Claude Code 2.1.299
190changes- Compact idle 200K+ token conversations before the one-hour prompt cache expires.
- Add CLAUDE_CODE_DISABLE_FILE_MENTIONS so @ mentions pass paths instead of file contents.
- Set the effort and fast mode new sessions start with from agent view.
- Stop Write from silently converting Shift-JIS and other encodings to UTF-8.
- Fix macOS sandboxed commands reaching denied paths through symlinks.
New features
18Disable inlining of @ file mentions
Added CLAUDE_CODE_DISABLE_FILE_MENTIONS environment variable: Claude gets the paths of @ file mentions to read with its file tools, not the files' contents
Configure idle compaction threshold
Added CLAUDE_CODE_IDLE_COMPACT_MIN_TOKENS to change idle compaction's 200K-token minimum; "idleCompaction": false in settings.json turns it off. Both need v2.1.290 or later
▤ In the docsSet effort and fast in agent view
Added /effort and /fast in agent view to set what new sessions start with; a bare /model, /effort or /fast there now opens a picker
Idle compaction in the terminal
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
▤ In the docsMore changes (14) Fewer changes
Read prompt attachments in mods
Added $.prompt.media so a prompt.submit hook of a mod installed by managed settings can read the bytes of an image or document attached to the prompt
Session identity for managed mods
Added $.session.identity() for mods installed by managed settings: how the session is signed in, and its organization and email when managed forceLoginOrgUUID lists that organization
More text and spacing options for mods
Added size, weight, mono and tabular to a mod's Text and named spacing steps (xs to xl) to its Box; bold is deprecated in favor of weight
Graceful shutdown for runner orchestrator
Added --shutdown-drain-sec to claude 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
More skill details for skill.prompt hook
Added the skill's frontmatter, source, plugin name and file path to what a mod's skill.prompt hook receives
Safe mode line in agents view
Added a "safe mode" line to the agents view peek panel for background sessions started with --safe-mode
Plan mode bypass permissions notice
Added a one-time note when plan mode won't ask before most edits or commands because bypass permissions is available
▤ In the docsExclude tools from plugin evals
Added --disallow-tools and a disallowed_tools case field to claude plugin eval to keep specific built-in tools, such as Agent, out of eval runs
Turn off attribution in Desktop Code tab
Added support for turning off the attribution block in Claude Desktop's Code tab: set CLAUDE_CODE_ATTRIBUTION_HEADER=0 in the env of a Claude apps gateway's code policy
▤ In the docsNew attribution field behind gateway
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
Queue door for session.append hook
Added a queue door to the session.append mod hook, so a mod can make the transcript's queue records differ from the prompt as sent
Note on skipped mod hooks
Added a note in /plugin and claude plugin validate naming a mod's hooks that are skipped where an organization manages Claude Code
More /usage warnings
Added /usage warnings for max effort, workflows, always-on context and large tool results
[Claude Tag] Delete canvases in Slack
Added the ability for Claude in Slack to delete a canvas it created in a channel when someone asks it to
Improvements
41Refuse actions when hooks cannot answer
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
▤ In the docsScope MCP allow rules to named servers
Changed MCP allow rules to reach only the server they name: mcp__a no longer allows a server named a__b or a_; write mcp__<server>__* with the server's whole normalized name
Scope agent tools lists to named servers
Changed agent tools: lists to take only the named MCP server's tools; list a server whose name contains __ as mcp__<whole name>__*
Run UserPromptSubmit before expansion
Changed UserPromptSubmit hooks to run before a skill or custom slash command expands, so a hook that blocks the prompt now also stops its embedded shell commands
Stronger managed policy enforcement against mods
Improved enforcement of managed permission rules, managed hooks and organization MCP tool names against user-installed mods
More changes (36) Fewer changes
Clearer runner git credential messages
Self-hosted runner: Improved what setup and doctor say when the runner's machine has no git credentials
Clearer cloud session login error
Improved the error shown when a cloud session is refused because your claude.ai login was rejected or has expired: it now tells you to sign in again with claude auth login
Lighter handling of @-mentioned PDFs
Improved memory and disk use when PDFs are @-mentioned: their bytes are no longer kept in the session transcript
