Claude Code 2.1.288
89changes- Add Ctrl+F to find a session by name and jump between groups in the agents view.
- Add $.ui.selection() for mods to read the text last selected in fullscreen mode.
- Add --max-findings to /code-review to control how many findings it reports.
- Compact an overlong conversation in auto mode instead of blocking on every tool call.
- Apply the background command time limit only to unattended sessions, not interactive ones.
New features
7Add session search and group jumps
Added Ctrl+F to find a session by name and Alt+↑/↓ to jump between groups in the agents view; both, and rename, can be rebound in keybindings.json
Add $.ui.selection() for mods
Added $.ui.selection() for mods: returns the text you last selected in fullscreen mode and, when the selection lies within one transcript row, that row
Add --max-findings to /code-review
Added --max-findings <n>|all to /code-review to report more or fewer findings than the usual limit; the choice is reused until you pass --max-findings default
Add built-in gh api for cloud sessions
Added a built-in gh api to cloud sessions whose image has no GitHub CLI, and fixed the built-in sending control characters from file names, jq filters or GitHub errors to the terminal
Recover a prompt cleared with Ctrl+C
Added recovery for a prompt cleared with Ctrl+C: pressing Up on the empty prompt brings the draft back, including pasted text and images
More changes (2) Fewer changes
Prompt to re-authenticate for wider MCP scope
Added a re-authenticate prompt when an MCP server asks for more OAuth scope during a tool call
Announce permission mode changes for screen readers
Added a screen reader mode announcement of the new permission mode when you approve a plan, including with Shift+Tab
Improvements
18Compact overlong conversations in auto mode
Improved auto mode: when a conversation grows too long for the client-side safety classifier to review, it is now compacted instead of prompting for, or failing, every tool call
Limit background command time only when unattended
Changed the background command time limit to apply only in unattended sessions (-p, Agent SDK, CI, cloud); terminal, desktop app and VS Code sessions have no limit
More changes (16) Fewer changes
Keep short announcements visible longer
Improved screen reader mode: short announcements, such as a deleted word, now stay on screen until your next key press or until something above them on screen changes
Mark answered questions in question dialogs
Improved screen reader mode: answered questions in question dialogs now say "answered" beside their box
Clarify /usage-credits message when disabled
Improved the /usage-credits message shown to Team and Enterprise members whose organization has turned off usage credit requests
Skip waiting for alwaysLoad false servers
Improved cloud sessions: a new conversation's first turn no longer waits for a stdio MCP server whose config sets alwaysLoad: false
Clarify who acted in advisory notes
Improved "You should know" notes to say "we", "the main agent" or "you" depending on who was responsible for a decision
Explain database size-limit write refusals
Improved the error for an artifact database write refused at the database's size limit: it now states the limit and what frees space
Shorten reasons for unchecked command parts
Improved Bash permission prompts to give a shorter reason when part of a command can't be checked before it runs
Improve gh api for self-hosted runners
Self-hosted runner: Improved the built-in gh api: a refused gh command now prints its gh api equivalent, --paginate follows every page of a repository's lists, and a nested claude no longer removes it
Keep sessions connected during credential renewal
Improved Remote Control's recovery from an expired server credential: sessions stay connected during renewal and are kept if it gives up after a server outage
Ignore Sonnet 5.5 pin in auto mode
Changed the client-side auto mode classifier to ignore an ANTHROPIC_DEFAULT_SONNET_MODEL pin that names Claude Sonnet 5.5 or Opus 5.5 and use Claude Sonnet 5 instead
Rename claude project purge to claude purge
Changed claude project purge to claude purge; the old name still works and prints a notice
Open the best-matching session on Enter
Changed the agents view n: filter (and Ctrl+F search) so Enter opens the session whose name matches best instead of the top row
Save auto-compact window per model
Changed /autocompact to save the auto-compact window per model, so each model keeps its own setting when you switch
Wait for confirmation on MCP URL prompts
Changed MCP URL prompts from servers that can't report when you're done to wait for "I'm done, continue" before the tool call continues, so you can finish in the browser first
[Claude Tag] Follow related threads in other channels
Improved Claude to also follow a related Slack thread in another channel that it only read, so updates there reach the conversation that depends on them
[Claude Tag] Clarify Configure page save errors
Improved save errors on a channel's Configure page: too-long channel instructions now say to shorten them, and a save refused for lost access no longer says to try again
Bug fixes
64Add a flag to disable structured outputs
Fixed session titles, memory recall and prompt hooks failing on Mantle or behind gateways that reject structured outputs; added CLAUDE_CODE_DISABLE_STRUCTURED_OUTPUTS to turn structured outputs off
Prompt before dangerous rm in scripts
Fixed a dangerous rm (such as one on / or the home directory) inside a bash -c or sh -c script running without a prompt in bypassPermissions mode or under a shell allow rule (anthropics/claude-code#96300)
Prompt before risky BASHPID assignments
