{"id":"src_uDHhSwgE7ufu05-FTCFtr","slug":"claude-code-releases","name":"Claude Code","type":"scrape","url":"https://code.claude.com/docs/en/changelog","orgId":"org_HXldkKAl-bZUWW6lNnELP","productId":"prod_QMqJRAH7ip4wF-kRw5Q8f","productSlug":"claude-code","org":{"id":"org_HXldkKAl-bZUWW6lNnELP","slug":"anthropic","name":"Anthropic"},"isPrimary":false,"isHidden":false,"discovery":"curated","fetchPriority":"normal","metadata":"{\"noFeedFound\":false,\"hasVideoEmbeds\":false,\"feedUrl\":\"https://code.claude.com/docs/en/changelog/rss.xml\",\"feedType\":\"rss\",\"feedDiscoveredAt\":\"2026-04-20T13:12:08.691Z\",\"feedLastModified\":\"Fri, 08 May 2026 22:00:10 GMT\",\"githubUrl\":\"https://github.com/anthropics/claude-code\",\"markdownUrl\":\"https://code.claude.com/docs/en/changelog.md\"}","notice":null,"kind":"tool","stars":156639,"starsFetchedAt":"2026-11-22T11:52:37.229Z","releaseCount":390,"releasesLast30Days":17,"avgReleasesPerWeek":5.7,"latestVersion":"v2.1.330","latestDate":"2026-11-20T22:32:56.000Z","changelogUrl":null,"hasChangelogFile":true,"lastFetchedAt":"2026-11-22T11:52:37.229Z","lastPolledAt":"2026-11-22T11:52:18.504Z","changeDetectedAt":null,"trackingSince":"2025-06-30T00:00:00.000Z","releases":[{"id":"rel_jNOcICaxjQdbBA67X_8mN","version":"v2.1.330","type":"feature","title":"v2.1.330","summary":"Claude Code now applies an Anthropic Enterprise Admin Console organization's settings and cloud provider or gateway after claude auth login, and ships easel, a built-in mod that draws charts on request and under MCP metrics results. Fixed compaction freezes lasting minutes, WebFetch hangs, and worktree deletion that could remove a directory another session was working in.","titleGenerated":"Claude Code 2.1.330 adds Enterprise Admin Console sign-in and easel chart mod","titleShort":"Enterprise org sign-in; easel charts MCP metrics","breaking":"minor","importance":3,"content":"## What's changed\n\n- Added sign-in to organizations managed in Anthropic's Enterprise Admin Console: after `claude auth login`, Claude Code applies the organization's settings and uses its cloud provider or gateway\n- Added easel, a built-in mod that draws charts when asked and under MCP metrics results, on Pro, Max and, where available, Team and Enterprise on macOS and Linux; turn automatic charts off in `/config`\n- Added `CLAUDE_CODE_FULL_SIGN_IN_LINK=1` to print the full sign-in link, if the short one shows an error or a script reads values from it\n- Self-hosted runner: Added `--start-dir` / `SELF_HOSTED_RUNNER_START_DIR` to have Claude start its work in a folder of a repository, while settings and hooks load as without it\n- Added a notice in cloud and self-hosted runner sessions when a plugin marketplace can't be fetched because its repository isn't one of the session's\n- Self-hosted runner: Added a notice when Anthropic-managed git refuses a plugin's own repository, naming the plugin and what to do\n- Added `commandPrinted` to `claude plugin install --json` results, true when the plugin's install command wrote to stderr, and `commandFailed` when it exited with an error, was killed or timed out\n- Added `claude_code_self_hosted_runner_order_to_registered_seconds` to the self-hosted runner's `/metrics`: seconds from work order issued to registration accepted, on orchestrator-spawned runners\n- Added `claude_code_self_hosted_runner_order_to_first_session_seconds` to the self-hosted runner's `/metrics`: seconds from work order to first session handed over, on runners started for a session\n- Fixed Claude Code freezing for up to several minutes right after compacting a very long conversation\n- Fixed WebFetch freezing the session for minutes on some pages, and returning an empty result for some others\n- Fixed `ExitWorktree` and the worktree exit dialog deleting a worktree that another running Claude Code session was working in\n- Fixed changes to user settings being lost when several Claude Code processes saved them at the same time\n- Fixed `--continue` and `--resume` opening an empty session or failing when the last saved message had a missing or invalid timestamp\n- Fixed the first prompt after a sleep of two minutes or more waiting three minutes on a connection that had died, on Linux and WSL\n- Fixed turns failing with \"JSON schema is invalid\", and session files growing by many megabytes, when many MCP tools have an input schema that the API refuses\n- Linux: Fixed Bash commands waiting seconds and showing no file diff for the rest of a session in slow repositories with large ignored folders such as `node_modules`\n- Windows: Fixed Bash commands running slower for the rest of a session after Git Bash took 10 to 60 seconds to start, on machines with no custom start-up files\n- Fixed `/rewind` sometimes restoring an older version of a file when two queued prompts, or stacked skills, ran together right after a turn that changed it\n- Fixed text and images added when approving a permission prompt being dropped when the approved command or tool then failed\n- Fixed MCP resources from claude.ai connectors and plugin servers not being attached when @-mentioned in the terminal's prompt box\n- Fixed an MCP server that subagents declare being shut down when one of them finished, while other subagents or the session were still using it\n- Fixed Claude in Chrome staying \"not connected\" after a brief connection problem, even though Chrome was open and signed in\n- Fixed terminal sessions on Amazon Bedrock (Mantle) started with no model selected ignoring a pinned Opus model, and still calling an unavailable default model after announcing a fallback\n- Fixed the rest of what you were typing being lost when a plan arrived in fullscreen mode\n- Fixed claude.ai connectors with different names being dropped when they share a server URL (anthropics/claude-code#101856)\n- Fixed plugins going missing from the installed list when several `claude plugin` commands ran at the same time\n- Fixed task dependencies and metadata being lost when several task updates ran at the same time\n- Fixed incomplete backups of `~/.claude.json` being kept, and displacing good ones, when the disk was full\n- Fixed a long pause in Code Mode after code that read hundreds of files in projects with path-scoped rules or Read deny rules\n- Fixed the built-in chart tool failing when Claude called it in Code Mode\n- Fixed Code Mode ignoring an MCP tool's request to end the turn from a script in auto and bypass permissions modes\n- Fixed Code Mode in auto mode dropping earlier thinking and rewriting the prompt cache when a mod added a note after a script loaded a skill\n- Fixed Code Mode not restoring a file that a script had read after Esc and `/compact`\n- Fixed `/context` undercounting system tools in Code Mode before Claude's first reply\n- Fixed a stopped background subagent sometimes not staying stopped\n- Fixed Claude still waiting for a background agent that was stopped just as it finished, in sessions run by an app or the SDK\n- Fixed background sessions sometimes waking in their previous permission mode after approving a plan\n- Remote Control: Fixed the note about background agents ended by a Stop from a phone or the web showing in the terminal as if you had typed it\n- Remote Control: Fixed the mobile apps hiding the Diff and pull request buttons, or showing a stopped session as working, after turning it on mid-conversation\n- Fixed environment variables written by a SessionStart hook being lost with `claude -p --continue`\n- Fixed `claude -p` exiting 0 with no message when its output went to a file on a full disk: it now prints an error and exits 1, though a full NFS export may go unnoticed\n- Fixed the system prompt keeping the earlier model's instructions when the model was changed after `/context` but before the first message\n- Fixed Claude sometimes replying to its commit attribution reminder and ending the turn early after the model changed mid-turn; it can still happen where managed settings set the PR attribution line\n- Fixed Esc not ending an LSP tool call that was waiting for a slow or stuck language server\n- Fixed LSP tool calls and diagnostics failing when several calls or edits arrived while a language server was still starting\n- Fixed `/mcp` listing plugin servers as built-in: they now stand under \"Plugin MCPs\", `--mcp-config` and built-in servers under \"Other MCPs\", and a server's page names its plugin\n- Fixed an \"MCP server blocked by enterprise policy: easel\" warning showing at every start when an MCP allowlist or denylist excludes the built-in chart tool, which stays blocked\n- Fixed `/stats` counting new activity in UTC days instead of your local time zone\n- Fixed `/ultrareview` failing or reviewing the wrong code after the shell moved into another repository: it now stops and names both folders\n- Fixed Grep reporting no matches when the pattern held a line break without multiline mode; it now reports the rejected pattern\n- Fixed Jupyter notebooks saved with a byte order mark being reported as invalid JSON by the Read and NotebookEdit tools\n- Fixed link reference definitions being dropped from CLAUDE.md and memory files that contain an HTML comment\n- Fixed running Monitors keeping the earlier conversation in memory until they end\n- Fixed some Claude Desktop sessions on third-party providers, other than scheduled tasks, ending a turn with \"Prompt is too long\" when a PDF larger than the context window followed a long conversation\n- Fixed the choice between two user or project agent files declaring the same name depending on the order the file search listed them\n- Fixed `claude daemon logs` never exiting inside a git repository when there is no log file\n- Fixed markdown hiding a character or two after bold or italic text when a backslash came before an emoji later in the paragraph\n- Fixed Edit refusing some valid hook commands in Claude settings files\n- Fixed `Read` with a page range showing both halves of a scanned two-page spread on each of its pages\n- Fixed line breaks and class stereotypes missing from Mermaid diagrams in HTML artifacts\n- Fixed the Project instructions picker in `/config` showing only one of its four choices in a 24-row fullscreen window\n- Fixed a subagent's own subagents and forks running on the plan-mode model in plan mode under `opusplan` or `haiku`\n- Fixed WebFetch page summaries ignoring the `/effort` level when a gateway refuses the small fast model\n- Fixed `claude -p` on Google Cloud's Agent Platform ignoring `--settings '{\"fastMode\":true}'` when the model comes from `ANTHROPIC_MODEL` or a settings file; such runs now bill at the fast mode price\n- Fixed startup under `forceLoginOrgUUID` being refused after one temporary server error or dropped connection\n- Fixed `claude mcp remove` giving no way forward for a claude.ai connector: it now says to turn it off in /mcp or remove it on claude.ai\n- Fixed the /mcp \"Connect on claude.ai\" check saying \"not connected\" when claude.ai has a connector at the server's URL without its query\n- Fixed `/install-github-app` and `/design-login` showing a generic failure for an account on hold instead of the on-hold notice and appeal link\n- Fixed a running MCP task being reported as stopped when you move a session to the background; the background session can now follow the task, but not if its server uses stdio or sets headers\n- Fixed self-hosted runners ignoring a session until they exited, after its failure report was lost or the session was given back to the same runner\n- Fixed self-hosted sessions losing the prompt cache and earlier thinking when the list of repositories a restart couldn't prepare changed\n- Self-hosted runner: Fixed tool calls behind a permission prompt vanishing with no result when a new runner took over the prompt, in sessions with telemetry off\n- Fixed resumed self-hosted runner sessions starting without their pushed work in large repositories when `--push-outcome-on-release` is set\n- Fixed self-hosted runner sessions being shown as working when Claude Code hangs at every start and `--startup-timeout-min` is from 2 to about 6\n- Fixed `pkill -f` in self-hosted runner sessions on Linux ending its own shell with exit code 144, or the runner (needs a `pkill` with `--ignore-ancestors`)\n- Fixed scheduled tasks on self-hosted runners having no tool to notify their owner when telemetry is off or a third-party provider is used\n- Self-hosted runner: Fixed one slow `spawn-runner` hook keeping the orchestrator from claiming sessions for its free slots; it now polls every 4 to 6 seconds while a slot is free\n- Self-hosted runner: Fixed a 403, such as an IP allowlist's, logged as a bad secret or token: most such lines now lead with the server's reason, so update log alerts that match the old words\n- Fixed self-hosted runners' warn mode for repository settings stopping at the first finding of most kinds: it now goes on, up to 50 for a session, and says if it stops early\n- Self-hosted runner: Fixed runners that start on a full disk taking sessions and failing them; they now wait for room and exit 1 after a minute with no session, unless started from a work order\n- Self-hosted runner: Fixed a failed start suggesting a retry when the runner's machine had no git credentials; the error now says whom to ask and what to do if the repository is wrong\n- Fixed the self-hosted runner's `last_poll_age_seconds` reading 0 while the runner had never polled successfully\n- Fixed the notice about a skipped scheduled task that names no organization: it now says to start Claude Code with `--resume`, because `/resume` inside a running session doesn't run the task\n- Fixed `claude plugin validate` wrongly warning that a mod's `AbovePrompt` or `AbovePromptList` hook never runs in VS Code\n- Fixed `/design-sync` deleting a file it had just uploaded when Claude Design stored that file under a changed name\n- Reverted 2.1.329's removal of the automatic retry when `max_tokens` exceeds a model's or gateway's limit\n- Improved the claude.ai sign-in link shown by /login, `claude auth login` and `claude setup-token`: it is now much shorter\n- Windows: Improved Grep and Glob speed by no longer starting a console host for each built-in ripgrep search on a local path\n- Improved the speed of file diffs\n- Improved `/config` search: most settings whose name starts with what you type come first, \"effort\" finds Model, and \"dark\" or \"light\" finds Theme\n- Improved `/resume`: sessions where you only opened a screen such as `/model` and left are no longer listed\n- Improved Reconnect in `/mcp` and `/plugin`: a server's page now stays open, with the cause in view, when the server still fails or needs sign-in\n- Improved the error on macOS when sandbox deny lists are too long: it now names the long lists and how to shorten them\n- Improved `/plugin` in fullscreen mode: the page of a plugin with errors now starts with those errors\n- Improved `/ultrareview`'s cost confirmation on Team and Enterprise plans: it now says how much of your monthly spend limit is left when that is less than a review usually costs\n- Improved self-hosted runner sessions on Amazon Bedrock or Google Cloud's Agent Platform to say so when they start on another default model because your account refused the first\n- Windows: Improved the error when a session's folder is on a drive Claude Code can't verify: it says why, and whether retrying can help\n- Improved Code Mode: Claude is now told how to look at images and PDFs\n- Improved Code Mode: after Esc, \"Interrupted\" now shows under the stopped command in place of a red error\n- Improved Code Mode: the terminal freezes for far less time after a subagent's script makes thousands of tool calls\n- Changed recurring scheduled tasks and `/loop` to run up to 30 days instead of 7 while their session keeps running, and say when they end; tasks made before this release still end at 7 days\n- Changed more `rm` commands aimed at the root, home, a top-level folder or the working folder to ask for approval; where nobody can answer, they're refused\n- Changed `deniedMcpServers` name entries to hide only the named claude.ai connector, not others at its URL; to block them all, deny its exact URL\n- Changed `claude --teleport` in some setups to print the git commands for switching branch instead of running them; with `-p` it exits 1 until you switch\n- Changed unrecognized model names containing \"sonnet\" or \"haiku\" to use the shorter system prompt, like other unrecognized names; `CLAUDE_CODE_SIMPLE_SYSTEM_PROMPT=0` gives every model the full one\n- Changed skills that Claude runs with a `model` setting, other than `context: fork` skills, to leave subagents, forks, compaction and `/recap` on the session's model\n- Changed a subagent's model fallback to carry over: agents with no model and forks it then starts run on the fallback, and the agents they start use its Bedrock region\n- Self-hosted runner: Changed `spawn-runner` hooks to start while earlier ones still run, up to `--hook-concurrency` per orchestrator (4 by default); set it to 1 if yours can't run beside another\n- Self-hosted runner: Changed Bash commands to skip the file diff when nothing needs it, so transcripts can lack it; set `bashEditDiffEnabled` to true in managed settings to get every diff\n- Changed self-hosted runners to stop a session's start at once, without trying other runners, when a checkout hook exits with status 108\n- Changed prompt hooks to accept model aliases such as `haiku` or `sonnet` in their `model` field\n- Changed `/design-sync`'s fallback build of the reference Storybook to use only a Storybook installed in the project\n- Changed subagent progress summaries in terminal sessions signed in to a Claude apps gateway to be written from short notes instead of the whole conversation, using far fewer tokens\n- [VSCode] Added `/fork` to fork a conversation after its last reply (anthropics/claude-code#69272)\n- [VSCode] Added `/workflows`, which opens the agent map with running workflows\n- [VSCode] Added the `claudeCode.escapeStopsClaude` setting; turn it off so Escape does not stop Claude (anthropics/claude-code#91536)\n- [VSCode] Added the `claudeCode.pinPrompt` setting to let your message scroll away instead of staying pinned\n- [VSCode] Added the `claudeCode.contextUsageThreshold` setting for when the context indicator shows (anthropics/claude-code#93036)\n- [VSCode] Added a `prompt` argument to the \"Open in Side Bar\" command for keybindings and extensions\n- [VSCode] Added Move up and Move down to a session group's menu (anthropics/claude-code#98654)\n- [VSCode] Fixed VS Code opening and focusing its terminal on every prompt of a connected terminal session\n- [VSCode] Fixed copying a code block adding a trailing newline that ran the command on paste (anthropics/claude-code#96976)\n- [VSCode] Fixed holding Enter on `/rewind` restoring files, and a double click on Remove or Delete removing a memory, MCP server or plugin marketplace, without confirmation\n- [VSCode] Fixed Enter on a question card sending your earlier answer instead of picking the focused one, and arrow keys not working after an \"Other\" answer; answered questions reopen on your pick\n- [VSCode] Fixed a held-down Enter key sending a half-written message or running a typed slash command\n- [VSCode] Fixed Cmd+A, C, V, X and Z not working in the rename box and dialogs on macOS\n- [VSCode] Fixed Ctrl+F and Ctrl+P not moving the caret in the message box on macOS\n- [VSCode] Fixed the Accept and Reject buttons doing nothing on a proposed diff moved to a new window\n- [VSCode] Fixed Accept and Reject Proposed Changes showing over the chat, the terminal tab and other files; they now show only on the proposed diff\n- [VSCode] Fixed a slow-starting Claude tab taking the keyboard away from the file you had gone on typing in\n- [VSCode] Fixed unsaved changes and new problems being missed on macOS and Linux for files whose path contains `#`, `?