Interruptible sandbox read rule expansion
Improved sandboxed Bash on Linux: Esc and Ctrl-C now work while wildcard read rules are being expanded for a command
Shorter unknown model ID warning
Improved the startup warning for an unrecognized model ID: it is now one short line under the welcome banner with a docs link, instead of a long raw message
Safer mod sharing in plugin-authoring
Improved the plugin-authoring skill's steps for sharing a mod: Claude asks public or private and checks the files before creating a GitHub repository
Better Unknown skill suggestions
Improved the Skill tool's "Unknown skill" error to suggest a skill's listed name when Claude wrongly adds the anthropic-skills: prefix to it
Clarify what /reload-skills does
Improved /reload-skills to say it re-sends the full skill list to Claude, which is how to get skills back after compaction
Clearer worktree base commit errors
Improved the error shown when a worktree cannot be created because its base commit does not resolve: it now says how the lookup ended and the likely reasons, such as a repository without a commit
Explain bare-repository worktree errors
Improved the error for creating a worktree at a bare-repository layout's root: it now explains the layout instead of an unrelated git error
Watch up to 25 artifacts
Improved artifact watching: a local session can now watch up to 25 published artifacts at once, up from 10
Richer background agent notifications
Improved background agent notifications: they now say when an agent is only waiting on its own background work, got a message too late for its result, or was stopped by a usage limit
Limit nesting in hook return values
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)
Stop following connector proxy redirects
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
Stop following MCP proxy redirects
Changed MCP requests in cloud and self-hosted runner sessions to no longer follow HTTP redirects from the session's MCP proxy
Skip skills with unsafe names
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 a name: is ignored
Tag mod roots in plugin test trees
Changed ui.render trees in claude plugin test: each mod's Box or Text root now carries part: { plugin } naming the mod, so exact-tree assertions must include it
Align WebFetch effort with web search
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
Time out unanswered mod calls in tests
Changed claude plugin test to fail a call to a mod's served $ method that has not answered within ten seconds, as a session does
Ask on flagged Claude in Chrome actions
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
Default to MCP protocol 2026-07-28
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=legacy opts out
Full text from first-party MCP servers
Changed MCP tool descriptions and server instructions from Anthropic's first-party MCP servers at api.anthropic.com to reach Claude in full instead of being cut at 4,096 characters
Limit PermissionRequest approvals in Cowork
Changed auto mode in Cowork cloud sessions: a PermissionRequest hook can no longer approve shell commands or subagents; those prompts wait or are denied
Resume command uses session ID
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
Shorter worktree resume command
Changed the resume command printed at exit of a worktree session: it uses -w and -r if the long flags do not fit the terminal. If neither fits a kept worktree, rows above it say where to run it
Time out unanswered /btw questions
Changed /btw side questions in non-interactive sessions to return a timeout error when still unanswered after 11 minutes
Delay skill UserPromptSubmit hooks
Changed a skill's own front-matter UserPromptSubmit hook to first run on the prompt after the one that invokes the skill
Include subagent turns in turn.start
Changed the turn.start mod hook to also hear subagents' and teammates' turns, not only the main loop's; tell them apart by e.agentId (absent on the main loop)
Drop git add . hint in /ultrareview
Changed the errors shown when /ultrareview can't upload a repository to stop suggesting git add ., which could stage or commit untracked files such as .env
Drop Cyber Verification Program prompt
Changed the cyber safeguard notice to no longer ask people to apply to the Cyber Verification Program
[VSCode] Faster startup without Python
Improved startup time in workspaces without a Python environment by starting Claude while the Python extension loads
[VSCode] Instant model and command menus
Improved startup: the model picker and slash command menu now fill in right away in a new window on a folder you've opened before
[Cloud sessions] Clearer deploy command display
