /config and /model to name your current model when it supports fast mode, and removed "(Opus only)" from the VS Code row

Stop loading nested plugin skills on plugins

Changed nested skill loading: .claude/skills folders inside ~/.claude/plugins no longer load when Claude works on files there

Self-hosted runner: Exit 0 on refused runs

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: Refine how Claude Code finds programs

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

Exclude built-in hooks from telemetry counts

Changed OpenTelemetry hook events in non-interactive sessions to leave two built-in hooks out of num_hooks and num_success after Bash and PowerShell calls

Raise Code tab message batch limit to 20

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

Deny undocumented afkTimeoutMs field

Changed AskUserQuestion in claude -p and SDK sessions to deny a host's permission answer that sets the undocumented afkTimeoutMs input field

Tighten managed extraKnownMarketplaces

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

Apply org policy to embedded Chrome tools

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

Layer managed MCP servers across matches

Changed 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

▤ In the docs

Gateway: Ignore ANTHROPIC_*_BASE_URL on upstreams

Gateway: Changed Bedrock, Vertex and Foundry upstreams to ignore the ANTHROPIC_*_BASE_URL variables in the gateway's environment. Set base_url on the upstream to use another endpoint

▤ In the docs

Match claude.ai's Cyber Verification wording

Changed the Cyber Verification Program sentence on a cyber safeguard block to use the same words as claude.ai

Protect AskUserQuestion/plan results in Cowork

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

Trim long command text shown to Claude

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

Recommend Inter and Space Grotesk fonts

Changed artifact design guidance to treat Inter and Space Grotesk as good default typefaces

Narrow the account-email check's refusal

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

Lock Remote Control sessions to their folder

Changed Remote Control sessions to keep the folder they were started in: a connected client can no longer move a session to another folder

Set aside data from undeclared marketplaces

Changed managed extraKnownMarketplaces take-overs from an undeclared source: saved data is set aside for cleanupPeriodDays, and options and secrets until claude plugin restore-set-aside

Bug fixes

74
More changes (70) Fewer changes

Fix MCP OAuth secrets leaking on redirect

Fixed MCP OAuth requests sending secrets such as refresh tokens to another host when a server redirected them

Fix wrong usage-credits note for Fable

Fixed the "Draws from usage credits" note showing after a switch to Fable on Enterprise seats whose plan includes Fable

Fix fallback model declining requests

Fixed the fallback model sometimes declining a request or mentioning a stopped response when the first model was blocked before replying

Fix repeated SEP-2352 sign-in warning

Fixed claude mcp login printing a SEP-2352 issuer warning each time you sign in again to a server that Claude Code registered with

Fix NotebookEdit cell loss on network paths

Fixed NotebookEdit deleting or overwriting a notebook's first cell when the call named no cell, on network paths (UNC shares, WSL folders, /net mounts)

Fix Edit errors on network paths

Fixed Edit changing the first of several matches, replacing a whole file on an empty old_string, or editing a notebook as text, on network paths (UNC shares, WSL folders, /net mounts)

Fix /config default permission mode display

Fixed /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

Windows: Fix FileChanged hooks on lowercase drives

Windows: Fixed FileChanged hooks stopping for paths set through watchPaths after the first shell command, when the session's working directory had a lowercase drive letter

Fix --plugin-dir writing into builtin folder

Fixed --plugin-dir . and --plugin-dir ./builtin writing generated type files for Claude Code's built-in plugins into a builtin folder

Fix overcounted tool definition tokens

Fixed context usage estimates overcounting tool definitions when exact token counts are unavailable

Fix duplicate /compact in Remote Control

Fixed /compact showing twice in Remote Control on web and mobile

Fix duplicate plugin downloads per worktree

Fixed plugin auto-updates and /plugin marketplace updates downloading a plugin again for each worktree or folder of a repository

Fix lost subagent reports in auto mode

Fixed auto mode losing a background subagent's report that had started a turn when the conversation was compacted before Claude answered it

Fix skills missing after folder returns

Fixed project skills, commands and agents staying missing after their folder was removed and came back mid-session, such as with git stash, including after entering a worktree