` or `%20`, such as a `C#` folder\n- [VSCode] Fixed whole-line selections counting one line too many in the selection chip, in @-mentions and in what Claude is told, and file links leaving out the last line of a range\n- [VSCode] Fixed a conversation forked in the panel re-running an old, unanswered prompt by itself after a window reload\n- [VSCode] Fixed a forked conversation running the original conversation's `/loop` schedules a second time\n- [VSCode] Fixed /clear giving no sign that the session still runs without a newly trusted folder's CLAUDE.md and settings\n- [VSCode] Fixed dialogs closing, and losing typed text, when a text selection was dragged past their edge\n- [VSCode] Fixed advisor calls showing as \"Unsupported content type\" (anthropics/claude-code#73499, anthropics/claude-code#64158, anthropics/claude-code#90471, anthropics/claude-code#73637)\n- [VSCode] Fixed the session list showing no status for the sidebar chat's session (anthropics/claude-code#93898)\n- [VSCode] Fixed the plan preview's Add Comment button vanishing after a fast double click on a word\n- [VSCode] Fixed long lines cut off in the Edit Notebook Cell row, permission rules squeezed in a narrow panel, hard-to-see question card marks, and menus moving the chat after a declined trust prompt\n- [VSCode] Fixed signing in to an account on hold showing a generic failure instead of the on-hold notice and appeal link\n","publishedAt":"2026-11-20T22:32:56.000Z","fetchedAt":"2026-11-20T23:27:51.313Z","url":"https://code.claude.com/docs/en/changelog#2-1-330","media":[],"coverageCount":0},{"id":"rel_HDpQ2VEVw3TFHF5ZzNEig","version":"v2.1.329","type":"feature","title":"v2.1.329","summary":"Fixed ~/.claude.json sometimes being reset to defaults when a network stall or file lock made it unreadable during a save, along with repeated macOS sign-outs when Claude Code could read the Keychain but not save to it. Saved scheduled tasks in new signed-in sessions mostly need scheduling again, and 2.1.326's automatic retry at the max_tokens limit is removed.","titleGenerated":"Claude Code 2.1.329 stops ~/.claude.json resetting to defaults","titleShort":"~/.claude.json no longer resets; saved tasks must be rescheduled","breaking":"major","importance":4,"content":"## What's changed\n\n- Added a trust dialog warning that says where a folder's hooks are declared\n- Added a warning in the conversation, in Claude Desktop and other non-interactive hosts, when hooks in managed settings fail to load\n- Added this session's scheduled tasks, `/loop` wakeups, the goal and this conversation's forks to `/tasks`; x stops one after asking\n- Added a \"Server message\" row to a server's page in `/mcp` that shows the JSON-RPC error message of a remote HTTP MCP server that refuses the connection\n- Added a \"Marketplace auto-update\" row to `/status` that names managed marketplaces that get no automatic updates, and why\n- Added support for turning on Code Mode while OpenTelemetry events and metrics are enabled; each tool call the code makes sends its events; tracing still keeps it off\n- Fixed `~/.claude.json` sometimes being reset to defaults when a network stall or file lock made it unreadable and cleared during a save\n- Fixed repeated sign-outs on macOS when Claude Code could read the Keychain but not save to it. That session now stops at expiry and says why, and other sessions stay signed in\n- Fixed `/terminal-setup` corrupting VS Code settings or keybindings files around block comments, and reporting success while losing the Shift+Enter binding where keybindings.json has an unclosed string\n- Fixed files accepted through the IDE diff being saved without their last newline, and losing their byte order mark (BOM) when an edit was accepted by saving the diff in VS Code\n- Fixed picking a cloud session in `/teleport` replacing your open conversation: the pick now starts a new session, as `/clear` does, and `/resume` brings the old one back\n- Fixed `/usage-credits` asking Pro and Max users to log in again after opening the browser, and the link disappearing after 8 seconds if the browser doesn't open and a task stops for usage credits\n- Fixed `/usage` saying usage credits are off when they had run out or reached the monthly limit, and `/model` not telling Pro and Max users when that limit blocks a model that runs on usage credits\n- Fixed the spend limit menu showing your spend as your monthly limit and wrapping its rows under the pointer in narrow windows\n- Fixed a status line script that prints without end freezing the prompt and using gigabytes of memory\n- Fixed an empty, frozen screen after approving managed settings while the --resume picker, the trust dialog or another set-up screen was waiting\n- Fixed choosing Default in `/model` replacing your saved model with your organization's default when the Default row already showed your saved model\n- Fixed an MCP server being signed in again after you cleared its authentication while a token refresh was running\n- Fixed background MCP tasks that stalled after a slow reconnect: Claude Code now reconnects most HTTP MCP servers by itself when one is left on an older protocol version\n- Fixed PDFs, images and `.docx`, `.xlsx` or `.pptx` files that MCP servers or websites send as `application/octet-stream` being saved as .bin, which kept Claude from reading the PDFs and images\n- Fixed `/code-review <pr number>` sometimes reviewing the checked-out branch instead of the pull request without saying so: it now reads the pull request with `gh pr diff`, or says when it can't\n- Fixed `/code-review` stopping at a permission prompt, on most models, to look for a `~/.claude/CLAUDE.md` that doesn't exist\n- Self-hosted runner: Fixed `--push-outcome-on-release` sometimes deleting a session's branch on the git host\n- Fixed Code Mode hiding a script's finished edits and the stopped command after Esc, and losing them on exit\n- Fixed a runaway loop in Code Mode filling the machine's memory; it is now stopped at a limit\n- Fixed a background agent running on when it was stopped while it was being resumed\n- Fixed background shell commands using memory for the earlier conversation until they exit\n- Fixed Remote Control using memory for the earlier conversation after a compaction or `/clear` in sessions run by Claude Desktop, IDEs or the SDK\n- Fixed Stop not ending a request in Claude Desktop sessions on Bedrock while it waits on a hung AWS credential_process or sign-in call\n- Fixed Claude sometimes failing to send you a file on its first try on Opus 4.8 and Opus 5\n- Fixed Claude not being told that a connected MCP server's tools could not be loaded, and taking it for a server with no tools, in sessions with tool search\n- Fixed `claude mcp login` and `/mcp` saying a service doesn't accept sign-ins from Claude Code when the service's client registration had been retired (HTTP 410 Gone)\n- Fixed another case of claude.ai-synced plugin hooks failing with \"Plugin directory does not exist\" while a session was still running on macOS\n- Fixed auto mode asking for manual approval to `rm` a symlink that points outside the working directory or at itself\n- Fixed `/sandbox` on Linux listing `allowUnixSockets` entries as allowed: they don't open those sockets there, and `/sandbox` now says so\n- Fixed pressing Esc twice to clear a mistyped code ending the sign-in in `/install-github-app`\n- Remote Control: Fixed sessions going back to their first effort level after a restart while the app still showed the level you picked (not yet for sessions that start in Ultracode)\n- Fixed a removed plugin's MCP server staying connected in headless sessions after `reload_plugins` or when the account's plugin list drops it\n- Fixed the Claude Docs connector's tools going missing when a plugin or MCP server at the same address failed to connect, and its error wrongly suggesting `/login`\n- Fixed a new skills, commands or agents folder being missed when `--add-dir` names the working directory or Claude Code starts in the home folder\n- Fixed sessions started from a pre-warmed process ignoring the `autoCompactWindow` setting in the project's `.claude/settings.json`\n- Fixed `CLAUDE_AUTOCOMPACT_PCT_OVERRIDE` staying in effect in a running session after it is removed from settings.json\n- Fixed the Agent tool's `effort` parameter being ignored for named teammates in agent teams\n- Fixed `/auto-mode-setup` ignoring recent auto mode denials when it scans a project for rules to suggest; it may now offer allow rules from them, and you review each before it's saved\n- Fixed a false \"Couldn't save this change\" warning after adding a permission rule or directory in a subfolder or worktree whose own `.claude/settings.local.json` is unreadable or malformed\n- Fixed cloud sessions telling Claude to run a network proxy check after a failed request, which auto mode then blocked\n- Fixed headless rewind and fork going ahead while a message sent right after a model change still waited on it\n- Fixed Claude not being told when a background task was stopped while a cloud session was restarting\n- Fixed `/team-onboarding` reporting a guide as shared on Team and Enterprise plans when the server refused the upload\n- Fixed panels and questions being cut off at the prompt in fullscreen, `/help`, history search and the background tasks list losing rows or titles in short windows, and tables wrapping in dialogs\n- Fixed resumed sessions showing \"Rejected\" or \"Declined\" for edits, questions and plans that were cut off when the session ended\n- Fixed `/passes` showing the wrong number of passes after a server error, and offering usage credits for sharing when no guest passes are left\n- Fixed `claude --help` output being cut short when piped to another program\n- Fixed `/terminal-setup` turning on \"Use Option as Meta key\" in Terminal.app on macOS 27, and tips there suggesting Option+Enter instead of Shift+Enter\n- Self-hosted runner: Fixed setup failing on Git LFS repositories cloned through Anthropic-managed git; sessions now start on pointer files, and `git lfs pull` fetches the content\n- Fixed sessions on self-hosted environments not starting from a git remote with an SSH host alias, GitHub SSH certificates or a GHE.com SSH user; an unresolved alias's error gives two git commands\n- Fixed remote MCP servers in self-hosted runner sessions staying without tools, or (over https) disconnected, after a timeout or network error\n- Fixed self-hosted runner sessions being shown as having stopped responding while Claude Code was still starting\n- Fixed Code Mode keeping in memory every large text file that code read, and memory growing each time a background subagent was woken\n- Fixed Code Mode stopping a script at its ten-minute limit while an MCP form, an MCP link prompt or a prompt that always asks was waiting for an answer\n- Fixed Code Mode scripts receiving a \"moved to the background\" message as data from most MCP calls over two minutes: the script now waits up to ten minutes, then gets an error\n- Fixed mods that list only some tools silently turning Code Mode off; such a mod no longer keeps it off, so use managed settings for that\n- Fixed Code Mode ignoring an MCP tool's request to end the turn when the tool was called from a script that Claude waits for\n- Fixed a Code Mode script's `let x;` keeping the value that an earlier script gave `x`\n- Fixed a Code Mode script's call with no argument failing with a missing-parameter error on tools that need none\n- Fixed Code Mode scripts losing access to tools through `tools` in sessions with more than 4,096 tools\n- Fixed the `/code-review` findings card often showing \"high\" for a review that ran at medium, on most models\n- Fixed key hints in permission prompts, questions, `/model` and `/rewind` ignoring `keybindings.json`\n- Fixed auto-scroll, the PR status badge, the auto-compact warning and the ultracode tag not following the `/config` switches that control them in a running session\n- Fixed a custom cancel key not working in `/add-dir`, `/tui` feedback and the save-workflow form\n- Fixed arrow keys in the save-workflow form moving the selection in the workflow view behind it\n- Fixed `/config` saying nothing, or \"dismissed\", after changing many settings\n- Fixed Read and @-mentions showing one line too many for files that end in a newline\n- Fixed `/status` not naming the model that opusplan and haiku use in plan mode\n- Fixed `claude doctor` advising a Keychain unlock inside a macOS sandbox, where unlocking can't help\n- Fixed the \"Jump to bottom\" pill staying hidden after copying text and scrolling away, the row under it going blank when its label changed, and quick PageUp presses scrolling too far in fullscreen\n- Fixed a screen update sometimes being skipped when memory is low\n- Fixed long code blocks losing a blank line and repeating another in scrollback\n- Fixed prompt suggestions sometimes showing placeholder text such as `_(no suggestion)_`\n- Fixed the line about earlier messages on resume blaming a damaged file when the file was intact\n- Fixed characters disappearing from the prompt when its text held a color code that sets several styles and a keyword or command was highlighted\n- Fixed `/rename` and `UserPromptSubmit` hook titles in `claude remote-control` sessions not showing on claude.ai and mobile\n- Fixed the Remote Control diff listing a file twice when the branch renamed it and it was then edited\n- Improved the automatic clean-up of old sessions on computers with managed settings\n- Improved `/tui` and other in-place restarts to wait briefly for a login refresh, so they are less likely to sign you out\n- Improved memory use in long sessions that edit many files, and in long Claude Desktop sessions on macOS and Linux with many large MCP tool results\n- Windows: Improved start-up and the first search on PCs where Claude Code's search program starts slowly\n- Improved auto mode's handling of a short reply such as \"yes\" to Claude's question when telemetry is off, behind a gateway to the Anthropic API, or on Microsoft Foundry\n- Improved memory use of idle Claude Desktop and SDK sessions after a long turn, and after `--resume` or `--continue` of a session that holds screenshots or large tool results\n- Improved how fast a Claude Desktop Code session with Remote Control on is ready for the next message after an answer\n- Improved cloud sessions to retry plugins that failed to download at startup\n- Improved Code Mode: the terminal freezes for far less time after a script makes thousands of tool calls\n- Improved Code Mode: after starting a subagent, Claude now gets the same guidance on waiting as outside Code Mode\n- Updated computer use on macOS to newer safety rules, so some actions are refused and some apps have less access\n- Changed saved scheduled tasks: in new signed-in sessions, most saved by earlier versions need scheduling again, though `--resume` of their own session still runs them\n- Changed `/code-review` with no target on most models to review the whole branch against the default branch, and fixed it losing the diff on multi-commit branches\n- Changed auto mode to run `git status` less often so most requests wait less in large repositories; a git command that discards changes may be blocked until the working tree is checked\n- Changed `/code-review <pr number> --fix`, and `/code-review` given a pull request URL, to stop without reviewing when `gh pr diff` is refused or fails\n- Changed 'Exit and stop tasks' to end this session's scheduled tasks so they don't run again on resume, unless you exit while Claude is responding\n- Changed some teammate panes to start through /bin/sh, so they may not read shell start-up files\n- Changed computer use on macOS to give up on an app launch that has not finished after 30 seconds\n- Gateway: Changed the start-up warning shown while `access_control.allow_cidrs` is empty to a notice at every start, boxed in text logs: never expose the gateway to the public internet\n- Self-hosted runner: Changed a second host name's clone of the same owner/name to go under `_repos/clones/<host>/` in the base folder; no clone moves\n- Self-hosted runner: Changed repositories with unusual characters in their address path to clone anew in a new folder; rules and volumes naming the old one stop applying, and the old clone stays put\n- Self-hosted runner: Changed `--repository-folder` to take most folders in the base folder at `--capacity 1` (above 1, one there stops start-up); another repository's session overlapping one is refused\n- Changed Code Mode so a PreToolUse hook that errors no longer refuses a script's tool call, as outside Code Mode; set `\"onFailure\": \"block\"` on a command, HTTP or MCP tool hook that must refuse\n- Changed Code Mode to refuse a script's read of a line range over 256 MiB, which could use gigabytes of memory, and fixed reads of very long lines timing out\n- Changed Code Mode to also start when telemetry, nonessential traffic or feature flags are turned off; an administrator can keep it off in managed settings\n- Changed Code Mode to run a script that declares its own `log`, `preview` or `sleep`, which it used to refuse\n- Changed macOS login renewal to test the Keychain with a `Claude Code-write-check` item\n- Changed the `/tasks` header and the footer's background count to one number covering shells, monitors, MCP calls, agents, workflows and scheduled tasks\n- Changed Stop and SubagentStop hooks to list a Monitor's script as `monitor`, not `shell`, in `background_tasks`\n- Removed 2.1.326's automatic retry at the limit named in a `max_tokens` error; behind a gateway whose model takes fewer output tokens, set `CLAUDE_CODE_MAX_OUTPUT_TOKENS` to that limit\n- [VSCode] Added a line above the message box after you decline the trust prompt, with Review trust to bring the prompt back without reloading the window\n- [VSCode] Fixed being signed out of Claude Code in every window and terminal when the sign-in server took 10 to 30 seconds to answer the extension's token refresh\n- [VSCode] Fixed keys pressed while an input method (IME) is composing moving focus on or declining a permission prompt, running menu rows and closing dialogs (anthropics/claude-code#106358)\n- [VSCode] Fixed the permission prompt for Bash, Edit, Write and other built-in tools not showing why a safety check asks\n- [VSCode] Fixed a half-written plan comment being lost or moved to another passage, and an input method's Enter sending the comment, ending a session rename or picking