Improved the deploy command shown when you create a self-hosted environment or environment key: it shows in full, wraps between arguments, and has a visible copy button
[Cloud sessions] Disable routines of deleted environments
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] Edit wrong messages in place
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] Join large channels quietly
Changed Claude in Slack to join channels with more than 100 members quietly, with no intro message
Bug fixes
131Fix image loss after an oversized image
Fixed long conversations losing newly attached images on every turn after one oversized image was refused by the API
Fix silent cutoffs after non-streaming fallback
Fixed replies cut off at the output token limit ending the turn silently, with no retry or error, after a request fell back to non-streaming mode
Fix Read and Write file encodings
Fixed Read garbling UTF-16LE files with a byte-order mark, and Write dropping byte-order marks or silently converting Windows-1252, Shift-JIS, GBK or UTF-16BE files to UTF-8 (now refused)
Fewer rm prompts on shell scripts
Fixed rm prompts on shell -c scripts that don't run rm. An rm built at run time in a script Claude Code can't read no longer prompts unless CLAUDE_CODE_STRICT_INLINE_SHELL_RM_PROMPT=1
Fix unprompted mod creation in accept-edits
Fixed the Write and Edit tools creating a mod in a session's mods folder without asking in accept-edits mode when the config directory's path has no .claude in it
Block sandboxed writes to plugin dirs
Fixed a sandboxed shell command being able to write into a --plugin-dir folder, whose code Claude Code reloads and runs outside the sandbox
Fix skipped hooks on deep inputs
Fixed plugin hooks being skipped when they pass a very deeply nested tool input or message on with next()
Fix MCP deny rules for double underscores
Fixed deny and ask rules for an MCP server (mcp__<server>) not applying when the server's normalized name contains a double underscore
Keep __proto__ keys in mod hook values
Fixed mods' hooks being handed values with __proto__ keys silently removed, which could skip a tool.check guard on a tool call another mod made
Fix macOS sandbox symlink bypass
Fixed sandboxed commands on macOS being able to read or write a denied path by its real location when the denied path passes through a symlink
Align sandbox glob parsing with settings
Fixed sandbox denyRead and allowRead globs with braces, parentheses or backslashes in a folder's name being read differently by Claude Code's settings checks than by the sandbox
Recheck sandbox globs after symlink swaps
Fixed sandbox glob allowRead and allowWrite entries not being re-checked when the directory they start from is replaced by a symlink
Tighten worktree isolation for awk
Fixed worktree-isolated sessions accepting some awk programs the isolation guard should refuse
Block hook-allowed writes in plan mode
Fixed plan mode running a write (an MCP tool's, a file's, a shell command's) that a PreToolUse hook or a mod's tool.check hook allowed
Fix hooks failing open without Git
Windows: Fixed PreToolUse and PermissionRequest hooks that set no shell failing open when Git for Windows isn't installed; a guard that fails now blocks the call
Gateway enforces blocked model policies
Fixed the Claude apps gateway serving models that a policy's availableModelsMatch: exact or deniedModels blocks; it now answers 400 and leaves them out of its model lists
Fix managed settings fetch with forceLoginOrgUUID
Fixed managed settings and organization policy not being fetched when the stored login's plan was changed locally on machines whose MDM or managed-settings file sets forceLoginOrgUUID
Run policy checks before connecting MCP
Fixed MCP servers added to a running headless or SDK session being connected before some policy checks ran; a blocked server is now reported as failed without being contacted
Don't start policy-blocked MCP servers
Fixed claude mcp list, claude mcp get and subagent MCP servers starting a server that a policy check blocks; it is now reported as blocked without being started
Preserve MDM forced login settings
Fixed forceLoginOrgUUID and forceLoginMethod from MDM or the managed-settings file being hidden or replaced by server-managed settings
Fix mods altering managed hook env
Fixed a user-installed mod's $.env.set changing the environment or shell prefix that command hooks from managed settings start with
More changes (110) Fewer changes
Fix wrong org in denied session error
Fixed the error for a denied cloud session, which could point to the wrong organization when a key from apiKeyHelper was in use
Fix false compaction failure near 200K