Fix wrong background command timeout text

Fixed claude -p sometimes telling Claude that background commands get 30 minutes when they are stopped after 10

Fix pkill/pgrep matching own processes

Fixed pkill, pgrep and pidwait in cloud sessions matching the session's own parent processes

Fix lingering browser/computer-use processes

Fixed Claude in Chrome and computer use server processes staying alive after their parent exits behind a proxy that stalls connections

Fix hook output file mix-up

Fixed a large hook output pointing Claude to a file that held another hook's output from the same tool call

Fix wrong PreToolUse ask reason shown

Fixed apps that embed Claude Code getting their own PreToolUse hook's ask reason instead of another hook's when both asked

Fix lost PreToolUse ask on hook defer

Fixed a PreToolUse hook's ask being lost when another hook's defer finished first and could not be honored

Fix rare crash on permission prompts

Fixed a rare crash on permission prompts and background tasks

Fix skills missing after entering worktree

Fixed project skills, commands and agents disappearing from the menu in SDK and desktop sessions after Claude entered a worktree

Fix /desktop missing ~/Applications install

Fixed /desktop on macOS saying Claude Desktop is not installed when it is in ~/Applications

Fix missing vcs attributes on pre-started app

Fixed OpenTelemetry repository attributes (vcs.*) missing from Claude Desktop sessions that start on a pre-started process

Fix WebFetch treating no-text as content

Fixed WebFetch passing "No response from model" to Claude as page content when the model reading the page returned no text; the call now fails with a clear error

Fix startup prompt delay on clock rewind

Fixed screen reader mode delaying the prompt at startup when the system clock moved back while Claude Code was starting

Fix panel not returning after dialog closes

Fixed a command's panel sometimes not coming back after a dialog closed, leaving no panel or prompt on screen until the next redraw

Fix MCP disconnect from name clash in Code tab

Fixed a project's MCP server being disconnected after the first message in a Claude Desktop Code tab session when the desktop app has an MCP server or extension of the same name

Restore recovery from mid-reply TLS failures

Fixed a turn ending with "API Error: EPROTO: The TLS handshake failed" when the connection broke partway through a reply; the partial reply is kept or the request is retried, as before 2.1.302

Fix high memory use reading large files

Fixed high memory use in long sessions that repeatedly read single lines from large files

Fix fast mode failing instead of falling back

Fixed fast mode losing replies, switching models or retrying for minutes when fast serving is rate-limited, overloaded or failing; it now tries standard speed on the same model first

Fix excess agent reports in one request

Fixed one request carrying far more background agent reports than intended after Claude was stopped while several were waiting

Fix stuck messages from Code tab sessions

Fixed messages from a Claude Desktop Code tab session to your other sessions staying paused after you typed while Claude was still working

Fix claude agents ignoring early flags

Fixed 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

Fix false login-expired in Remote Control

Fixed a case where Remote Control showed as off with "login expired" in the desktop app while its connection was still up, because the app was briefly unable to hand over a sign-in token

Fix queued Remote Control message timing

Fixed a Remote Control message sent with "Queue for later" reaching Claude in the middle of the running turn

Fix git status read as clean on failure

Fixed the git status given to Claude at session start reading as clean when git status failed; it now reads as unavailable

Fix first file wrongly shown as staged

Fixed the git status given to Claude at session start showing the first listed file as staged when its changes were not staged

Fix hang on connectors with long names

Fixed a hang or out-of-memory crash when a connector in a cloud session lists resources with very long names

Fix stale connected state in Remote Control

Fixed Remote Control showing as connected after its connection closed for good while the session's history was still uploading

Fix screen reader mode turning off

Fixed screen reader mode turning off when you return to a conversation from agent view

Fix stray y taken as yes in screen reader

Fixed screen reader mode y/n questions taking as yes a stray "y" at the end of text typed as the question opened

Fix wrong host-not-allowed message

Fixed cloud sessions telling Claude that a host is probably not allowed when it is allowed and the network proxy could not connect to it