an @-mentioned file\n- [VSCode] Fixed the limit banner not saying when usage credits are in use, and the session survey covering that banner and the usage credits card\n- [VSCode] Fixed screen readers not announcing a permission request that arrived while the Account & Usage dialog was open until that dialog was closed\n- [VSCode] Fixed a proposed-edit diff tab staying open, with an Accept button that did nothing, after its terminal Claude Code session disconnected\n- [VSCode] Fixed short tool inputs and outputs cut off at the panel's edge with no way to read the rest (they now open in an editor tab), and dragging to select text also opening one\n- [VSCode] Fixed the Up key putting an earlier message over text that an input method (IME) is composing in the message box (anthropics/claude-code#101294)\n- [VSCode] Fixed dropping an Explorer folder onto the chat inserting nothing on VS Code 1.103 to 1.139 or in a floating chat window, and the drop highlight vanishing with a false \"Hold Shift\" hint\n- [VSCode] Fixed a permission prompt moving keyboard focus away from the details you were scrolling as its buttons switched on\n- [VSCode] Fixed file links in a Claude tab opening in another editor group when `claudeCode.lockEditorGroups` is off (anthropics/claude-code#108884)\n- [VSCode] Fixed ranges written with a single tilde (`3~5 … 7~10`) losing their tildes and showing struck-through text between them; only `~~text~~` is struck through now\n- [VSCode] Fixed dictation errors wrapping into a narrow column, the spinner turning while a question card waits, and a NUL symbol after each letter in the output of a failed UTF-16 command\n- [VSCode] Changed Down to stop at the last item of a permission prompt instead of wrapping to Yes\n- [VSCode] Changed command output to show hidden characters and control sequences other than plain colors as symbols, so Persian, Khmer, Lao, Sinhala and Malayalam text can look broken or out of order\n","publishedAt":"2026-11-19T22:14:41.000Z","fetchedAt":"2026-11-19T23:11:28.697Z","url":"https://code.claude.com/docs/en/changelog#2-1-329","media":[],"coverageCount":0},{"id":"rel_uZblYXtQyNSql-TpzodGj","version":"v2.1.328","type":"feature","title":"v2.1.328","summary":"Interactive sessions with no permission mode now start in auto mode for Opus 4.6 and Sonnet 4.6 on Bedrock, Google Cloud's Agent Platform, Microsoft Foundry and the Claude apps gateway. MCP tool hooks gain \"onFailure\": \"block\", and a scheduled task save no longer erases a project's other saved tasks when the tasks file can't be read.","titleGenerated":"Claude Code 2.1.328 adds Opus 4.6 auto mode on Bedrock and Foundry","titleShort":"Interactive sessions start in auto mode on cloud providers; MCP tool hooks can block","breaking":"minor","importance":4,"content":"## What's changed\n\n- Added auto mode for Claude Opus 4.6 and Sonnet 4.6 on Amazon Bedrock, Google Cloud's Agent Platform, Microsoft Foundry and Claude apps gateway; interactive sessions with no permission mode start in it\n- Added `\"onFailure\": \"block\"` for `mcp_tool` hooks, as for command and HTTP hooks: a hook whose call fails blocks the action\n- Added a warning when a message from a session that skips permission prompts may be held and dropped by a cloud session\n- Added a notice in `/status` and `claude doctor` when an OpenTelemetry exporter has no endpoint and its data would likely be lost\n- Added `CLAUDE_CODE_DISABLE_DEFERRED_TOOL_RESUME`, which lets a host stop `claude -p --resume` and `--continue` from running a tool call that a hook deferred\n- Added support for apps that embed Claude Code to switch on the policy plugin's `allowManagedModsOnly` option; an admin's managed settings still win\n- Added a startup warning to the Claude apps gateway when a policy's `maxEffortLevel` is not a valid effort level\n- Self-hosted runner: Added `--repository-folder` to name the folder that sessions use for a repository at `--capacity 1`, in place of its path under the base folder, which can't be a symbolic link\n- Self-hosted runner: Added a start-up warning when the host's settings.json sets a default permission mode that sessions ignore\n- Self-hosted runner: Added `--hook-label` to `claude self-hosted-runner orchestrator`, to hand chosen session labels to the `spawn-runner` hook\n- Self-hosted runner: Added `--metadata key=value` to `claude --environment`, to label a session for the `spawn-runner` hook\n- Self-hosted runner: Added `claude_code_self_hosted_orchestrator_session_queue_wait_by_reason_seconds` to the orchestrator's `/metrics`, so a re-opened session's wait can be told from a new one's\n- Self-hosted runner: Added an orchestrator metric, `claude_code_self_hosted_orchestrator_sessions_given_up_total`, that counts by cause how often Anthropic gave up starting a runner for a session\n- Fixed Bash commands in zsh stopping with a \"not found\" error at a word that starts with `=`, such as `echo ====` or `[ a == b ]`; `=cmd` no longer expands to a command's path\n- Fixed automatic compaction reusing the session's earlier system prompt instead of rebuilding it the way `/compact` does\n- Fixed markdown rendering freezing on some very long or malformed input\n- Fixed rare screen freezes\n- Fixed Esc often being lost at a permission prompt or question when you typed straight after it; in terminals that send Alt as Esc and a key, an unused Alt or Option key now declines there\n- Fixed a paste or a terminal's reply being typed into the prompt when a slow connection cut it in two\n- Fixed saving a scheduled task erasing the project's other saved tasks when the scheduled tasks file can't be read, on third-party providers or with telemetry or feature flags turned off\n- Fixed a scheduled task with a typo in the project's scheduled tasks file being deleted the next time that file was saved\n- Fixed a case where a resumed cloud session could lose its compaction summary or recent messages when a message was saved late\n- Fixed the token counts on the spinner and for subagents counting a retried reply twice, missing replies that were not streamed, and standing still while Claude thinks\n- Fixed the error shown when usage credits run out mid-response, which named a rate limit\n- Fixed auto mode failing every turn outside Claude Desktop with `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS` set behind a gateway rejecting betas, on Bedrock, Agent Platform, Foundry or Claude apps gateway\n- Fixed `CLAUDE_STREAM_IDLE_TIMEOUT_MS` not lengthening the 5-minute connection limit on gateways, custom base URLs and cloud providers\n- Fixed some Claude Desktop sessions on third-party providers, other than scheduled tasks, ending for good with \"Prompt is too long\" after large tool results; the largest are now left out\n- Fixed MCP tools whose input schema starts with `anyOf`, `oneOf` or `allOf` going missing on third-party providers or with telemetry or feature flags turned off\n- Fixed an MCP tool with an invalid input schema failing requests on third-party providers or with telemetry or feature flags turned off; the tool is now left out and Claude is told why\n- Fixed effort being turned off for a subagent's model for the session if Bedrock, Google Cloud's Agent Platform or Foundry refused its progress summary's effort setting (regression in 2.1.327)\n- Fixed Code Mode stopping a script when a permission prompt outside auto or bypass mode, a plan approval or a question stayed open for ten minutes\n- Fixed Code Mode scripts in auto and bypass-permissions modes failing to call an MCP server that had connected while the previous script was running\n- Fixed Code Mode scripts not running in the background in auto or bypass permissions mode in a session chosen in the `claude --resume` picker\n- Fixed Code Mode losing a loaded skill, hook context or a note typed with a permission approval when resuming a session in which one reply ran several scripts\n- Fixed Claude not being told that a Code Mode script was cut off when the session was resumed from the picker or in a mode that asks for permission\n- Fixed Code Mode silently using `undefined` for top-level variables that an earlier failed script never set\n- Fixed Code Mode scripts not being told when a failing command's long output was cut\n- Fixed Code Mode refusing valid scripts that start with a `{` block\n- Fixed Code Mode refusing scripts that use both a top-level `await` and a top-level `return`\n- Fixed an @-mentioned small file not being attached in Code Mode after a script had read or written it\n- Fixed Code Mode offering `write()` and `edit()` to read-only subagents that can't call them\n- Fixed Claude starting a turn by itself in Code Mode when Esc was followed by a command such as `/model`\n- Fixed Claude in Code Mode not being told to run workflows only when you ask, or the workflow size you set in `/config`\n- Fixed Remote Control repeating the same disconnect message, and telling you to retry a session that can't reconnect\n- Fixed a waiting permission prompt being lost, and its answer ignored, when a remote session restarted while the service was slow\n- Fixed cloud sessions resuming a cleared conversation after a rewind followed by `/clear`\n- Fixed a remote session losing a turn's tool calls and results when its runner or worker was lost after a dropped model stream had been retried\n- Fixed cloud and Remote Control sessions that just started, or have an @ in their title, being unreachable in cross-session messaging from a session with Remote Control on\n- Fixed Claude being told that cloud sessions can never reply; it now learns when a reply can come and how else to get the result\n- Fixed `/diff`, the Changes pane of Remote Control sessions and the diff beside an edit leaving out committed work after a command moved into another repository\n- Fixed Remote Control not showing a running workflow's agents when a session run by an IDE extension or other SDK app is opened over an hour after the workflow last made progress\n- Remote Control: Fixed an idle Desktop or VS Code session showing as running after Remote Control is turned on\n- Fixed the model saying background agents were still running after an SDK interrupt had stopped them\n- Fixed Send now in the desktop app and SDK hosts telling Claude that a stopped web search found no results\n- Fixed Cmd+letter being typed into the prompt in terminals that report Cmd\n- Fixed Ctrl+click opening a link twice in GNOME Terminal, side mouse buttons acting as clicks, and the first wheel notch in the other direction being ignored after a pause, in fullscreen mode\n- Fixed Shift+Enter in Konsole outside tmux and ssh, and Alt shortcuts on non-Latin keyboard layouts in terminals that speak the kitty keyboard protocol\n- Linux: Fixed pasting in other apps freezing while Claude Code is suspended with Ctrl+Z, when `xclip` or `wl-copy` is installed\n- Linux: Fixed Ctrl+V missing the clipboard image, pasting an old one or hanging\n- Fixed text copied in fullscreen mode including the quote bar, losing spaces at line wraps when a drag scrolled upward, or missing words under the \"Jump to bottom\" pill\n- Fixed an image path pasted into a `!` command becoming an attachment\n- Fixed Ctrl+C showing no exit warning while the `?` shortcuts list is open\n- Fixed Ctrl+R search in fullscreen mode and the `/theme` color editor picking the wrong row when keys arrive together, and keys being dropped there and in the search of `/config` and `/plugin`\n- Fixed wrong or missing colors inside GNU screen\n- Fixed many background notifications cutting into the top border of open dialogs in fullscreen mode\n- Fixed a message to a background agent showing a long file path in place of a pasted image\n- Fixed `/goal` removing a skill's Stop hook, or ending early on another hook's result, when both had the same wording\n- Fixed `/bug` saying cancelled after the report was sent\n- Fixed Esc not dismissing the transcript share question for Pro and Max users\n- Fixed `/hooks` losing your place when going back\n- Fixed `/export` writing a file named `.txt` when the name was left empty\n- Fixed `/terminal-setup` in Zed replacing a symlinked `keymap.json` with a plain file\n- Fixed `/install-github-app` adding a code review workflow when asked to update the existing workflow file\n- Fixed `/plugin` reopening a skill, or closing the open page, when a slow save or removal finished after you went back\n- Fixed `/plugin` asking to uninstall the wrong plugin, or ignoring Esc, after leaving a plugin while a change was still running\n- Fixed typing and arrow keys not working on the `/plugin` Installed tab after a search with no results or a click on the tab row\n- Fixed holding Enter in `/plugin` turning off several installed plugins\n- Fixed the `/usage` plugin total stopping at 2,000 tokens on 1M context models, and low token estimates in `/usage` and `/plugin` when no model is picked\n- Fixed Esc on the Bedrock access keys step closing `/login` instead of going back\n- Fixed `/exit` and Ctrl+D describing running work differently, and `/exit` warning about a workflow that had stopped\n- Fixed Claude calling an artifact private after reading one in a shared folder or publishing to another organization's, and saying only named people can open one when its folder's people can too\n- Fixed a plugin's toast sometimes showing another plugin's text\n- Fixed a mod saved just after a turn began sometimes reloading mid-turn instead of at the turn's end\n- Fixed one malformed `OTEL_RESOURCE_ATTRIBUTES` entry dropping the variable's other labels from metric points, events, and spans. Now only that entry is left off\n- Fixed agent hooks running past their timeout while compacting after a request was refused as too large\n- Fixed a crash at startup when `FORCE_COLOR` is set to a negative number\n- Fixed a crash in secret masking when the machine is almost out of memory\n- Fixed the Bash sandbox on Linux renaming `.git/HEAD` after each command when the branch name is not valid UTF-8\n- Self-hosted runner: Fixed all running sessions being stopped when a proxy or web filter answered a poll with HTTP 200 and a body such as `{}`; the runner now retries such a poll\n- Self-hosted runner: Fixed deletion of a kept clone's objects or refs, and of commits that existed only there, after a temporary disk error; that start now fails instead\n- Fixed self-hosted runners stopping when their token ran out on a host whose clock is slow or was set back\n- Self-hosted runner: Fixed sessions with Anthropic-managed git losing git access to GitHub after a script or hook in the session ran `claude` again\n- Fixed self-hosted runners with Anthropic-managed git being unable to fetch or push where the machine's git config sets `http.proactiveAuth`\n- Self-hosted runner: Fixed a runner that exited without telling the server when setup failed after it had registered; it now deregisters with a reason, and an on-demand runner reports the failed start\n- Self-hosted runner: Fixed archiving or requeueing a session not taking effect while it waited for another session to prepare the same repository (`--capacity` above 1)\n- Self-hosted runner: Fixed a session with a damaged `.claude.json` failing every later start under `--trust-workspace false`; it now starts on defaults, with MCP servers turned off there on again\n- Self-hosted runner: Fixed a failed start's message losing the words that say what failed when a checkout hook or git printed a lot\n- Fixed the self-hosted runner skipping its post-session hook when a disk stalled for a few seconds; on a dead disk a session's end can take 25 seconds longer by default\n- Fixed self-hosted runners: a session that resumes in a new Claude config directory no longer loses its task list\n- Fixed Claude not being told when a self-hosted session starts without one of its repositories, and why\n- Self-hosted runner: Fixed `--correlation-id` of `claude --environment` not reaching the `spawn-runner` hook\n- Fixed the self-hosted runner setup wizard's key rotation advice, which now warns that revoking a key stops the sessions running under it\n- Improved `claude -p` and SDK sessions to fetch managed-settings plugins that have their own git repository, in the background, for a later session\n- Improved how soon managed-settings plugins from a git marketplace load on a fresh machine in `claude -p`, SDK and Claude Desktop sessions; a short run may exit up to 30 seconds later\n- Improved the time to resume large sessions\n- Improved memory use when resuming a session with many images\n- Improved memory use and response time in long SDK and desktop sessions that run slow commands or tools\n- Improved Code Mode's speed in auto and bypass-permissions modes by removing a 0.75 second pause after most scripts\n- Improved prompt caching for new Code Mode sessions that have MCP servers, and Claude is now told when a server is still connecting and when the list of servers changes\n- Improved Code Mode's token use with shorter tool descriptions\n- Improved Code Mode to show the command a script is running, and its output, while it runs\n- Improved Code Mode edits: an edit that changes nothing no longer stops a script, and Claude is told where a failed edit matches or comes closest\n- Improved Code Mode: Claude is now told which tool calls were still running when a script failed\n- Improved how Claude starts dev servers in Code Mode's auto and bypass modes: Claude is told to run them as background tasks that show in `/tasks`\n- Windows: Improved Code Mode by telling Claude that scripts' shell commands run in Git Bash and that PowerShell aliases run instead of functions with the same name\n- Improved starting several in-process teammates at once: far fewer now fail with \"Lock file is already being held\"\n- Improved the computer use approval dialog: it now shows each app's level of access and plainer warnings\n- Improved cross-session messaging: Claude now explains that Remote Control is off when a cloud session or another computer is not listed\n- Improved `claude -p` text output to say on stderr when a message from another session is held or dropped, and how to allow it\n- Improved Claude's answers on who can open an artifact it published: Claude is now told to check the artifact again instead of repeating the last publish's answer\n- Improved resumed self-hosted runner sessions: Claude now checks whether its earlier work is still there, and says what is missing\n- Changed plan mode and Bash prompts for Opus 4.6 and Sonnet 4.6 on third-party providers to use and offer auto mode, as on newer models; `disableAutoMode` turns both off\n- Changed Code Mode to no longer run UserPromptSubmit hooks, managed ones included, on scripts' output; to redact that output, use a PostToolUse hook with `updatedToolOutput`\n- Changed `/code-review` on most models to report only bugs and CLAUDE.md violations; run `/simplify` for reuse, simplification and efficiency cleanups\n- Changed `/code-review` to mark each finding Important, Pre-existing or Nit on most models, and to end each summary with a suggested fix; a clean review's list now reads \"Code review found no issues.