Fixed sessions on a gateway-specific model name behind a custom ANTHROPIC_BASE_URL ending with "Prompt is too long - automatic compaction failed" near 200K tokens instead of sending the request
Fix mod prompts extending headless runs
Fixed a mod that submits a prompt at the end of every turn keeping a headless run going past --max-turns, --max-budget-usd, an interrupt, or the end of the host's conversation
Fix follow-up text order in cloud sessions
Fixed cloud sessions sometimes showing the model's follow-up text after a tool group's one-line label
Unload remotely disabled mods
Fixed installed mods that kept running in an open session after being turned off remotely; they now unload between turns
Fix Edit and Retry on stopped messages
Fixed Edit and Retry in Cowork and claude.ai cloud sessions failing on a message stopped while its UserPromptSubmit or mod prompt.submit hooks were still running (regression in 2.1.296)
Fix credential leak in teammate description
Fixed an agent teammate started without a description keeping part of a credential from its prompt in the task description shown for it
Fix claude.ai view freezing on slow links
Fixed the claude.ai view of a Remote Control or cloud session freezing when a large batch of events could not finish uploading over a slow connection
Fix debug logging memory growth and hang
Fixed memory growth under heavy debug logging (--debug, --debug-file), and a hang on exit when the debug log is on a disk or network mount that stops accepting writes
Fix unbounded memory with stderr debug
Fixed memory growing without bound when debug output goes to stderr and the parent process stops reading it
Fix double-click collapsing transcript rows
Fixed double-clicking a word in an expanded transcript row in fullscreen mode collapsing the row instead of selecting the word
Fix empty .mcp.json reported invalid
Fixed claude mcp list, claude doctor and /status reporting an empty .mcp.json as invalid
Fix /btw jumping to bottom
Fixed /btw jumping the transcript to the bottom when asked while scrolled up in fullscreen
Fix WSL2 sandbox mkdir failure
Fixed sandboxed commands failing on WSL2 with bwrap: Can't mkdir /mnt/c/Program Files/ClaudeCode when the sandbox can write to the Windows drive
Fix scheduled wakeups during background tasks
Fixed scheduled wakeups (ScheduleWakeup) not firing while a background task was running in Agent SDK, IDE and claude remote-control sessions
Fix dropped prompt.submit rewrites with images
Fixed a mod's prompt.submit rewrite being silently dropped, sending the prompt as typed, for some prompts with a pasted image or an image that can't be read
Fix lost Remote Control prompts
Fixed a Remote Control prompt being lost when another queued prompt in the same batch failed to deliver
Keep Remote Control prompts in history
Fixed a Remote Control prompt that had passed its hooks being dropped from prompt history when Stop interrupted the next queued prompt's hook
Fix apparent config loss on Linux
Fixed sign-in, MCP servers and project trust appearing lost on Linux after a sandboxed command left an empty .config.json in the config directory
Fix false success in /sandbox
Fixed /sandbox reporting success when its setting could not be saved because the project folder is not writable; it now says nothing changed and why
Leave auto mode on unsupported models
Fixed SDK and headless sessions staying in auto mode after the host switched to a model that doesn't support auto mode
Connect MCP servers named constructor
Fixed MCP servers named constructor or prototype in an MCP config never being connected
Fix permission prompts for some MCP names
Fixed tools of an MCP server named constructor or toString asking for permission in auto mode and bypass permissions mode
Fix spurious synced plugin error
Fixed a spurious "synced[0] commands path not found" plugin error at startup when another Claude Code process had just updated a claude.ai-synced plugin
Fix Artifact retry advice on 403
Fixed the Artifact tool advising a retry when a publish that requests connectors or other access was refused by the server or a proxy (HTTP 403); it now says so and what to tell the user
Fix stale account plugins in Cowork
Fixed Cowork cloud sessions staying on an old account plugin after a plugin reload that timed out or a restart that reused a recent sync
Keep documents in mid-turn messages
Fixed document blocks being dropped from your own message sent while Claude is working
Fix Chrome prompts with no browser
Fixed Claude in Chrome asking for permission while no browser is connected; it now says the extension is not connected
Show update notice beside idle hint
Fixed the fullscreen "Update installed" notice being hidden behind the idle "/clear to save tokens" hint; it now shows beside the hint
Fix endless mod hook reruns