Fixed Bash tool permission check to prompt before a BASHPID assignment whose value the shell would evaluate as arithmetic, instead of allowing it silently
Fix credentials.files protection on git config
Fixed sandbox.credentials.files entries on git config files not taking effect while permissions.blockReadsOutsideWorkingDirectories is on
Block calls instead of skipping failed hooks
Fixed PreToolUse and PermissionRequest hooks being skipped when matching them failed or the tool's input could not be serialized to JSON; the call is now blocked
Enforce model list on cloud session restart
Fixed restarted cloud sessions restoring a model that the organization's enforced model list refuses
More changes (58) Fewer changes
Recover from mid-response API timeouts
Fixed mid-response API timeouts failing the turn: non-interactive sessions and subagents now continue from the partial response, and thinking-only responses are retried
Fix auto-compact skipped on zero token usage
Fixed long conversations failing with "Prompt is too long" instead of auto-compacting when the last reply reported zero token usage
Fix --resume dropping restored context
Fixed --resume sometimes dropping files and other context that a compaction had just restored
Fix lost last response in resumed sessions
Fixed a resumed session sometimes not saving the last response of a turn, so that the next --resume showed the prompt unanswered
Fix resume loading a truncated transcript
Fixed resume occasionally loading a transcript cut short when the same session rewrote the file during the load
Preserve thinking when resuming older sessions
Fixed resuming a conversation started on 2.1.286 or earlier dropping the model's earlier thinking
Fix wrong permission rule in auto mode
Fixed auto mode denials pointing Claude at a Bash permission rule when the blocked tool was not Bash
Fix auto mode stuck on local classifier
Fixed auto mode on Bedrock and Mantle switching to the local classifier for the rest of the session after a request to an older model, such as a WebFetch summary or a sonnet subagent
Fix replies on a server-refused model
Fixed cloud sessions that restarted on a newly picked model replying with that model after the server refused it
Fix Cowork stuck on waiting-for-input
Fixed Cowork cloud sessions staying marked as waiting for input after a WebFetch permission prompt for an unapproved URL went unanswered for five minutes
Fix missing prompt suggestions on phone join
Fixed prompt suggestions not appearing on a phone that joins a Cowork cloud session started on another device
Fix mod buttons running the wrong action
Fixed a mod's button sometimes running a different button's action when pressed on a view drawn before Claude Code restarted
Fix blank pane on unparsable diffs
Fixed a plugin's pane showing nothing when one Code element held a diff that does not parse; it now draws as plain code
Substitute placeholders in plugin LSP config
Fixed plugin LSP servers receiving literal ${user_config.*} and ${CLAUDE_PLUGIN_ROOT} placeholders in initializationOptions and settings instead of substituted values or manifest defaults
Fix tool.call hook breaking worktree subagents
Fixed a plugin's tool.call hook making Bash fail and file searches read the wrong folder in subagents that run in a worktree
Fix git-subdir installs on older git
Fixed git-subdir plugin installs failing, or caching an incomplete plugin, on older git (before 2.39, e.g. Ubuntu 22.04's 2.34)
Show Configure options for --plugin-dir plugins
Fixed plugins loaded with --plugin-dir not showing "Configure options" in /plugin
Fix background sessions ending on plugin reload
Fixed background sessions ending when a plugin was reloaded or disabled while one of its timers or reads was still running
Fix repeated prompts for unquoted heredocs
Fixed sandboxed heredocs with an unquoted delimiter (python3 <<EOF) asking for approval on every run under sandbox auto-allow when the body holds only plain text and simple $VAR references
Fix crash opening background tasks dialog
Fixed fullscreen sessions exiting with "unrecoverable interface error" when opening the background tasks dialog while a plugin or mod showed rows above the prompt
Fix false delivered status for held messages
Fixed Claude reporting a message to another session as delivered when that session held it: the notice now says it wasn't delivered and names the session, and in SDK sessions Claude can now learn of it mid-turn
Fix unknown values in blocked_on_user spans
Fixed OpenTelemetry claude_code.tool.blocked_on_user spans reporting unknown source or decision in -p and SDK sessions and for PreToolUse hook approvals
Emit tool_decision for unanswered prompts
Fixed permission asks that ended unanswered, in -p or on an interrupted turn, emitting no tool_decision event
Allow Edit and Retry on pre-compact messages
Fixed Edit and Retry in Cowork cloud sessions refusing a message sent before /compact even though its history was still saved
Stop retrying for hours on stream failures
Fixed unattended sessions (CLAUDE_CODE_RETRY_WATCHDOG) retrying for hours after a very long response stream failed; Claude Code now streams again, and gives up after three timeouts
Report failed credential saves on /login
Fixed /login reporting "Login successful" when credentials could not be saved to secure storage; it now shows the failure, and offers a retry when the new login didn't take effect (anthropics/claude-code#73861)
Fix Stop during Bedrock lookup forcing fallback
Fixed a Stop during Bedrock credential lookup sometimes moving the session to a fallback model instead of ending the request
Fix duplicate auth refresh sign-ins on wake