\"\n- Changed `deniedMcpServers` entries of the form `claude.ai <name>` to apply to that connector in cloud and self-hosted runner sessions too\n- Changed managed `PostToolUse` hooks to no longer fire for calls refused before they ran: no log entry, rewrite, block or stop there; `PostToolBatch` gets most, not marked as refused\n- Changed `/bug` to ask Yes or No before sending, so repeated Enters no longer send a report or open GitHub\n- Changed subagents in `claude -p`, the Agent SDK, desktop and VS Code to wait for their own subagents by default when `CLAUDE_CODE_FORK_SUBAGENT=0` is set\n- Changed `claude -p` with `CLAUDE_CODE_FORK_SUBAGENT=0` to count a background subagent and those it waits for against one 10-minute wait; raise `CLAUDE_CODE_PRINT_BG_WAIT_CEILING_MS` for longer chains\n- Changed a shell command rewritten by a PermissionRequest hook to a sandbox-excluded command to run outside the sandbox\n- Changed the Claude in Chrome welcome screen and `/chrome` to say that Claude Code asks before Claude acts on a site, unless your permission mode or a rule in `/permissions` answers\n- Changed Claude Desktop and Agent SDK sessions on third-party providers or with telemetry or feature flags off to end the turn when closed with a question open, instead of continuing without an answer\n- Changed pasted text to be marked for Claude as pasted, on third-party providers or with telemetry or feature flags turned off\n- Changed subagent progress summaries in terminal sessions on the Anthropic API to be written by the default small fast model from short notes, also with telemetry off, using far fewer tokens\n- Changed how a sandbox that can't start is reported, unless `sandbox.failIfUnavailable` is set: the next sandboxed Bash command is refused and says that the sandbox is off for the session\n- Changed `/list-agents` to show its list to you alone; Claude lists sessions itself when needed\n- Self-hosted runner: Changed the built-in `gh api` to be off by default where settings guard Claude's GitHub tools by name; `--use-anthropic-gh true` turns it on\n- Self-hosted runner: Changed a `claude` that a wrapper script runs before the session's own `claude` is up, or after it has ended, to have no Anthropic-managed git\n- Self-hosted runner: Changed `CLAUDE_RUNNER_SESSION_UUID` and `CLAUDE_CODE_REMOTE_SESSION_UUID` to both be set for the session and its `checkout`, `post-session` and `spawn-runner` hooks\n- Changed self-hosted runners at `--capacity 1` to refuse a session earlier when the path of a repository it pushes to is a symbolic link\n- [VSCode] Added a notice when a message from another session is held, with a link to the settings file that can allow it\n- [VSCode] Fixed the permission prompt cutting off its No button, reply field or title when the panel is low or a command, URL or description is long\n- [VSCode] Fixed failed or blocked tool calls showing no reason or a reason cut short, and Glob and Grep counting an error as results\n","publishedAt":"2026-11-19T01:14:04.000Z","fetchedAt":"2026-11-19T02:57:48.747Z","url":"https://code.claude.com/docs/en/changelog#2-1-328","media":[],"coverageCount":0},{"id":"rel_4PcjlXjhEcduVbZI3ysCk","version":"v2.1.327","type":"feature","title":"v2.1.327","summary":"Stopping the Remote Control server in worktree mode no longer deletes running sessions' git-ignored files such as .env, and project-scope claude mcp add and remove no longer strip other top-level keys from .mcp.json. The release also adds a stdin form of claude mcp add-json that keeps tokens off the command line and a /status list of broken agents, skills, commands, keybinding errors, and hook matchers that never match.","titleGenerated":"Claude Code 2.1.327 stops worktree cleanup from deleting ignored files","titleShort":"Worktree stop no longer deletes .env; .mcp.json keys preserved","breaking":"minor","importance":4,"content":"## What's changed\n\n- Added a confirmation before `/install-github-app` changes a repository\n- Added `claude mcp add-json <name> -` to read the server's JSON from stdin, so tokens can stay off the command line; not available inside a Claude Code session\n- Added a line when a resumed session's earlier messages can't be loaded\n- Added a list in `/status` of broken agent, skill, command and output style files, keybinding errors, an unavailable sandbox, ignored permission settings and hook matchers that never match\n- Added a note in `/config` when other settings override a row\n- Added `x-claude-code-managed-settings-checksum` to the gateway hint headers; sent where managed settings set `CLAUDE_CODE_GATEWAY_HINT_HEADERS=1`, including where they already do\n- Added support for naming a session's billing group with the `attributionProject` setting, for Claude apps gateways that set `admin.billing_group_tag`\n- Added `/tui classic` as the name for the non-fullscreen renderer; `/tui default` still works\n- Remote Control: Fixed stopping the server in worktree mode deleting a running session's git-ignored files, such as `.env`\n- Fixed self-hosted runners with `--use-anthropic-git-proxy` deleting a non-empty directory named by `GIT_CONFIG_GLOBAL` at startup: the runner now keeps it, ignores the variable and warns\n- Fixed `claude mcp add` and `claude mcp remove` with project scope deleting other top-level keys from `.mcp.json`, such as `$schema` or another tool's `servers`\n- Fixed `/export` overwriting an existing file without asking: the dialog now asks, and `/export <name>` keeps the file and says to run `/export`\n- Fixed a command you had allowed still running after you pressed Stop, when the approval arrived after a restart, in cloud and self-hosted runner sessions with no feature flags loaded\n- Fixed Enter closing `/teleport`'s load error screen\n- Fixed a click followed by Enter in `/plugin` installing or disabling a plugin\n- Fixed a half-typed `!` command running on an extra Enter after the usage limit menu\n- Fixed some `--disallowedTools` rules with parentheses in their content\n- Fixed auto mode ignoring some wildcard PowerShell allow rules that name a script with `.\\`, such as `PowerShell(.\\scripts\\build.ps1 *)`\n- Fixed a running Code Mode script in auto mode getting permission prompts and denials once Claude started a second script that auto mode also allowed\n- Changed some pre-started background sessions to be discarded after an account change\n- Changed a workflow paused at a usage limit in the terminal to stay paused after a login, until the earlier login returns or you type a prompt\n- Fixed a nested `claude` with its own token being refused by `forceLoginOrgUUID`: commands that don't get the session's token no longer get its `CLAUDE_CODE_OAUTH_SCOPES`; set it on any that need it\n- Fixed headless sessions losing stored messages on resume when a message was sent again with the same UUID after the process was killed\n- Fixed cases where a session resumed after an empty reply or a killed command lost an answer, or brought back a rewound prompt, after another crash or a move to the background\n- Fixed a remote session forgetting messages sent after Stop was pressed during a restart, if Claude Code then crashed\n- Fixed tasks created after `/clear` being lost, and cleared tasks coming back, when a cloud or self-hosted runner session with no feature flags loaded resumed later\n- Fixed a compaction right after a reply cut at the output token limit losing that reply's tool call and its result\n- Fixed a settings change from `/config` and similar commands failing to save with no warning; a line under the prompt now says so\n- Fixed `/rewind` closing without a word when some files could not be restored\n- Fixed pasted images being dropped when the prompt starts with a file path such as `/tmp/a.txt`\n- Fixed agents passed with `--agents` being lost once plugins loaded or reloaded\n- Fixed a plugin's agent being dropped when resuming a session from the `--resume` picker\n- Fixed Claude Code refusing to start when `https_proxy` or `HTTPS_PROXY` is a bare `host:port` and neither its input nor its output is a terminal (the desktop app, an SDK, cron)\n- Fixed Claude Code on Linux sometimes returning to the shell while restarting itself at startup, such as after the first sign-in to a Claude apps gateway or when turning on Code Mode\n- Fixed `claude -p` hanging before its first request when the fetch of a model's settings stalled, such as at a proxy; it now goes on without them after 10 seconds\n- Fixed the screen staying blank or frozen after a brief out-of-memory error\n- Fixed a rare display fault that could freeze the screen or end the session\n- Fixed cloud sessions breaking when a program left thousands of connections open to the session's network proxy\n- Fixed a turn hanging when a mod's `$.model.complete` or `$.model.classify` call met a usage limit: the call now gets the API's error at once; mods should handle that error\n- Fixed stalls on a compaction reply with thousands of unclosed tags and on some very long WebFetch URLs; WebFetch now refuses a URL over 2,000 characters before any permission prompt\n- Fixed a message with thousands of `@` mentions stalling before Claude replies; only the first 100 mentions are now looked up as files\n- Fixed the prompt freezing after rejecting a write to a very large file; a rejected write now shows a short preview\n- Fixed a crash when a burst of keys reached the search box in `/plugin`\n- Fixed a false crash report and exit code when a suspended session resumes, or at exit, where a Linux security policy blocks a process from signalling itself\n- Fixed remote MCP servers staying disconnected after a longer outage such as a VPN drop: most servers on other machines reached by hostname over https are now retried in the background for about a day\n- Fixed Edit refusing with \"File has not been read yet\" after a compaction on unchanged files outside the project that Claude had already read in full\n- Fixed Grep repeating or skipping results when paging with `head_limit` and `offset`; content and count results are now sorted by file path\n- Fixed Read failing on Jupyter notebooks that contain an output of an unknown type\n- Fixed NotebookEdit removing the notebook file's final newline\n- Fixed the Read tool reporting 512KB instead of the real size when a file of 10 MB or more is too large to read at once\n- Fixed the Bash tool's diff of changed files not naming files that are binary, too large or too many to show\n- Fixed artifact publishes failing with \"the source file changed between approval and publish\" when a synced or remote-backed folder had only changed the file's timestamp\n- Fixed a new artifact that you approved being refused in a cloud scheduled task's session, in organizations that require approval of each new artifact\n- Fixed a new artifact being refused after an app built on the Agent SDK approved it, in organizations that require approval of each new artifact\n- Fixed `/artifacts` showing part of your list, or \"No artifacts yet\", when a request for more of the list failed: it now tries again, then warns if it can't load the whole list\n- Fixed in-process agent teammates on a different model from the lead compacting at the lead's context limit and on the wrong model; they now use their own limit and model, at that model's rate\n- Fixed a finished subagent resuming on the model and effort of a sibling subagent that messaged it\n- Fixed the main conversation not being told a background agent had finished when the subagent that started it ended by itself without reading the result, outside auto mode\n- Fixed Claude reporting a background agent as still running when it had waited 30 seconds or more on a permission prompt in the desktop app, on the web and in the SDK\n- Fixed Claude not being told in auto mode when a background agent's reply to a follow-up message was dropped, and reduced repeat wake-ups there from agents waiting on background commands\n- Fixed the desktop app, Claude Code on the web and VS Code showing a raw task notification as a subagent's prompt after its background task woke it\n- Fixed messages sent from another session with \"later\" priority joining the running turn, instead of waiting for it to end, in Remote Control and Claude Desktop sessions\n- Fixed messages to cloud and Remote Control sessions being reported as failed with a stale address after a timeout or rate limit; Claude now says when a message may have arrived\n- Fixed cloud sessions giving no error when a `UserPromptSubmit` hook blocks a message sent while Claude works\n- Fixed failing hooks showing nothing: SubagentStop in background agents, SessionEnd at `/clear` and `/resume`, and CwdChanged and FileChanged in fullscreen\n- Fixed a Stop hook that failed to run being shown as having reported a problem\n- Fixed focus view keeping a finished reply hidden while a Stop hook was running\n- Fixed Claude getting a long managed PostToolUse hook output twice when a mod added context to the tool call\n- Fixed `/model` listing the wrong models in some terminal sessions when a project's `.claude/settings.json` sets `ANTHROPIC_BASE_URL` to an LLM gateway\n- Fixed `/statusline` writing to `~/.claude/settings.json` when `CLAUDE_CONFIG_DIR` points to another folder\n- Fixed `claudeMdExcludes` patterns on Windows not matching a path whose drive letter is written in the other case\n- Fixed `claude agents` and `claude rm` refusing to remove a fully pushed session (\"The worktree has unpushed commits\") when its worktree has a file or folder named `HEAD` at the top level\n- Fixed long ListAgents results hiding most sessions on accounts with many cloud or Remote Control sessions: long lists now come in pages\n- Fixed slow turns in resumed sessions for organizations with tens of thousands of connector tool permissions\n- Self-hosted runner: Fixed `claude -p` started by a session's script or hook waiting out rate limits, because it inherited the session's own retry setting\n- Fixed the spend limit menu reporting a changed limit when the server refused it\n- Fixed a second `x` on a task's details in `/tasks` stopping the next task too\n- Fixed Esc being ignored on a permission prompt that had just appeared\n- Fixed lists that load their rows late sometimes ignoring Enter and the arrow keys\n- Fixed `/autofix-pr` holding the keyboard after Esc while it stopped the cloud session; the dialog now closes at once, and a stop that fails is reported afterwards with a link\n- Fixed a blank screen and stray mouse codes in the shell after quitting with ctrl+c just after ← opens the agents view\n- Fixed Ctrl+C right after the trust question leaving the terminal in fullscreen\n- Fixed the shell flashing on screen when ← opens the agents view in fullscreen mode\n- Fixed terminal display faults: most scrambled scrollback in Ghostty on resize, link addresses hidden in Windows Terminal under old tmux and in VTE 0.50.0, and no selection highlight with 16 colors\n- Fixed some garbled rows in tmux with `allow-passthrough` on inside kitty, Ghostty or iTerm2\n- Fixed `/mcp` saying no servers were configured after one was added during a session or declined; it now says what to do\n- Fixed `/hooks` leaving out hooks given by `--settings`\n- Fixed the `/` menu listing a row for each skill of the same name, though every row ran the first\n- Fixed `/teleport` showing no message when the chosen session failed to resume\n- Fixed n and N in transcript search skipping matches\n- Fixed cancelling a transcript search (ctrl+o, then /) in a long session landing far from where the search started\n- Fixed up and down losing the cursor's column over a short or blank line\n- Fixed Tab in shell mode filling in a file name after `&&`, `;` or `|`, where a command goes\n- Fixed the selected session leaving the screen in `claude --resume` and `/resume` when branch names or folder paths are long\n- Fixed the MCP tool list in `/mcp` and `/plugin` losing your place when going back from a tool\n- Fixed `/resume` listing one session fewer than the window has room for, and a 'Resend your message' hint after most spend caps, where resending can't help\n- Fixed an MCP server's long question being cut off after one row, text fields hiding long pastes with no sign, and a doubled divider line in `/plugin`'s options and `/login`'s Bedrock and Vertex steps\n- Fixed the \"+N lines\" count being too high for command output with links or colors, and the transcript's `?` help hiding how to exit in a short window\n- Fixed Ctrl+C twice not closing `/chrome` from its browser list\n- Fixed a number and Enter typed too early at an agent's plan approval in screen reader mode causing the next number to be dropped too\n- Fixed Claude trying to read design system token files that do not exist when starting a deck or design\n- Fixed exit not saying that a worktree was kept\n- Fixed `/hooks` help naming input fields that do not exist\n- Fixed question dialogs naming ctrl+g after the external editor key was rebound\n- Fixed two sign-in error messages that pointed to the Claude subscription option on computers where `forceLoginMethod` is set to `console`\n- Fixed `/config` saying the dialog was dismissed instead of \"Set theme to …\" after changing the theme\n- Bug fixes and reliability improvements\n- Improved the speed of a session's first Bash command: shell start-up files may now run before it's approved, even if it's refused; set `CLAUDE_CODE_DISABLE_SHELL_HEAD_START=1` to turn this off\n- Improved commits on newer models and in PowerShell: Claude is now told not to commit files that likely contain secrets\n- Improved sandboxed shell commands: Claude now sees when a command exempt from the sandbox stayed in it because its permission check relied on the sandbox\n- Improved the refusal for shell commands with no `rm` that the dangerous `rm` check can't parse, so Claude can resend harmless ones in plain syntax instead of giving up\n- Improved resuming very long sessions: a transcript read that ends early is now retried once, so messages are lost less often\n- Improved subagent answers that hit the output token limit twice: they now more often come back whole or say up front that they are partial\n- Improved long replies on unrecognized models: a reply cut at 32,000 output tokens now continues with a limit of 128,000\n- Improved the reminder about a completed subagent after compaction: if Claude can't read the agent's output file, it is now told to ask the agent with `SendMessage`\n- Improved the warning shown when an API key and an Anthropic profile are both set: it now says which one is in use and how to switch to the profile\n- Improved `/plugin`: the question before removing a marketplace now says that its plugins' settings and saved data are deleted, and a marketplace's page no longer shows another marketplace's error\n- Windows: Improved start-up time with no Git Bash and no `pwsh` on PATH by checking PowerShell's standard install locations at once; System32's `powershell.exe` now wins over one earlier on PATH\n- Improved startup time on Windows when PATH points to Git for Windows' bin folder, by starting git.exe directly instead of through its launcher\n- Windows: Improved PowerShell tool speed by up to about half a second per command\n- Improved how soon the Bash tool's diff of changed files is ready in auto and bypass permissions modes\n- Improved