Fixed a mod hook, with or without a .catch, being run again without end when it waits on another mod's noun that does its work outside the call
Fix gateway key sent to Anthropic
Fixed the server-managed settings request to Anthropic carrying a third-party gateway's API key when a managed settings file sets ANTHROPIC_BASE_URL
Longer stream timeout for custom URLs
Fixed streams through a custom ANTHROPIC_BASE_URL being aborted after 180 seconds with no data arriving; they now wait 300 seconds, whether or not feature flags are loaded
Send updated mod answers after invalidate
Fixed a mod's tool.describe or prompt.section answer that changed after $.ui.invalidate never reaching Claude; the new answer is now sent on the next turn
Fix settings ignored for string server lists
Fixed a settings file being ignored when enabledMcpjsonServers or disabledMcpjsonServers held a single server name (letters, digits, - or _) as a plain string
Fix hook time budget in plugin test
Fixed claude plugin test pausing a hook's time budget while a served $ method it had called kept making $ calls after answering
Fix memory after resuming subagent-heavy sessions
Fixed high memory use after resuming a cloud or Remote Control session that had run many subagents
Refuse only failing keys in mod values
Fixed a value a mod's hook passes to next(), returns, or sends to a $ method being refused whole when one of its getters throws a RangeError; now only that key is refused
Fix usage shown above 100%
Fixed /usage, the approaching-limit warning (terminal and VS Code) and the status line sometimes showing more than 100% used (anthropics/claude-code#92238)
Fix CJK and emoji text wrapping
Fixed wrapped text mixing CJK or emoji with Latin words losing its last line or leaving a blank row at some terminal widths
Fix ignored answers during worker swaps
Fixed a cloud session sometimes ignoring the answer to a permission prompt or a question when the answer was given while the session's worker was being replaced
Fix nested text styles in tables
Fixed bold, italic and other nested text styles running over table borders or being lost on wrapped rows
Refresh changed account plugins in Cowork
Fixed Cowork cloud sessions running the old version of an account plugin changed, removed or disabled from another device until the session restarted
Fix has_hooks for hooks modules
Fixed the plugin_loaded telemetry event reporting has_hooks as false for plugins whose hooks are a hooks module, built-in plugins included
Fix full-size images sent mid-task
Fixed images attached while Claude is working (VS Code, desktop app, SDK) being sent at full size, which led to repeated image errors in long conversations
Fix image coordinate hint for large images
Fixed the image coordinate hint being missing or slightly off for large near-square images on newer models
Stop sending gateway API keys to Anthropic
Fixed feature flag, startup, fast mode, profile and feedback requests to Anthropic carrying the API key configured for a third-party gateway (ANTHROPIC_BASE_URL); Anthropic API keys are still sent
Fix mod hot reload after failures
Fixed a mod's hot reload ignoring a fix saved right after a failed reload; saving a mod file without changing it no longer reloads the mod
Fix mods refused for imported hook functions
Fixed mods loaded from source files being refused when a hook hands $ to a function declaration imported from another file of the mod
Fix duplicate tool calls on resume
Fixed a regression in 2.1.296 where cloud and headless sessions could run an approved tool call twice if a new message arrived as the session resumed after a restart
Fix wrong minute for scheduled tasks
Fixed scheduled tasks being announced for the wrong minute: the scheduling tools now report each job's actual next run time
Report invalid settings in claude -p
Fixed claude -p saying nothing when a settings file fails validation: text-output runs now list the file and the invalid entry on stderr
Fix long pastes losing text
Fixed long pastes in terminals without bracketed paste losing all text after an invisible control character, such as a mis-decoded curly quote
Fix MCP HTTP hangs on notifications
Fixed MCP connections over HTTP hanging for 30 seconds and failing when a server never ends the body of its reply to a notification (anthropics/claude-code#99014)
Fix /reload-plugins missing new installs
Fixed /reload-plugins not loading a plugin installed from another terminal a moment earlier
Fix blank rows under narrow CJK text
Fixed blank rows appearing under CJK and emoji text squeezed into a one-cell column, as in a very narrow terminal
Fix text cuts beside wide characters