Fixed a second gcpAuthRefresh/awsAuthRefresh browser sign-in opening when a laptop wakes from sleep while another Claude Code process is signing in
Fix plugin agent config in agent teams
Fixed agent teams: a plugin-defined agent spawned by name now runs with its own prompt, tools, disallowedTools and effort instead of the defaults
Fix SIGTERM ignored alongside SIGCONT
Fixed headless (-p / SDK) sessions occasionally ignoring SIGTERM when a supervisor such as timeout or systemd sends SIGCONT alongside it
Fix duplicate MCP calls on oversized results
Fixed MCP tool calls sometimes running twice when a remote server's result was over 16 MB or could not be parsed
Fix memory MCP tools missing for subagents
Fixed subagents in Claude Desktop's Code tab getting none of the tools of a user-configured MCP server named memory
Skip prompts for screenshots without auto mode
Fixed Claude in Chrome asking before every screenshot and page read on a site you allowed when auto mode is unavailable (such as with disableAutoMode or an older model); typing, navigation and JavaScript still ask
Fall back to HTTPS without SSH key
Fixed claude plugin install failing for GitHub-source plugins on macOS and Linux machines with no GitHub SSH key: the clone now falls back to HTTPS and prints a notice
Fix missing design systems in Artifact tool
Fixed Claude leaving out your organization's design systems when starting slides or a design with the Artifact tool on Team and Enterprise plans or machines with managed settings
Fix keyboard failing after self-restart on Windows
Fixed the keyboard not working on Windows after Claude Code restarts itself (first sign-in to a Claude apps gateway, provider setup, /tui)
Fix stall launching agents with many tools
Fixed a stall when launching an agent whose tools: lists very many Agent(...) entries
Fix Claude 3 sessions failing after PDF
Fixed sessions on Claude 3 Opus and Claude 3 Sonnet failing on every turn after a whole PDF entered the conversation
Fix false success on failed binary download
Fixed the npm auto-updater reporting success when the platform-native binary failed to download and only the placeholder claude stub was installed
Fix Remote Control archiving live sessions
Fixed Remote Control cleanup archiving a session that is still connected or was just re-attached by another Claude Code process
Show both SSH and HTTPS fetch errors
Fixed owner/repo plugin marketplaces showing only the second attempt's error when both the SSH and HTTPS fetch fail; both errors are now shown, with the transport tried first on top
Load rules on Write and Edit
Fixed path-scoped .claude/rules and nested CLAUDE.md files not loading when Write or Edit creates or changes a file in their scope (previously only Read loaded them)
Time out hanging LSP tool calls
Fixed LSP tool calls hanging indefinitely when a language server uses dynamic capability registration or stops responding; requests now time out after 60s (per-server requestTimeout)
Skip idle_prompt while background agents run
Fixed idle_prompt notification hooks firing while background agents are still running (anthropics/claude-code#93672)
Use server limits for the first request
Fixed the first request in a fresh environment or after a model switch using the built-in output limit and auto-compact window, not the server's; that request may now wait up to 1.5 seconds
Hide stale hint after sending queued messages
Fixed the "What should Claude do instead?" hint showing on the Interrupted row after sending queued messages with ctrl+enter
Fix unread sign-in replacing saved login
Fixed /login in a --bare session running a sign-in the session never reads, which could replace your saved login; it now says which credentials work
Fix missing agent_id in InstructionsLoaded hook
Fixed the InstructionsLoaded hook omitting agent_id and agent_type when a subagent's file access loads a rule or nested CLAUDE.md; rules and nested CLAUDE.md files loaded on file access now also report effort
Fix Agent tool rejecting all subagent types
Fixed the Agent tool in claude mcp serve always reporting no available agents and rejecting every subagent_type
Fix cursor not following search input
Fixed the terminal cursor not following the typed text in the fullscreen transcript viewer's search and in /theme's custom color search
Fix rule numbers opening search in /permissions
Fixed /permissions in screen reader mode: typing a rule's number now picks it instead of opening the search box
[VSCode] Fix stuck Needs authentication status
Fixed a claude.ai connector staying on "Needs authentication" after you authorize it: the MCP servers dialog now offers Check connection
[VSCode] Fix New Conversation affecting every view
Fixed the opt-in New Conversation shortcut (Cmd/Ctrl+N) starting a conversation in every visible Claude view instead of only the one you are in
[VSCode] Fix chat view resuming after archive
Fixed the chat view resuming the next saved session after you archive the one it shows; it now starts a new conversation instead
[Cloud sessions] Fix switch locked off while loading
Fixed the Cloud sessions switch in Claude Code admin settings staying locked off while an unrelated security setting was loading or had failed to load
[Cloud sessions] Fix Stop not cancelling queued message
Fixed pressing Stop while a self-hosted runner was still starting not cancelling the queued message, which could then run once the runner was up
[Claude Tag] Remove the always-failing scope option
Fixed Claude Tag admin settings offering a "Remove this scope" option, which always failed, on channels whose settings were created automatically
Fix false remote-off reports in plugin test
Fixed claude plugin test reporting mods as turned off remotely when it had only read an out-of-date saved setting