speed of the Bash permission check on very deeply nested missing paths on macOS and Linux, and of secret masking on very long text with a URL\n- Improved transcript search to find text in edits and written files\n- Improved speed of reading large PNG images, and the quality of the JPEG sent in their place\n- Improved the Artifact tool to read an uploaded file by the address the artifact's page shows; other forms still fail, most with a message that says to read uploads one per call, by id\n- Improved the Artifact tool to ignore `prompt`, `limit` and `label` on some calls that do not use them instead of failing the call\n- Improved MCP server pages in `/mcp`: local, remote and agent servers share one layout, menus end in Back, and actions are no longer numbered outside screen reader mode; use the arrow keys and Enter\n- Improved narrow terminals: `/status`, forms and detail screens lay out better, `/model` and `/effort` fit at about 40 columns, and key hints such as `←/→ to adjust` stay\n- Improved `/install-github-app`: the repository, secret and API key steps use the standard list and keys, with shorter messages, a spinner and clearer errors\n- Improved `/desktop`: the download question is a standard list instead of y/n\n- Changed `ExitWorktree` to ask before removing a worktree that may hold unsaved work; refused in don't-ask mode or with nobody to answer, unless an allow rule, auto or bypass permissions mode allows it\n- Changed `claude mcp add --scope project` to stop, instead of overwriting the file, when `.mcp.json` lists its servers at the top level with no `mcpServers` key\n- Changed the arrow keys on most permission prompts to stop at No instead of wrapping around; a number past the last answer moves to No\n- Changed path checks to refuse some symbolic links\n- Changed what counts as inside a working directory, so some paths now ask for permission\n- Changed `permissions.blockReadsOutsideWorkingDirectories` to ask about more Bash commands in bypass and auto modes, and to deny some in auto mode\n- Changed `permissions.blockReadsOutsideWorkingDirectories` to refuse more Bash commands that name a path under a Read or Edit deny rule, where it used to ask\n- Changed auto mode's notice about falling back to billed classifier requests to say that an upcoming release removes the fallback; each action it would have checked will be asked about or denied\n- Changed auto mode to ignore more allow rules that run any package, like `Bash(pnpm dlx *)`, and path or version forms of ones it already ignored (`/usr/bin/npm run:*`); the safety check decides those\n- Changed `/sandbox` and most `/config` value lists to be unnumbered outside screen reader mode; use the arrow keys and Enter\n- Changed Bash commands whose variables expand to very long text to be denied, with a message that says to use shorter values or a file\n- Changed the PowerShell tool to ask before one more command, even under a rule naming it; if nothing answers, it's refused unless a whole-tool rule, auto mode or bypass permissions mode allows it\n- Changed a subagent resumed by a sibling's message to follow the session's permission settings; it may now ask, or be refused where nobody can answer\n- Changed `--worktree` with a relative `--add-dir` beside the launch folder to give the Bash sandbox that folder, not one beside the worktree\n- Changed some `--worktree` starts with a relative `--add-dir` above the launch folder, such as `..`, to give the Bash sandbox, and a `--bg` session, only the worktree\n- Changed Claude in Chrome to refuse uploading some credential files; attach those to the page yourself\n- Changed artifact approval in organizations that require it for each new artifact: an artifact that only a plugin created or allowed can be refused\n- Changed sessions with nobody to ask to refuse a new artifact at once, with no permission prompt, in organizations that require approval of each one\n- Changed the Artifact tool to refuse calls once settings turn Artifacts off, in sessions already running too\n- Changed the Artifact publish prompt to stop offering \"don't ask again\" in subagents and after a background task result, someone else's message or a compaction\n- Changed `claude plugin eval` to keep its deny rules over a plugin's permission hook, and to flag calls the hook let through\n- Changed MCP server logs to record the mode, action, and field names of elicitation requests and hook answers, not their full contents\n- Changed most prompts to ignore answers other than Esc for a moment after resuming with `fg`\n- Changed Code Mode to no longer refuse to start when an installed mod hooks tool calls; an administrator can still keep Code Mode off in managed settings\n- Changed subagents launched without a model to stay on the model they started on when resumed, including after `/model`\n- Changed the first subagent on a model to wait up to 1.5 seconds (10 in `claude -p`) for that model's settings from Anthropic, instead of starting without them\n- Changed restarted background subagents, and turns that start by themselves, to wait for plugins to load\n- Changed prompt history to leave out most launch prompts of sessions that another Claude Code session started, such as `claude --bg` run by Claude\n- Changed subagent progress summaries in terminal sessions on Amazon Bedrock, Google Vertex AI and Microsoft Foundry to be written from short notes, not the whole conversation, using far fewer tokens\n- Changed non-interactive (`-p`) mode to run fewer git commands before the first request\n- Changed first-run setup to skip terminal setup in remote VS Code, Cursor and Devin Desktop sessions, where it cannot change settings\n- Changed fullscreen text selection to leave out the glyph of system, thinking and inbound message lines\n- Changed Anthropic-hosted cloud sessions to ignore more of a repository's flag and plugin delivery settings; its `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC` now acts at most as `DISABLE_TELEMETRY`\n- [VSCode] Added \"Claude Code: Add to Chat\" to the Explorer's right-click menu, which @-mentions the selected files and folders in the chat\n- [VSCode] Fixed the permission prompt not showing the reason that a PreToolUse hook gives when it asks for confirmation (anthropics/claude-code#85950)\n- [VSCode] Fixed long tool and file names being cut off in a permission prompt's title\n- [VSCode] Fixed a terminal session's diff showing the previous proposed edit after you typed in it and Claude edited the same file again\n- [VSCode] Fixed Escape also stopping Claude's response when it closes the feedback or \"Build a custom style\" dialog, the @-mention menu, the model, mode or \"+\" menu, or another popup\n- [VSCode] Improved speed when Claude makes several edits in a row: queued edits share one wait for new problems; a problem may be reported with a later edit\n- [VSCode] Improved @-mention file search speed in large workspaces when `rg` is on your `PATH`; results now rank much as in the terminal, with less tolerance for typos\n- [VSCode] Improved startup: the model picker is ready at once in folders opened for the first time, empty windows and worktrees, except on third-party providers\n- [VSCode] Changed a chat edit's diff to stay open and keep asking for approval when a terminal session in the same window sends a prompt\n- [VSCode] Removed unused tools from the IDE connection (openFile, saveDocument, getOpenEditors and related)\n","publishedAt":"2026-11-17T23:17:45.000Z","fetchedAt":"2026-11-18T02:41:28.494Z","url":"https://code.claude.com/docs/en/changelog#2-1-327","media":[],"coverageCount":0},{"id":"rel_qYj1XfEjijH1uXrEVQ_c9","version":"v2.1.326","type":"feature","title":"v2.1.326","summary":"Sandboxed commands on macOS, Linux and WSL no longer write to some sensitive home folders by default, and file tools now refuse writes to Claude Code's sign-in files such as .credentials.json. Added support for turning on Code Mode while agent teams are enabled, plus fixes for resumed sessions losing recent turns after a crash and background sessions losing the LLM gateway your shell exported.","titleGenerated":"Claude Code 2.1.326 tightens sandbox writes and hardens Bash permissions","titleShort":"Sandboxed commands write to fewer folders; Bash permissions hardened","breaking":"major","importance":4,"content":"## What's changed\n\n- Added support for turning on Code Mode while agent teams are enabled; if you already turned it on, sessions with teams on now start in Code Mode\n- Added a startup warning when an `--allowedTools`, `--disallowedTools` or `--tools` value has an entry that doesn't look like a complete rule\n- Added a message explaining why MCP servers stop loading when an administrator adds `managed-mcp.json` during a session\n- Added a row to `/status` that says when your provider or an environment variable such as `DISABLE_TELEMETRY` turns features off\n- Added a note in `/plugin` and `claude plugin list` when a malformed `plugin.json` field such as `author` or `homepage` is ignored\n- Added Ctrl+O to show hidden startup notices\n- Added a hint in the empty prompt that says what to do after a request fails\n- Added an `esc to stop` footer hint in a subagent's view where Esc would stop it\n- Added `OTEL_RESOURCE_INCLUDE_HOST` to include `host.name` and `host.id` in OpenTelemetry resource attributes (off by default)\n- Added `overflowX=\"scroll\"` on a mod's `Box`, and `maxLines` and `wrap=\"clip\"` on its `Text`\n- Fixed `cp` of a file being refused when only its source was covered by an `Edit` deny rule\n- Fixed Bash commands that delete or overwrite files (`find -delete`, `cat a > b`, `sort -o`) being summed up in the transcript as a search or a read\n- Fixed a misleading \"Dangerous rm operation\" warning in the Bash permission prompt on some commands with no rm that Claude Code can't read reliably\n- Fixed background sessions losing the LLM gateway your shell exported after a crash, after a reopen from another folder, or when the background service lacks it: they now stop before any model request\n- Fixed the background service ending your own processes, such as an SSH login, when a process number on its list had been reused by another user's process\n- Fixed rules added in `/permissions` or at a main-conversation tool prompt, and `/add-dir` folders, looking saved when the settings file couldn't be written; the terminal now warns and names it\n- Fixed a failed MCP `headersHelper`, or one not run in an untrusted folder, making other sessions start without that server, for most servers with no stored OAuth refresh token\n- Fixed managed settings with certain key names stopping Claude Code from starting or affecting other managed files\n- Fixed settings files with values nested thousands of levels deep crashing Claude Code or applying only some of the time; values past 64 levels are now dropped with a warning\n- Fixed a settings file saved with an error mid-session being skipped silently; a warning now names the file\n- Fixed the subagent model set in managed settings not applying in Claude Desktop's Code tab; it now overrides your own, as in a terminal, unless those settings also pick a provider or gateway\n- Fixed sandboxed commands failing to start on Linux when the home or config directory is reached through a symlink into `/net` and sandbox settings deny reads of it\n- Fixed empty placeholder files, including `.git/config.lock`, sometimes being left in the project on Linux when Claude Code was stopped during a sandboxed command\n- Fixed Up then Enter answering Yes at the folder trust prompt shown under an already trusted parent folder\n- Fixed an extra Enter at the end of setup accepting the offer to make auto mode the default\n- Fixed a message typed over the startup auto mode offer being lost, and its Enter accepting the offer\n- Fixed a note typed on a permission prompt being sent to Claude after Tab had hidden it\n- Fixed PreToolUse hooks in managed settings leaving no OpenTelemetry `tool_decision` event or `permission_denials` entry when they refused a direct tool call\n- Fixed letters typed over `/sandbox` or a delete question in `/permissions` moving the selection\n- Fixed resumed sessions losing recent turns after a crash when the previous turn ended with an empty reply\n- Fixed resumed sessions losing earlier conversation history when Claude Code ran low on memory\n- Fixed a re-sent file overwriting Claude's changed copy in Remote Control and cloud sessions on Linux with no feature flags loaded: the copy moves to a `.kept-` folder, or the new file isn't saved\n- Fixed a background session being killed and its request lost when you opened, stopped or removed the older session whose conversation it had resumed\n- Fixed `claude purge` hiding its prompts when stdout is redirected, where one unseen Enter could delete a project's data; they now show on the terminal\n- Fixed the spend limit menu removing the monthly limit without asking when Backspace and Enter were pressed as it opened\n- Fixed `/rewind` restoring only some files after running `/branch` twice while a slow backup copy was still running\n- Fixed `/rewind` saying \"No code changes\" when a shell command changed files it can't restore; a turn whose changes weren't tracked now reads \"Code changes unknown\"\n- Fixed `/rewind`: Esc now cancels its summary, its list shows more than one past prompt in fullscreen, and a long prompt no longer hides its warning about losing later file changes\n- Fixed Ctrl+C or Ctrl+D twice in `/skills` quitting Claude Code and dropping changes; it now saves and closes the panel\n- Fixed the usage limit menu erasing your draft and opening while you were typing\n- Fixed pasted text and images being dropped when ↑ recalls a draft cleared with Esc Esc\n- Fixed an edited typed answer to Claude's question sending the old text when you moved on without pressing Enter\n- Fixed going back to a question you answered from its list putting the cursor on the first option, where Enter replaced your answer\n- Fixed `/feedback` and `/bug` sending the report when you typed a space on the confirmation screen, and losing the typed report after Ctrl+O or Ctrl+Z\n- Fixed a prompt typed after `--add-dir`, `--allowedTools` or another list option being dropped without a message\n- Fixed `/plan <description>` ignoring the description when plan mode was already on\n- Fixed valid PDFs being dropped from the conversation after the API rejected a different PDF read in the same batch\n- Fixed a session ending after a single failed memory allocation while drawing the screen\n- Fixed compaction on Bedrock, Vertex and Foundry, and with telemetry turned off, sometimes replying to one of your messages instead of summarizing, or leaving Claude's last reply out of the summary\n- Fixed image-heavy sessions on Vertex AI getting stuck on a 400 error about the request size exceeding a 30 MB limit, instead of compacting and retrying\n- Fixed an error when `max_tokens` exceeded the model's limit: Claude Code now retries the request with the maximum the Anthropic API reports\n- Fixed `grep` in the Bash tool silently skipping some text files\n- Fixed a crashed Node or Python hook showing only the first line of its stack trace; the error is now shown, and Ctrl+O shows up to 10,000 characters of the output\n- Fixed `/goal clear` or a new `/goal` not sticking when typed while Stop hooks were running and background work was still in progress\n- Fixed the automatic session recap being sent after a reply billed to usage credits; `/recap` still works\n- Fixed auto mode staying on its local classifier, which counts toward usage, because of a stale saved value, such as one from an older Claude Code\n- Fixed \"Sign in with your Console account\" in `/login` replacing your Claude subscription login when the browser sign-in used your Claude plan instead of a Console organization\n- Fixed `/login` ending on \"Login successful\" while an approved `ANTHROPIC_API_KEY` was still used; it now says requests still use the key\n- Fixed \"Not logged in\" staying after turning on the API key in `/config`; \"Not logged in\", `/login` and `/status` now name an unused `ANTHROPIC_API_KEY`\n- Fixed `claude auth logout` leaving credentials behind, and token refresh failing, when only a Console profile's credentials are on a drive that rejects folder timestamps\n- Fixed a past conversation resumed from agent view being sent on the background service's sign-in when the service doesn't have the terminal's API key or token\n- Fixed model requests failing on a one-off authentication error (401) in Claude Desktop's Code tab: they are now sent once more first\n- Fixed `forceLoginOrgUUID` and `forceRemoteSettingsRefresh` startup errors telling you to sign in again when a network block page (HTTP 403) was the cause\n- Fixed `claude auth login` not showing a pasted sign-in code, so a wrong paste couldn't be seen or corrected before Enter\n- Fixed `/login` and `claude setup-token` showing nothing after you submit a pasted code; they now show \"Finishing sign-in…\"\n- Fixed Esc doing nothing while the first-run sign-in waits for the browser or a pasted code; it now goes back one screen\n- Fixed `/logout` in fullscreen mode exiting without leaving the logged-out message on screen\n- Fixed the shell prompt being drawn over the last screen after quitting with Ctrl+C or Ctrl+D at a sign-in or first-run step\n- Fixed Ctrl+C twice doing nothing on the first-run theme choice, and freezing the terminal on setup screens or a background session at its startup question\n- Fixed your OFF answer to the updated terms being replaced when keys were typed while it was saving, at startup and in `/privacy-settings`\n- Fixed the `/privacy-settings` toggle sometimes saving the opposite of what the screen showed on a quick double press, changing your setting on keys pressed after Esc, and not saying when a save failed\n- Fixed `/privacy-settings` saving your choice on the wrong account after signing in to a different account in another terminal\n- Fixed `/privacy-settings` saying \"set to true\" when you picked OFF right after ON\n- Fixed skills and slash commands failing to load in Code Mode under auto mode when an inline `!` command needs approval; Claude is now asked to run it\n- Fixed background subagents started in Code Mode failing when they preload a skill with an inline `!