Fixed text cut in the middle beside wide characters leaving a cell unused or being cut twice, and an emoji vanishing where a color changed inside it
Fix bare mode restart notices
Fixed bare mode notices in background sessions telling you to restart without --bare, which keeps the mode; they now name /stop and claude --resume, as safe mode ones do
Explain Left arrow on queued messages
Fixed the Left arrow key not saying why it stayed put over a queued message while Remote Control is on
Fix $.http.fetch refusing bodiless responses
Fixed $.http.fetch in mods refusing a 304, 204 or 205 response that declares a Content-Length over the 4 MiB body limit
Fix /config push notification toggles
Fixed /config push notification toggles sending your account the previous setting instead of the one you just chose
Fix unfinished replies after worker swap
Fixed a cloud session leaving a reply unfinished when its worker was replaced just after the reply hit the output token limit
Fix PDF page reads on Linux remote
Fixed Read failing on PDF page ranges in remote sessions on Linux that run without elevated privileges
Fix Enter answering held-back prompts
Fixed Enter answering a question or permission prompt that had been held back while you typed an unsent message
Retry stalled attachment downloads
Fixed a file attached to a message in a remote session being left out when its download stalled part way; the download is now tried again
Separate merged stream-json prompts
Fixed queued prompts sent as content blocks over --input-format stream-json running together without a line break when merged into one turn
Keep --exec-path wrapper MCP servers
Fixed self-hosted runner sessions shutting down MCP servers added through an --exec-path wrapper's own --mcp-config when connectors change mid-session
Fail plugin eval runs on API errors
Fixed claude plugin eval passing runs that the API refused: a run that ends on an API error now scores 0 without grading and makes the suite exit 1
Fix prompt wrapping with complex emoji
Fixed permission prompts cutting off or wrapping too early when the reason text contains keycap or joined emoji
Fix --bare dropping specially named MCP servers
Fixed --bare sessions leaving out an MCP server named constructor or prototype that --channels asked for
Fix resume reattaching wrong Remote Control
Fixed resuming a conversation from before /clear reattaching to the Remote Control session of the conversation after it
Fix verbatim prompts rerun as commands
Fixed a prompt starting with /, sent with the Agent SDK's verbatimPrompts option, running as a slash command when CLAUDE_CODE_RESUME_INTERRUPTED_TURN re-ran it after a restart
Keep safe mode for background sessions
Fixed background sessions started with --safe-mode silently leaving safe mode when the session was woken or restarted
Fix scheduled tasks when DST ends
Fixed scheduled tasks and loops firing again after every turn, or firing early, during the repeated hour when daylight saving time ends
Fix diff names with edge spaces
Fixed the diff view listing a new file whose name begins or ends with a space under the wrong name
Fix MCP calls during background reconnect
Fixed MCP tool calls in headless and SDK sessions sometimes failing or running twice when a background reconnect closed the connection they were running on
Fix /btw in non-interactive sessions
Fixed /btw side questions in non-interactive sessions not being cancelled when the session ends, and running a second time when the same request arrived again mid-run
Fix mcp add with --background argument
Fixed claude mcp add <name> -- <command> --background being started as a background session
Deliver --bg text to background sessions
Fixed a prompt or reply of --bg or --background to a background session being taken for the flag, which could start new sessions over and over instead of delivering it
Fix duplicate shell line in commands
Fixed a custom slash command's embedded shell line running twice when a newer message interrupted the command while its prompt was being checked
Clean up worktrees with reftable storage
Fixed --worktree sessions leaving a folder behind in .git/worktrees in repositories using git's reftable ref storage
Fix staged deletions in self-hosted runner
Fixed claude self-hosted-runner sessions using Anthropic-managed git showing tracked files as staged deletions when starting in a reused clone whose remote is still empty
Announce Ultracode to headless turns
Fixed Ultracode not being announced to scheduled tasks and other delivered headless turns when ultracode is on in settings
Send MCP headers during OAuth flows
Fixed MCP OAuth sign-in, token refresh and sign-out not sending configured headers/headersHelper values, which broke servers behind an authenticating proxy
Fix spurious Artifact publish diffs