Fix MCP server showing no tools

Fixed an MCP server sometimes showing as connected with no tools after it was reconnected or the session's directory changed while its tools were still loading

Fix Bash commands running in wrong folder

Fixed Bash commands sometimes running in the wrong folder when Claude sent several at once and an earlier one changed folders

Retry uploads after proxy 507 errors

Fixed files Claude sends from cloud and Remote Control sessions failing at once when an upload gets an HTTP 507 from a proxy; the upload now gets a second try

Fix hangs reading transcript under low memory

Fixed new messages not being saved, and claude -p, SDK and VS Code sessions hanging, when memory ran out while the transcript was being read

Fix indefinite hang on stalled TLS

Fixed sign-in, token refresh and logout waiting indefinitely when something on the network accepts the connection but never answers the TLS handshake

Refuse files with hidden-character names

Fixed a sent file showing under a different name in the Claude apps when its name hid text-direction or control characters; Claude now refuses to send such a file

Fix sub-agent summaries running past tool calls

Fixed sub-agent progress summaries running a request to its end after the model replied with a tool call

Fix $PWD-based Bash reads asking to confirm

Fixed Bash reads such as grep -r foo "$PWD/src" asking for permission when the same command with the path typed out runs without asking

Serve gateway despite unknown newer field

Fixed Claude apps gateway refusing to start when a desktop setting holds a field newer than the gateway, such as a managed MCP server's oauth.omitResource; it now warns and serves it

Fix empty output with KSH_OPTION_PRINT

Fixed Bash commands coming back empty for zsh users whose .zshrc sets KSH_OPTION_PRINT

macOS: Fix MDM policy read timeout handling

macOS: Fixed "Unable to read managed policy settings" at startup when a valid MDM profile was slow to read: the read is retried once, and a timeout is now reported as a timeout

Fix agent view erasing saved input

Fixed agent view erasing another agent view's saved input text when it restarted for an update with nothing typed

Fix silent remote restarts on blocked hooks

Fixed remote sessions giving no reply after a restart when a UserPromptSubmit hook blocked a replayed permission answer; these hooks no longer run on that replay

Fix stray y accepted at remote-control prompt

Fixed claude remote-control accepting a y typed without Enter at its first-run question when the terminal's input ended

Fix missing spend limit warnings

Fixed Claude Desktop not showing spend limit warnings behind a Claude apps gateway whose policy has no code: rules

Fix crash from shrinking tool results

Fixed fullscreen sessions with some plugins ending with an error when a tool result that had got shorter scrolled off the top of the screen

Fix extra API calls from prompt suggestions

Fixed prompt suggestions making extra API requests when the model replied with a tool call

Refuse blank-glyph Bamum skill names

Fixed skill and command names holding a Bamum letter that some fonts draw as blank: they are now refused

Fix silent claude -p exit on NFS loss

Fixed 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

Fix background agents marked user-stopped

Fixed background agents being reported as stopped by the user, and refused a resume, when a service interrupted a cloud session and nobody pressed Stop

Fix NotebookEdit reporting false success

Fixed NotebookEdit reporting success to hooks and telemetry when the edit failed, for example when the notebook changed while the permission prompt was open

Fix startup failure on null MCP record

Fixed a blank screen or no answer at startup, and empty claude mcp list output, when a stored MCP sign-in record was null

Guide Artifact tool past large collections

Fixed the Artifact tool telling Claude to add filters when an artifact's database collection is too large to sort; it now says to read the collection page by page and sort the results

Fix dropped images on early restart

Fixed a restarted cloud session dropping a message with a large image when the restart came before Claude's first reply to it

Fix unanswered subagent message on restart

Fixed a restarted cloud session leaving a subagent's message to the main conversation unanswered

Tell Claude when a workflow is stopped

Fixed Claude not being told when you stop or pause a running workflow, so it no longer says the workflow is still running

[VSCode] Hide hidden characters in plugin names

Fixed a plugin's permission request showing a name with hidden characters: it now reads "a plugin"

Change details