` command\n- Fixed mods being unable to use Bash, Read and other tools through `$.tool.call` and `$.tool.check` in Code Mode\n- Fixed parallel MCP tool calls reading another call's saved result when two long results arrived in the same millisecond (anthropics/claude-code#105138)\n- Fixed tool search sometimes loading another MCP server's tool when asked for a tool by its exact name; you may be asked again for that tool\n- Fixed MCP prompts disappearing when two servers have nearly identical names\n- Fixed a tool call to an MCP server that an SDK host had just removed reconnecting to it, or reporting that it needs authentication\n- Fixed Claude not being told that a remote MCP server's question was declined when the server then failed the tool call\n- Fixed MCP tools with a root `anyOf` or `oneOf` showing Claude only the first branch's form of an argument; text such as `42` can now arrive as a number where another branch takes one\n- Fixed Claude not being told when an approved MCP resource tool was unavailable after a cloud session's worker was replaced\n- Fixed `/mcp` not showing why a local (stdio) MCP server failed to start\n- Fixed Esc doing nothing while an MCP server reconnects: Esc, or Ctrl+C pressed twice, now closes the wait and the reconnect continues in the background\n- Fixed Enable in `/mcp` showing no progress for a slow server, and a second Enter disabling it again\n- Fixed `claude mcp add-json` answering most mistakes with \"Invalid configuration: : Invalid input\"; it now says what to change\n- Fixed the Ctrl+C hint on a background MCP task's question saying \"cancel\" when the second press leaves the question to answer later\n- Fixed pressing `c` to copy an MCP sign-in URL leaving a stray `c` in the paste field\n- Fixed Tab and Shift+Tab not moving between fields in an MCP server's input form, and a required field there keeping its tick after being emptied\n- Remote Control: Fixed a session disconnecting mid-reply when the server was briefly unavailable\n- Remote Control: Fixed the QR code in `/remote-control` being cut off in short terminal windows such as 80×24\n- Remote Control: Fixed terminal sessions showing as idle on your phone while background subagents were still running; Stop on your phone or the web now ends them together with the turn\n- Fixed Remote Control showing a session as running while it waited on a prompt that can only be answered in the terminal, and as idle when turned on while a tool was running\n- Fixed a reply in a remote session running again after a restart when Stop had been pressed while Claude Code was restarting\n- Fixed Stop in cloud and self-hosted runner sessions sometimes waiting for slow MCP servers to connect before taking effect\n- Fixed files Claude sends from cloud and Remote Control sessions failing at once when a proxy answers an upload with its own HTTP 502, 504 or 520 to 524; the upload now gets a second try\n- Fixed claude.ai chats and Cowork tasks sometimes not sending a file after you chose \"Send it\" at the check for secret keys\n- Fixed claude.ai chats and Cowork tasks asking about a plugin file only because it holds a Word, Excel or other zip-type file; such files are now read for secret keys first\n- Fixed `claude self-hosted-runner` asking for new tokens nonstop when a refreshed token was already due, such as when the machine's clock runs ahead\n- Fixed `/autofix-pr` reporting a run as cancelled when its cloud session could not be stopped; it now links the session so you can stop it\n- Fixed `/teleport` hiding a session from its list in narrow terminals\n- Fixed artifact publishing refusing files in a background session's temporary folder\n- Fixed subagents whose `disallowedTools` lists `Artifact` still being offered artifact tools in some cloud sessions\n- Fixed Claude calling a just-published artifact private when people with access to its folder can open it\n- Fixed the terminal not showing an artifact's link after Claude publishes or updates it\n- Fixed the main conversation of a terminal session waiting in silence for a background subagent that another subagent had messaged: it's now told which agent the subagent reports to\n- Fixed background agents being told you refused their tool call when a session ended by Ctrl+C twice at the exit dialog, or after an interface crash\n- Fixed Up and Esc moving the main conversation's queued messages into the prompt box of a subagent or teammate you are viewing\n- Fixed the agent rows under the prompt showing an agent as busy after its turn ended: the row now says what it is waiting on, and `/tasks` names the subagent behind each background command\n- Fixed stopped and failed agents looking like running ones in the agent rows under the prompt, or showing \"Done\" when run in parallel, and a stopped agent's last command still showing \"Running…\"\n- Fixed an opened background agent showing a normal spinner while its requests were being retried\n- Fixed the \"background agents launched\" line stopping at two when Claude starts three or more agents in one reply\n- Fixed every earlier turn's \"done\" line showing \"1 shell still running\" while a background task ran\n- Fixed subagent tool counts in auto mode including the call that delivers the agent's final report\n- Fixed the welcome header outside fullscreen mode showing the session's model while viewing a subagent or teammate that runs on a different model\n- Fixed images pasted while viewing a background agent arriving as the text `[Image #1]`\n- Fixed long background agent and background command results wrapping under the bullet instead of lining up with their text\n- Fixed `claude --worktree` showing nothing while a slow worktree is set up\n- Fixed the worktree exit dialog counting a new folder as one uncommitted file\n- Fixed workflow resume telling Claude every finished agent would be reused after a call failed; Claude is now told how many results were reused and where reuse stopped\n- Fixed a hang on malformed glob patterns in settings, plugin marketplace entries and Grep file filters\n- Fixed plugin and marketplace requests waiting forever when `ALL_PROXY` is the only proxy set and the proxy never answers; they now fail after 25 seconds\n- Fixed the `/` menu, `/skills` and `/plugin` repeating rows and losing the cursor when two skills or commands in one place share a name\n- Fixed search in `/resume`, `/plugin`, `/permissions` and `/config` ignoring IME input, dictation and fast typing\n- Fixed the cursor disappearing after clearing a search on the `/plugin` Installed tab\n- Fixed a mod's `Text` with `wrap=\"truncate-end\"` losing rows on the terminal when it holds a line feed\n- Fixed `/model` showing your saved model as \"Default (recommended)\": Default now names your plan's default model, and choosing it in the terminal switches you back to that default\n- Fixed a first skill, command or agent created during a session not being found right away\n- Fixed `/clear --help` and other built-in commands running at an idle prompt instead of showing usage; added `/help <command>`\n- Fixed a half-written reply sitting frozen with no spinner while Claude Code retried after a mid-response server error\n- Fixed the API retry notice sitting at \"Retrying in 0s\" during the next attempt, now \"retrying…\", and the \"will retry in\" countdown freezing when a reply stalled or with Reduce motion on\n- Fixed the transcript dropping a few rows, or the prompt jumping up, as a reply started streaming\n- Fixed a line of a long list or code block going missing from scrollback, and text jumping a row, while a response streams\n- Fixed a long queued message hiding Claude's reply while it streams, and erasing it from scrollback in the classic renderer\n- Fixed fullscreen text jumping when a reply finishes while you're scrolled up or after answering an Edit or Write prompt, and losing your place or jumping turns when resizing\n- Fixed the fullscreen transcript going blank when searching or jumping far back, and freezing partway through a fast scroll, in long sessions\n- Fixed the pinned prompt in fullscreen showing an older question while a reply streams, and being copied with the reply when you drag a selection past the top\n- Fixed the screen going blank and dropping keys after picking a model in `/model`, and `/model <alias>` and `/effort <level>` briefly hiding the prompt and dropping letters typed right after\n- Fixed the screen scrambling after `/tui default` in a long conversation, or when the `tui` setting changed in another window (anthropics/claude-code#94879)\n- Fixed the screen freezing for seconds when showing long newly created files in the permission prompt and the Ctrl+O transcript\n- Fixed some very long Bash or PowerShell commands freezing the session for seconds\n- Fixed Shift+Enter or Option+Enter pressed right after a paste sending the prompt instead of adding a new line\n- Fixed Up, Down, End and mouse clicks putting the cursor on the wrong row of a prompt that wraps\n- Fixed undo (Ctrl+_) losing the last words typed when pressed after deleting a line or word\n- Fixed code or a diff pasted into a prompt losing its indentation in the transcript, and a pasted diff's added lines showing as removed\n- Fixed `[Image #N]` landing in the middle of a word when typing right after pasting, and a pasted image not being sent after deleting its placeholder and pasting it back with Ctrl+Y\n- Fixed asterisks disappearing from your own prompts in the transcript, such as `src/**/*.ts` or `2*3*4`\n- Fixed `claude -p` showing nothing in the terminal while it retried an overloaded or failing API request, or when it failed with its text output redirected or captured\n- Fixed Windows paths in Claude's responses losing the backslash before folders like `.claude` or `_build`\n- Fixed bold text showing as raw `**` in Japanese, Chinese and Korean replies when it starts or ends with punctuation\n- Fixed `/memory` opening a read-only file with no warning that changes may not save\n- Fixed Claude editing `~/.claude/keybindings.json`, which has no effect in Claude Desktop, when asked to change a shortcut there\n- Fixed Claude saying that an administrator can turn on Claude in Chrome when your organization's policy disables it, in apps that provide its tools such as Claude Desktop's Code tab\n- Fixed the suggested next prompt sometimes being a bare tag, such as `<signal type=\"empty\" />`\n- Fixed long command output ending mid-line in the expanded view with no sign it was cut; it now names the saved file\n- Fixed collapsed tool lines reading \"Ran\" or \"Called\" while still asking permission, and \"Running…\" or \"Reading…\" again when a panel opened after a call was stopped or refused\n- Fixed failed, blocked or refused commands looking like passing ones in a collapsed tool line: it now shows \"· 1 failed\" or says they did not run\n- Fixed the \"done\" line showing after a turn that failed or was refused at a permission prompt\n- Fixed failed or refused skill and agent calls still showing \"Initializing…\"\n- Fixed `/btw` showing only a lead-in such as \"Let me check that file.\" with no notice when the model tried to use a tool (anthropics/claude-code#90314)\n- Fixed long `/btw` answers looking finished and ignoring PgUp, PgDn, Home, End and the mouse wheel\n- Fixed `/resume` search saying \"No sessions match\" for older sessions that exist, and a typed number opening a session instead of searching\n- Fixed pressing Ctrl+R again in fullscreen prompt search not moving to the next older match\n- Fixed Tab not completing file names without a slash in shell mode (`!`), and completed paths with spaces or shell characters not being quoted\n- Fixed a rebound `chat:submit` key doing nothing while the slash command or @ file menu was open\n- Fixed holding Esc over typed text opening Rewind after the text was cleared\n- Fixed Esc over typed text cancelling a pending `/loop`\n- Fixed \"Esc again to clear\" staying on screen after a second Esc would no longer clear the prompt\n- Fixed typing landing in the wrong place in a question's note after going back to it or editing it in an external editor\n- Fixed the \"paste again to expand\" hint hiding the mode and \"esc to interrupt\" after a paste was sent or deleted\n- Fixed \"paste again to expand\" hiding the middle of a paste longer than 10,000 characters in fullscreen mode, and a prompt collapsing while you type past that length\n- Fixed the arrow keys and `q` not working in a pager that `claude --help` is piped to, as in `claude --help | less`\n- Fixed Claude Desktop's Code tab blaming the internet connection when a certificate's signature couldn't be verified\n- Fixed Web Search showing nothing while its request was being retried\n- Fixed desktop notifications not being sent inside tmux (with `allow-passthrough` on), or to iTerm2 over SSH, on the default setting\n- Fixed `/copy` and `/export` saying \"copied to clipboard\" over SSH when the terminal may not have copied anything\n- Fixed `/usage-credits` saying \"Request sent to your admin\" when the request was refused; people who have billing access themselves are now pointed to the usage settings page\n- Fixed the transcript share question for Pro and Max users staying open after the next message, `/clear` or `/resume`\n- Fixed an upgrade hint appearing under an old usage limit message after the limit had reset\n- Fixed the effort level notice reappearing after every permission prompt or dialog, and the usage limit warning and other notices not appearing while the context warning showed\n- Fixed `/tui default` saying it had switched back to the classic renderer when a setting kept fullscreen on; after a `/tui` that can't switch, Claude Code now says what keeps the renderer\n- Fixed a new account's first session using a different renderer from later ones\n- Fixed Reduce motion and Show tips in `/config` applying only to the current folder\n- Fixed typing and spinners redrawing the whole pane in tmux 3.7\n- Fixed Option+T typing † instead of toggling thinking over SSH from iTerm2 when Option isn't set as Meta\n- Fixed `/terminal-setup` not naming the tmux settings that Shift+Enter needs, and promising Shift+Enter in Windows Terminal before 1.25, where it sends the prompt\n- Fixed Program Status Protocol (OSC 7501) terminals showing a finished turn after idle compaction, or no status for background sessions opened with `claude agents`, `claude attach` or `--resume`\n- Fixed terminals that support the Program Status Protocol (OSC 7501) not seeing Claude Code as waiting on you while the folder trust prompt is open (anthropics/claude-code#105801)\n- Fixed the terminal progress bar showing Claude as working while it waited for your answer or was suspended with Ctrl+Z, and disappearing in Ghostty after 15 seconds while Claude was still working\n- Fixed fullscreen giving no sign that a permission prompt or question was waiting while you were scrolled up\n- Fixed the fullscreen tool group headline naming a finished or queued call instead of the running one\n- Fixed finished tool groups in fullscreen counting a search or listing also as a shell command, and the tool timer there counting time at a permission prompt or queued behind another command\n- Fixed a multi-line shell command filling the screen in fullscreen while it runs\n- Fixed `!` shell commands showing no output for their first three seconds\n- Fixed a running command showing no output or timer when its collapsed row was clicked open in fullscreen mode\n- Fixed the jump-to-bottom pill disappearing in fullscreen while more of the reply was still below\n- Fixed the prompt disappearing in short fullscreen windows while a survey or notice showed above it\n- Fixed the custom status line disappearing and the screen jumping on Ctrl+C, in `!` shell mode and while pasting an image\n- Fixed a custom status line keeping its old width, folder and session name after a terminal resize, `/cd` or `/rename` (anthropics/claude-code#107291)\n- Fixed \"esc to interrupt\" missing beside a custom status line and showing while Esc would go to a panel or footer item\n- Fixed mouse escape codes being typed into the shell after exiting fullscreen mode\n- Fixed the Ctrl+Z message printing over the prompt box outside fullscreen\n- Fixed `claude logs` painting over and mixing with what was already in the terminal\n- Fixed the spinner tip flashing back and the screen bouncing as a reply finished\n- Fixed the \"Reading 1 file…\" line blinking out and the prompt jumping when a tool call finished streaming\n- Fixed the `claude --resume` session preview being garbled and not scrolling in fullscreen mode\n- Fixed Esc not leaving the \"Connect to your IDE automatically?\" question in `/ide`, where Ctrl+C twice quit Claude Code\n- Fixed first-run terminal setup saying nothing about what it changed or could not set up\n- Fixed `/status` showing installation warnings twice\n- Fixed the footer's PR badge showing the previous branch's pull request after switching branches\n- Fixed the \"Pushed to\" summary naming the fetched branch when a command fetched or pulled first, and \"Merged PR\" or \"Closed PR\" showing when `gh` refused the action and the exit code was hidden\n- Fixed URLs that wrap over several lines not being clickable in terminals with link support that Claude Code doesn't recognize, and labeled links in the Rex terminal\n- Fixed the Ctrl+O transcript hinting Ctrl+B to background a command in fullscreen, where it pages up, and showing no sign that Claude is still working\n- Fixed the Ctrl+R search, the model picker, plan approval, question previews, background shell output and `/plugin` install choices being cut off in short fullscreen windows\n- Fixed the `/context` grid, API errors such as \"Credit balance too low\", \"Waiting for API response\" and retry notices, \"esc to interrupt\" and \"Interrupted\" being cut off or jumbled in narrow terminals\n- Fixed wide tables breaking apart in plans and subagent prompts, cutting short words in half beside a long cell, and showing stray words and blank rows when drawn as label and value lists\n- Fixed code keywords, added diff lines, the slow-reply spinner and \"thinking\" being hard to read in some themes and terminals\n- Fixed code in diffs and file previews being invisible when the theme and the terminal background differ\n- Fixed the unused part of `/usage` bars looking full in the light and ANSI themes\n- Fixed the `/effort` slider ignoring mouse clicks in fullscreen mode\n- Fixed text being drawn out of place after skin-tone or joined emoji, symbols such as ✌ and ☝, and Hindi, Tamil and other Indic vowel signs in terminals that draw them wider\n- Fixed joined emoji pushing rows out of place in GNU screen\n- Fixed the last item of JSON lists and right-aligned numbers losing their indent in command and tool output\n- Fixed diffs numbering \"No newline at end of file\" as a line of code, which shifted the line numbers after it\n- Fixed long single-line messages showing \"(0 lines hidden)\" instead of how much text is hidden\n- Fixed notebook edits showing no tool row when Claude left out the cell type or deleted a cell, and showing `@undefined` when inserting a cell at the top\n- Fixed the \"Switch model?\" and \"Change effort level?\" questions showing no key hints\n- Fixed the `?