Fixed the Artifact tool reporting a spurious one-line difference against the live page, and refusing the publish, when a large page with a line break after <head> is published again unchanged
Fix $.ui.open on paneless surfaces
Fixed $.ui.open reporting a mod's pane as placed in sessions where no surface draws it, such as Claude Code for VS Code and claude -p
Run agent.spawn hook on resume
Fixed the agent.spawn mod hook not running when a finished subagent or a killed teammate is resumed: the hook's event now carries resume, and a mod can refuse it
Fix plugin validate for $.state references
Fixed claude plugin validate rejecting a mod's $.state references when plugin.json has no types key and the contract is at ./types/index.d.ts
Fix mod dimColor in dark ANSI
Fixed a mod's dimColor text looking the same as plain text under the dark ANSI theme and on 16-color terminals
Fix duplicate background subagent reports
Fixed background subagents in auto mode sometimes sending their final report twice, once as the result and once as a message
Find subagents by name after restart
Fixed SendMessage not finding a subagent by name after the CLI restarted and resumed the session (names given before the last compaction still need the agent ID)
Fix unanswered messages after compaction
Fixed cloud sessions sometimes leaving a message unanswered when the worker was replaced right after a compaction
Fix duplicate answers after restart
Fixed cloud sessions and CLAUDE_CODE_RESUME_INTERRUPTED_TURN sometimes answering a finished turn again after a restart, when a tool ended the turn and a skill loaded in the same step
Fix dropped Remote Control photos
Fixed photos sent over Remote Control being dropped when their copy could not be saved on your computer; Claude now still sees the photo
Fix missing Remote Control files
Fixed files sent over Remote Control sometimes not arriving in long-running sessions
Fix 30 to 35 MiB file transfers
Fixed files of 30 to 35 MiB sent over Remote Control or to a cloud session not arriving
Fix orphaned PowerShell processes
Windows: Fixed the process-liveness check leaving orphaned PowerShell processes behind on busy machines until memory ran out
Fix session-wide command not found
Windows: Fixed Bash commands failing with "command not found" for a whole session when Git Bash did not report its PATH at startup
Fix runner ignoring --retire-at
Self-hosted runner: Fixed sessions that wrote output before their first turn being left with no release timer, ignoring --retire-at and delaying shutdown
[VSCode] Fix lost unsent prompts
Fixed an unsent prompt and its attachments being lost when switching to another session and back
[VSCode] Fix blank chats keeping processes
Fixed unused blank chats keeping a background Claude process running after starting a new conversation
[VSCode] Fix mic error when signed out
Fixed the mic button showing an internal error message when you are signed out of claude.ai; it now offers to sign you in
[Cloud sessions] Fix memory spike on file re-send
Fixed a memory spike that could crash a session when many of its existing output files were re-sent to it at once
[Cloud sessions] Keep routine environment on save
Fixed saving an edited routine silently moving it to a different environment when its own had been deleted or wasn't in the list; the routine now keeps its environment
[Cloud sessions] Fix session ID search
Fixed Sessions search in a self-hosted environment's admin details finding nothing when you paste the session ID from a session's link
[Cloud sessions] Fix empty-list message on searches
Fixed the Sessions and Runners lists in a self-hosted environment's admin details saying the list is empty when a search just has no matches
[Remote Control] Fix sends to unreachable computers
Fixed Remote Control sessions whose computer can't be reached still appearing to send your messages; your text now stays in the box and Send is off until it reconnects
[Claude Tag] Fix silence on untagged posts
Fixed Claude in Slack staying silent on untagged posts that a channel's own instructions tell it to answer, such as posts that tag someone else, in channels with guests
[Claude Tag] Fix false cross-channel reply reports
Fixed Claude in Slack reporting a reply to another channel's thread as posted when it isn't a member there; it now says the reply wasn't delivered and how to fix that
[Claude Tag] Fix missing answers without python3
Fixed Claude in Slack sometimes ending a turn without posting its answer when a self-hosted runner's image has no python3
[Code Review] Fix spend-limit links
Fixed Code Review's monthly spend-limit comment and check-run card sending admins to Claude Code settings; they now link the Usage page, where that limit is set