` shortcuts list running its columns together in narrow terminals and being cut off at the bottom in short fullscreen windows\n- Fixed leftover hint text in screen reader mode on macOS and Linux after closing the `/permissions` search when the rule list was taller than the terminal\n- Improved Bash permission checks: some commands now ask first, or are refused where no one can answer\n- Improved permission checks in Code Mode; more sandboxed commands now ask or are refused\n- Improved the messages shown when Code Mode can't be turned on, so they say what you can do\n- Improved Bash sandbox protections on macOS, Linux and WSL\n- Changed edits in a session's mods folder to ask more often, and sandboxed commands to write to fewer folders; if every edit asks, remove any link there to an unreadable or network path\n- Improved Claude's sandbox instructions on Linux to explain that a server is reachable only from the command that started it\n- Improved how secrets are masked in push notifications from Remote Control and cloud sessions\n- Improved `/logout` and `claude auth logout` in the terminal to retry revoking the login, and to say where to revoke it if the server doesn't confirm\n- Improved Remote Control: replies from `claude remote-control`, Claude Desktop and VS Code sessions now stream to your phone or browser as they are written\n- Improved typing, streaming and resize responsiveness, especially in long sessions\n- Improved response time after reading or editing files, most noticeably on Windows\n- Improved idle CPU use: an idle session now wakes the machine less often\n- Improved CPU use when more than 20 subagents are running: token counts of background agents whose transcript isn't open now update every 5 seconds\n- Windows: Improved the speed of a session's first Bash command by about a second\n- Improved response time of the first message after `/compact` in git repositories\n- Improved response time in Claude Desktop for messages queued while Claude finishes a reply\n- Improved time to first response in Claude Desktop's new sessions: the first turn no longer waits up to 500 ms for local MCP servers whose tools load on demand\n- Improved how a PreToolUse hook's `updatedInput` is kept when more than one hook answers a tool call; the rewrite now runs in more cases\n- Improved the error when a PreToolUse hook exits 127, saying its check may not have run\n- Improved hook failures, blocks and timeouts in the transcript: failures are now red instead of grey, a hook that stops Claude is yellow, and the reason is shown under them\n- Improved the folder trust prompt, at launch and in `/cd`, to name the project settings files Claude Code will load; at launch it also marks files that run commands\n- Improved the folder trust prompt to say why Claude Code exited\n- Improved the message in non-interactive runs and background agents when a permission check fails with an error, so it names the cause\n- Improved Claude in Chrome permission requests from a terminal session over Remote Control: on claude.ai they now name the site Claude wants to act on, also when it is already allowed\n- Improved MCP tool errors in cloud sessions: when a call through the session's proxy is cut off after it was sent, Claude is told that the outcome is unknown\n- Improved `/update-config` in the terminal to use about 13,000 tokens of context instead of 93,000, and a rejected settings edit to quote only the changed settings\n- Improved Program Status Protocol (OSC 7501) reports: background commands, monitors and MCP tasks are listed, and a session waiting on them shows as working instead of done\n- Improved Monitor: when a watch expires with no events and its command wrote to stderr, Claude is now told so and told which file holds the output\n- Improved Claude in Chrome: a long `browser_batch` call now has 150 seconds to answer, up from 90, so fewer finished results are lost\n- Improved `/doctor` to share findings as it works instead of staying silent until the full report\n- Improved the error for a settings or `.mcp.json` file with broken JSON: it now gives the line and column\n- Improved commands that are off for your provider, sign-in or an environment variable such as `DISABLE_TELEMETRY` to say why instead of \"Unknown command\"\n- Improved the `CERT_SIGNATURE_FAILURE` connection error to name a possible cause after a CA renewal and what IT can change\n- Improved the message shown when a team's pooled group budget is used up: it now says when the budget resets\n- Improved the error shown when the API refuses a Console sign-in: it now says to run `/login` instead of pointing to an admin setting\n- Improved sign-in prompts: pasting a verification code or a web address now says what to do instead, and the prompt stays so you can paste the right code (anthropics/claude-code#97)\n- Improved the `/login` screen: the code goes in a proper text field, a rejected code says why under it, and pasting in a short window scrolls the field into view\n- Improved `/design-login`, `/install-github-app` and MCP sign-in screens to match `/login`; the first two now mask the pasted code\n- Improved the Claude apps gateway sign-in screens: the sign-in address is a link and `c` copies it\n- Improved `/plugin` Discover and marketplace lists: they fit short windows with the cursor and key hints on screen, and Home, End, PgUp, PgDn, the mouse wheel and clicks now work\n- Improved `/plugin` Marketplaces and Add marketplace: a compact list, a clearer remove confirmation, a clear text field and shorter errors when adding fails\n- Improved `/plugin` Installed tab: Home, End, PgUp, PgDn and clicks now work, and hidden rows are counted\n- Improved `/plugin` Errors and Stats tabs: long errors no longer fill the screen, uninstalling from an error asks first, and long skill names no longer break the table\n- Improved `claude plugin`, `mcp`, `update`, `doctor` and `auth logout` output on a terminal to mark success, failure and warnings the same way (piped output keeps its old form)\n- Improved wording across the terminal: `/plugin`, `/mcp`, sign-in, setup, settings, the agent view, `/teleport`, Remote Control, update messages, spinner tips, `/powerup` and `claude --help`\n- Improved key hints in many dialogs and the `?` lists: they show the keys that work, in one order and wording, and fit narrow windows; deeper screens in `/hooks` and `/permissions` say Esc goes back\n- Improved text fields in `/add-dir`, `/export`, `/permissions`, `/feedback`, `/install-github-app` and Bedrock and Vertex setup: one look, long text wraps, and a refused value can be corrected in place\n- Improved dialogs, panels and the transcript: errors, warnings and confirmations in many dialogs start with ✘, ⚠ or ✔, and loading lines, status rows, counts and durations look and read more alike\n- Improved long menus such as `/permissions`, `/resume` and `/model` to say how many rows are above and below, and the `/` and `@` menus to count hidden rows outside fullscreen and screen reader mode\n- Improved list navigation: more lists wrap at both ends, checkbox lists in slash commands take Home and End, and `/permissions` says when a search finds nothing\n- Improved the pickers in agent view to take Home, End and paging\n- Improved agent view and the agent rows under the prompt to mark the selected row with the same pointer and color as other lists\n- Improved the footer after Ctrl+C stops background agents: it now offers `↓ to resume subagent` while a stopped subagent is listed\n- Improved fullscreen mode: the prompt's borders now show how many lines of a long prompt are out of view\n- Improved `/autocompact` and `/scroll-speed` with a slider that shows every value, takes Home, End and (in fullscreen) clicks, and fits narrow terminals\n- Improved on/off settings to read `on` / `off` in `/config`, `/memory` and auto mode setup\n- Improved the send-now hint under queued messages: it now says \"Queued\" and no longer looks like part of your message\n- Improved terminal setup to list every change before making it\n- Improved alignment and spacing in `/btw`, `/help`, `/scroll-speed`, the settings error dialog and dialogs that suggest a plugin or setting\n- Changed terminal and VS Code sessions in organizations with the Claude Code + Cowork HIPAA configuration to start in auto mode by default; `disableAutoMode` turns it off\n- Changed terminal and VS Code sessions with no mode set to start in auto mode, not Manual mode, when your organization's policy hasn't loaded\n- Changed a `managed-mcp.json` added mid-session so it may apply at the next check; restart open sessions for it to apply in full\n- Changed Code Mode's `$` shell helper to refuse some values with an error; nothing runs\n- Changed auto mode to stop some shell commands that ran before\n- Changed some Bash commands on macOS that ran before to ask or be refused\n- Changed a hook's `\"decision\": \"block\"` to hold in more cases, so a hook that always prints one may start blocking calls\n- Changed sessions that start in auto by default to take a server-managed `defaultMode` once settings load, or Manual if settings fail with no stored copy and policy fails or requires it\n- Changed auto mode for organizations with the Claude Code + Cowork HIPAA configuration: on a direct Anthropic API connection, the server-side classifier now checks actions, as for other organizations\n- Changed the Bash sandbox to keep its state in some cases during a session, until you run `/sandbox` or `/cd`; Bash may ask first\n- Changed sandboxed commands on macOS, Linux and WSL to no longer write to some sensitive home folders by default; some can be allowed with `sandbox.filesystem.allowWrite` in user or managed settings\n- Changed the Bash sandbox to set aside unexpected entries in a repository's `.git` folder after a command\n- Changed the Bash sandbox on Linux and WSL to create some missing `--add-dir` folders before each sandboxed command, so commands that create them can fail, skip, or land inside them\n- Changed commands that a permission hook or SDK host rewrites while approving to stay in the sandbox if the original was checked as sandboxed; the approver can add `dangerouslyDisableSandbox: true`\n- Changed `claude mcp serve` to ignore `sandbox.excludedCommands`\n- Changed `/rewind` on macOS, Linux and WSL to leave some files untouched and report them as skipped when the sandbox is on\n- Changed the Bash sandbox to ask before some lines that set environment variables when auto-allow is off; unattended runs can refuse them\n- Changed Claude's file tools to refuse writes to Claude Code's sign-in files, such as `.credentials.json`\n- Changed the Bash permission check to ask before more commands reach Claude Code's sign-in files\n- Changed some file tools to refuse some device and process files\n- Changed some `cp` and `mv` commands to ask first\n- Changed some tool calls in bypass permissions mode to ask for approval, or be denied where no one can be asked\n- Changed the Bash permission check to ask about some commands with very long values it can't finish checking, or to refuse them where a deny rule may apply or no one can answer\n- Changed subagents isolated in a worktree to refuse, in any repository, some git commands in Bash that can rewrite branches shared with your checkout; sessions are not affected\n- Changed `ExitWorktree` to refuse removing a worktree while a background subagent or shell command may still be running in it, unless Claude passes `discard_changes`\n- Changed the computer use macOS permissions panel to ignore j and k; the arrow keys and Ctrl+P/Ctrl+N still move the selection\n- Changed the PowerShell tool to ask before some more commands; unattended runs refuse them unless an allow rule, auto mode or bypass permissions mode allows them\n- Changed the PowerShell tool to ask about some more commands, even in bypass permissions mode, and to refuse some of them\n- Changed some malformed PowerShell commands to be denied\n- Changed some PowerShell tool commands that define a function named like a built-in command to be refused, so Claude can rename it\n- Changed auto mode to ignore some wildcard shell allow rules, such as ones that name no program or that start with `$HOME`, so the safety check decides those commands\n- Changed auto mode to judge wildcard allow rules whose program is a Windows backslash path, so those calls can be blocked; exact rules keep their allow; we plan to restore rules that start with `.\\`\n- Changed a managed `autoMode.environment` list without `\"$defaults\"` to replace every built-in entry in Claude Desktop and Agent SDK apps too; add `\"$defaults\"` or copy the sensitive-target ones\n- Changed `--add-dir` and `/add-dir` in local terminal sessions to show the trust prompt for a folder you have not trusted yet; `claude --bg` refuses such a folder\n- Changed self-hosted runners under `--trust-workspace false` to ignore more repository settings; set such values in user or managed settings\n- Changed how permission answers over Remote Control and on self-hosted runners save \"always allow\" choices\n- Changed enabled plugins that were refused for a malformed `plugin.json` display field: they now load, with their hooks and MCP servers; administrators should check the plugins they deploy\n- Changed how `allowedHttpHookUrls` and `httpHookAllowedEnvVars` apply to hooks from managed settings\n- Changed HTTP hooks to ignore `Content-Length` and `Transfer-Encoding` in their `headers` setting; the request's length comes from its body\n- Changed plugin uninstall to also delete the files extracted from the plugin's MCP bundles (`.mcpb`, `.dxt`), unless another plugin uses the same bundle\n- Changed some worktree sessions to skip some plugin settings; a notice names what was skipped\n- Changed `/plugin` uninstall, remove and switch questions to lists answered with the arrow keys and Enter, or Esc to cancel; `y`, `n` and `u` no longer answer them\n- Changed automatic compaction in plan mode to use the model answering you, such as Opus with `opusplan`, so it can cost more\n- Changed Cowork cloud sessions to hold an organization's OpenTelemetry export until the first turn; a session that gets no message now sends nothing, so `claude_code.session.count` can fall\n- Changed Claude's messages to a session on another machine to be refused when your session uses an API key, auth token, profile or workload identity beside a saved claude.ai login\n- Changed `--bare` sessions on an unapproved API key: moving one to the background is now refused in most setups, and `claude --bare --bg` exits with an error if the background service lacks the key\n- Changed \"Continue automatically at usage limit\" to stop with a notice, sending nothing, when your login changes or ends during the wait\n- Remote Control: Changed how it connects through a proxy: requests now use a CONNECT tunnel, so a proxy that refuses CONNECT stops Remote Control with an error\n- Remote Control: Changed the session debug logs written under `--verbose` to go to `~/.claude/child-logs/`\n- Self-hosted runner: Changed Linux and macOS sessions with no `gh` to get a built-in `gh api` by default, via Anthropic, which adds their GitHub sign-in, if any; `--use-anthropic-gh false` turns it off\n- Changed agent view so that quitting with Ctrl+C twice, Ctrl+D twice or `/exit` after pressing `←` in an empty conversation ends that session, instead of leaving it running for 5 minutes\n- Changed `--await-initialize` to be refused with `--bg` or `--worktree --tmux`, which never read the initialize request, instead of starting without the fields it carries\n- Changed worktree creation, `/teleport` and IDE checkouts to run git with a stricter `PATH`; a git filter or helper may need its full path\n- Changed the card for approving a subagent's plan to two choices, \"Yes, and switch to default (ask each time) for this session\" and \"No\", chosen with the arrow keys and Enter (typed text moves to No)\n- Changed the `/claude-api` skill to build agents on Managed Agents by default, unless you want to host the agent loop yourself\n- Changed `/usage` in fullscreen mode to show plan limit bars before session cost for subscribers; `/cost` still starts with cost\n- Changed the wording of some messages that `claude` subcommands print; JSON output is unchanged\n- Changed first-run setup to preselect the \"Auto (match terminal)\" theme instead of Dark mode\n- Changed fullscreen `/cd` to refuse \"Yes\" while its trust prompt doesn't fit, with a line saying to make the window taller\n- Removed the tmux notice that said the mouse wheel needs `set -g mouse on` in fullscreen; the wheel scrolls without it\n- Removed the spinner tip that suggested `/clear` 30 minutes into a running turn\n- [VSCode] Added a `claudeCode.promptSuggestions` setting that turns prompt suggestions off\n- [VSCode] Fixed \"Rewind code to here\" in a forked conversation sometimes undoing one turn too many and deleting files created in that turn\n- [VSCode] Fixed the Sandbox dialog creating empty files and folders in the home directory and the project's `.git` folder while the sandbox is off\n- [VSCode] Fixed the saved sign-in being lost on some networks\n- [VSCode] Fixed sign-in ending when a verification code, a web address or a code from another sign-in attempt is pasted; the panel now says what to paste instead\n- [VSCode] Fixed the sign-in screen showing briefly each time a panel or window opened while signed in to Console without an API key\n- [VSCode] Fixed the usage-limit banner's \"Request more\" button for Team and Enterprise members coming back after each press without sending anything, after you signed in again during a session\n- [VSCode] Fixed `/ide` in a second Claude Code session silently losing the editor connection back to the first session; the newest connection now wins and the displaced session shows how to switch back\n- [VSCode] Fixed saved memories not being listed in the Memory dialog, or RAM use growing without end, when the memory folder held symlink loops, and a hang when it held a named pipe\n- [VSCode] Fixed a running command's live output stopping after one failed read\n- [VSCode] Fixed \"Open in Terminal\" showing only the shell's error when `claude` could not be started: in bash, zsh, fish and Git Bash, a notice now explains why and links to the install guide\n- [VSCode] Fixed the panel showing no \"Open artifact\" link after Claude publishes an artifact\n- [VSCode] Fixed the prompt suggestion hiding the \"attach selected text\" hint while the chat panel is unfocused\n- [VSCode] Fixed buttons with no visible border in the high-contrast themes, and the dark welcome art on the sign-in screen in Light High Contrast\n- [VSCode] Improved Edit and Write speed: no wait for new problems on files that are not open in a tab when the editor has been reporting none for such files\n- [VSCode] Changed a running command's live output in the chat to look like the command card's OUT row\n- [VSCode] Changed the text of extension errors in event listeners, views and activation to stay out of the editor's own error telemetry\n","publishedAt":"2026-11-16T22:11:05.000Z","fetchedAt":"2026-11-16T22:22:19.758Z","url":"https://code.claude.com/docs/en/changelog#2-1-326","media":[],"coverageCount":0},{"id":"rel_PDTpeSlKRQmK17wmUFUjo","version":"v2.1.325","type":"feature","title":"v2.1.325","summary":"A security issue in how zstd-compressed responses are decoded is fixed. Sandboxed Bash commands now follow the normal permission flow by default unless sandbox.autoAllowBashIfSandboxed is set, and read-only git commands ask first in untrusted repositories.","titleGenerated":"Claude Code 2.1.325 fixes zstd decoding security issue","titleShort":"zstd decoding security issue fixed; sandboxed Bash now prompts","breaking":"major","importance":4,"content":"## What's changed\n\n- Fixed a security issue in how zstd-compressed responses are decoded\n- Added a spinner tip in VS Code and similar integrated terminals about re-running a notebook cell, shown once Claude has read or edited a Jupyter notebook\n- Added a line under the error shown when your organization's policy blocks a request, saying how to continue: run `/rewind` and change the message, or run `/clear`\n- Added a note for Claude when a sandboxed `git checkout`, `git stash` or autostash pull cannot update protected files such as those in `.claude`\n- Fixed `/security-review` and skills' inline git commands stopping with an error in worktrees of bare repositories, `--separate-git-dir` checkouts and Android `repo` checkouts\n- Fixed subagents with their own model being moved to the plan-mode model when the model setting is `opusplan` or `haiku`\n- Fixed subagents on a different model sizing file reads and MCP results with the main session's model\n- Fixed sessions on many third-party gateways and Foundry stopping with a context length error on every turn instead of compacting\n- Fixed recent turns sometimes going missing from a resumed session after Claude Code stopped without a clean exit\n- Fixed shell commands running in the launch directory after the working directory was deleted; the command now fails with a message and can be re-issued\n- Fixed an approved plan edit being lost when the `plansDirectory` folder can't take it: the plan is now saved in the default plans folder, and a warning says where and why\n- Fixed a saved `sonnet[1m]` model setting being rewritten on a new install to a model id the API rejects\n- Fixed focus view hiding the reply to your prompt when a scheduled task replied after it\n- Fixed the prompt sometimes missing from a background subagent's view when fullscreen is off\n- Fixed long options in lists and permission prompts being partly overwritten when they wrap, and first-run screens losing lines on narrow terminals\n- Fixed startup warnings about permission rules and the \"N MCP servers need authentication\" notice being recorded as seen when they were scrolled out of view\n- Fixed the welcome header showing control characters or splitting across lines when an agent's name or the working folder contains them or a line break\n- Fixed the red error shown when Claude tries a command only you can run: a dim line now says it runs only when you type it at the prompt\n- Fixed `/stats` showing an Active days total that was one day too high\n- Fixed `/insights` saying a report was ready when no session qualified for analysis, and writing two values into its HTML report without escaping them\n- Fixed `/voice` saying voice mode changed when managed settings, `--settings`, or the project's `.claude/settings.json` or `.claude/settings.local.json` kept it as it was; it now says what holds it\n- Fixed the IDE welcome dialog showing Cmd+Esc for Quick Launch on Windows and Linux\n- Fixed Slack app tips showing when your organization's policy denies cloud sessions\n- Fixed `claude purge` hanging silently when there is no terminal to read keys from, and its project picker and `-i` prompts ignoring the keyboard when stdin is piped\n- Fixed `claude plugin eval --output-dir ''` writing the report into the current folder; it now exits with an error\n- Fixed `claude -p` text output on a terminal passing control sequences other than text styling; pipes and files are unchanged\n- Fixed leftover `.oauth_refresh.lock.owner.tmp.*` files piling up in the config directory\n- Fixed `/fast` saying \"Fast mode has been disabled by your organization\" under the keyless Console sign-in, where Claude Code had not asked the server\n- Fixed `/model` saying that your organization restricts a model when Claude Code could not check that, for example with no network\n- Fixed \"Sign in with your Console account\" in `/login` replacing your existing login when you pressed Esc right after the browser step\n- Fixed the browser showing the sign-in success page in some cases where sign-in failed or was canceled\n- Fixed `claude auth status` and other short commands leaving a lock behind on a Console profile, which made the next session hang for about a minute\n- Fixed background sessions started on your claude.ai login using (and billing) an API key or token from the shell that started the background service\n- Fixed more cases of a session moving to another sign-in in the background; these moves are now refused, with a message saying what works\n- Fixed a second Enter in the exit dialog stopping background work after the move to the background was refused\n- Fixed token counting on Amazon Bedrock sending the conversation to Bedrock when the `allowedProviders` managed setting does not allow it\n- Fixed AWS `credential_process` helpers and Google Cloud default credentials failing when Claude Code is started without a `HOME` environment variable\n- Fixed a session sometimes going back to an earlier permission mode after a subagent's plan was approved\n- Fixed one invalid value in an Agent SDK `applyFlagSettings()` call silently discarding all settings from earlier calls; a call that would do this is now refused with an error\n- Fixed sessions showing as waiting for input when a finished pull request only needed your own review or merge\n- Fixed an unrecognized `teammateMode` settings value being ignored without a warning: `claude -p`, the startup warning and `claude doctor` now list it as a skipped entry (still not applied)\n- Fixed `--worktree` skipping a `WorktreeCreate` hook from a plugin in `CLAUDE_CODE_PLUGIN_SEED_DIR` on the first run in a new home directory\n- Fixed brief-mode sessions, such as Dispatch, sending a reply again after the turn had already replied\n- Fixed Claude saying a working Atlassian, Asana or Sentry connector had failed in cloud sessions when a plugin listed it at a different URL\n- Fixed Claude Desktop and SDK sessions listing a folder's own MCP servers (`.mcp.json`, local) as host-added or not at all, and not connecting WebSocket ones, after a folder move or pre-warmed start\n- Fixed `/mcp` still listing a plugin's MCP server after a plugin reload removed it while it was connecting\n- Fixed an MCP server staying \"needs authentication\", and its `mcp_tool` hooks being skipped, after a token refresh that was refused only for a moment\n- Fixed MCP elicitation requests being cancelled without asking in headless mode after a server's connection was replaced during a tool call\n- Fixed MCP tool calls from an SDK host answering \"Not connected\" after a tool call had reconnected the MCP server\n- Fixed an MCP task getting stuck after its question was closed with Esc or asked again: the footer now says it needs your input, and Enter in `/tasks` shows the question again\n- Fixed background MCP tasks that stopped updating after another call to the same server lost the connection, when a later call reconnects\n- Fixed a background MCP task's question opening over a message you were still typing; it now waits until you send or clear the draft, and the spinner and turn timer keep running\n- Fixed `worktree.sparsePaths` and sparse plugin and marketplace clones leaving out the repository's root files on older git versions, such as 2.34\n- Fixed self-hosted runner sessions with several repositories being unable to use a repository added mid-session\n- Fixed Claude asking for permission before reading your own artifact or its data\n- Fixed deny and ask permission rules and hooks for the Artifact tool not applying in some cloud sessions\n- Fixed Code Mode scripts ignoring a worktree entered with `EnterWorktree`, writing `~` paths into the project, and leaving `/mcp` showing an MCP server as connected after its sign-in expired\n- Fixed words typed with a Yes at a permission dialog or plan approval not reaching Claude in Code Mode\n- Fixed Claude starting work by itself when you resumed a session that ended during a Code Mode script; it now waits for your next prompt\n- Fixed a missing OpenTelemetry `user_prompt` event for a message sent while tool calls run. The event has `delivery.mode: \"mid_turn\"` and the running turn's `prompt.id`\n- Fixed the OpenTelemetry `user_prompt` event leaving out earlier messages' text when queued messages merged into one turn in VS Code, or when one message had an image\n- Fixed a slash command's `allowed-tools` grants applying to background agents woken during or after its turn\n- Fixed `/feedback` and `/bug` to upload other sessions only when they were recorded by the signed-in account or organization\n- Fixed relative deny and ask permission rules (e.g. `Read(./.env)`) no longer covering the project after a `cd`; they now also apply in added directories\n- Fixed deny and ask permission rules starting with `mcp__plugin_*_` missing a plugin's MCP server or tool spelled in another letter case\n- Fixed deny rules not being applied to a permission approval that arrives after Claude Code restarts, in self-hosted runner and `claude remote-control` sessions\n- Fixed a case where a committed `.claude/settings.local.json` could be treated as the user's own when Claude Code was started in certain subfolders of a repository\n- Fixed some deny and ask rules not applying to certain Bash commands\n- Fixed a gap in Bash permission checks where some compound commands could run without asking; these now ask for permission\n- Fixed a case where a Bash allow rule could run a different program than the one it names\n- Fixed a connector tool's approval prompt in cloud sessions being answered automatically while plugins were still loading\n- Fixed some subagents changing an artifact that people outside your organization can see, or uploading files to it, without asking first when an allow rule is set or uploads were approved earlier\n- Fixed some artifact publishes skipping the confirmation for pages that people outside your organization can see when \"don't ask again\" had been chosen for Artifact\n- Fixed auto mode judging only the first change to an artifact that people outside your organization can see; its yes now covers that change alone\n- Fixed a managed `strictPluginOnlyCustomization` lock on agents and skills not taking effect in sessions that were already running\n- Fixed sandboxed commands being able to write to the native updater's download folder in some cases\n- Fixed a case where the Bash sandbox did not fully protect a worktree's settings and hooks\n- Fixed a case where the Bash sandbox did not fully protect a linked worktree's shared git folder\n- Fixed `/teleport`'s stash choice in some unusual git states: it now stops and says what to do\n- Windows: Fixed a start-up stall that could last about a second when a PATH entry sits under a folder your account cannot open\n- Fixed `/resume` on Windows refusing or hiding sessions whose folder was recorded with a different drive-letter case or slashes\n- Fixed an in-session `/resume` on Windows dropping a project's saved trust when the session was recorded with a different drive-letter case\n- Linux: Fixed sandboxed Bash commands failing with `bwrap: Can't create file` when the working directory was below an `allowRead` folder inside a `denyRead`, such as under a symlinked home directory\n- Fixed Claude Desktop's third-party deployments staying on \"Stopping\" until a token renewal finished (once it includes this version)\n- Improved subagents cut off at the output token limit: they are now asked for the whole answer again, more briefly, instead of returning only its end\n- Improved startup with a keyless Console sign-in or other Anthropic profile login: a token that is still valid is read without waiting for the credentials lock\n- Improved the account email check: it now also refuses your email as a client's identity in scripts a command runs, file writes, web fetches and MCP calls\n- Improved the `HTTPS authentication failed` message for private plugin marketplaces: it now says git's terminal prompts and askpass programs are disabled, and it also appears with newer git versions\n- Improved the add-marketplace confirmations, plugin recommendation prompts and plugin install warning to say that plugins can run code on your computer, with access to your files and shell\n- Improved the `claude --desktop` sign-in error to name the API key, provider or setting that stops a Claude account sign-in from working\n- Improved the `/voice` \"No audio detected\" message on macOS and Windows: it now says where to pick your microphone\n- Improved MCP tasks: Claude is now told when a task is waiting for your answer, and can point you to `/tasks`\n- Improved `claude remote-control`: it now exits with status 4 when it stops by itself (single-session runs still exit 0) and 69 when it can't register, so a service manager can start it again\n- Improved the messages shown when a background session can't use its session's API key or token: they no longer suggest adding the key to settings\n- Improved the bundled dataviz skill's part-to-whole advice for static images and for parts with a natural order, such as age bands\n- Changed sandboxed Bash commands to follow the normal permission flow by default; set `sandbox.autoAllowBashIfSandboxed: true` to keep running them without prompts\n- Changed managed settings that turn the sandbox on: to keep commands running without prompts, set `sandbox.autoAllowBashIfSandboxed: true` in your highest-ranking managed source\n- Changed read-only git commands to ask first in a repository that isn't the project's or one you trusted; name it with `--add-dir`, not the `additionalDirectories` setting, to run them without prompts\n- Changed scheduled tasks and `/loop` to ask before running a skill that declares `allowed-tools` or hooks; add `Skill(name)` to `permissions.allow` to run it unattended\n- Changed \"Always allow\" in Desktop, IDE and SDK hosts: a rule alone is offered only when saving it would allow that command again, so a scheduled task that relied on one can now wait for you\n- Changed the sandbox to refuse sandboxed writes to `.claude` settings, skills, agents, commands and hooks in added directories, as it does in the launch folder\n- Changed worktree isolation: an isolated worktree session or agent is now refused `git bisect run`, `git maintenance register|start` and `scalar`'s set-up commands, and is told what still works\n- Changed plugin MCP bundles (`.mcpb`, `.dxt`) to unpack under the plugins directory, not `.mcpb-cache` in the plugin's folder, so read-only plugin folders load; one fetched by URL downloads again once\n- Changed plugins of a remote marketplace that managed settings declare, enable and list in `prependPlugins` or `appendPlugins` to count as the organization's; `pluginDigests` is deprecated and ignored\n- Changed plugin loading to skip, and `claude plugin validate` to report, some unsafe paths and links in a plugin's folder\n- Changed nested skill loading: `.claude/skills` folders in `--plugin-dir` plugins no longer load from a session started above them; start Claude Code in the plugin folder instead\n- Changed EnterWorktree to refuse a `WorktreeCreate` hook path that goes through a symlink inside the project; have the hook print the real path\n- Changed tmux and iTerm2 teammates, and new sessions from the agents view opened with ←, to start in the session's folder instead of where the shell last ran `cd`\n- Changed prompt suggestions in the terminal to refresh less often after 8 turns in a row started by background or scheduled tasks, saving usage until you press a key or return to the terminal\n- Changed Esc on a background MCP task's question to \"answer later\": nothing is sent to the server, and the question stays in `/tasks`\n- Changed `CLAUDE_CODE_MAX_CONTEXT_TOKENS` to also set the context window of unrecognized model IDs that start with `claude-`, such as a gateway's `claude-sonnet-latest`\n- Changed `/login` to ask for permission to change your privacy settings\n- Changed `/cost`, `$.session.usage()`, `--max-budget-usd` and the cost and token metrics to count a mod's `$.model.complete` and `$.model.classify` calls\n- Changed the programs that Claude Code's sandbox starts for its own use, such as its network bridge, to run without Claude Code's sign-in tokens and API keys in their environment\n- Changed ripgrep, local git commands and the sandbox runtime's own helper programs to run without the mTLS client key's passphrase in their environment\n- Remote Control: Changed `/resume` and `/teleport` to disconnect first when the connected session has a share link, or when that can't be checked; run `/remote-control` to connect again\n- Changed Claude Code on the web, cloud scheduled tasks and Cowork cloud sessions to start despite `sandbox.failIfUnavailable`: the hosted container has no command sandbox\n- Changed requests for settings, permission rules, hooks or status over a cloud or self-hosted session's connection to return less or be refused; what applies in the session is unchanged\n- Changed `claude -p` with no cached account data to wait for that data before its first request, so a first run in a clean environment can start a few seconds later\n- Changed Claude's instructions to keep files you'll want out of its temporary scratchpad folder, which can be cleared\n- Changed marketplace `ref` to accept the full ID of any commit on a branch or tag; clients before this version can't clone such a pin, so a marketplace pinned this way stops loading there\n- Changed plugin marketplace and plugin downloads to use a CONNECT tunnel when `ALL_PROXY` is the only proxy set; a proxy that refuses CONNECT now fails them\n- Changed `-p` and SDK runs in a worktree or submodule mounted into a container to apply `.claude/settings.local.json` allow rules only once the folder is trusted\n- Changed self-hosted runners to refuse a session whose session folder or repository path is a symbolic link, other than a `checkout` hook's checkout\n- Removed \"ask your admin to raise it\" from spend limit and shared budget errors in cloud sessions and under Remote Control\n- [VSCode] Added prompt suggestions wherever the terminal shows them: Tab accepts the suggested next prompt; `promptSuggestionEnabled: false` in Claude Code's settings.json turns them off\n- [VSCode] Added a Console sign-in that creates no API key\n- [VSCode] Added a field to paste the redirect URL when signing in to an MCP server from the panel in SSH, WSL and container windows\n- [VSCode] Fixed the model picker's check mark staying on the old model after Claude Code restarted on another model or the model was switched over Remote Control\n- [VSCode] Fixed the effort selector offering levels above the effort cap set by your organization or the `maxEffortLevel` setting\n- [VSCode] Fixed the sign-in screen offering sign-ins that the managed `allowedProviders` setting refuses\n- [VSCode] Fixed the permission prompt for MCP tools not saying why Claude Code asks when a safety check raises the question\n- [VSCode] Fixed the message for organization settings that couldn't load saying to run a command again; it now says to try again, matching the tab's Retry link\n- [VSCode] Improved the note shown for an untrusted folder: it now says why the extension is off and how to turn it on\n- [VSCode] Improved the error shown when the system refuses to run Claude Code: it now says your IT team can allow a blocked program\n- [VSCode] Improved the sign-in card to say what to do when the browser shows a connection error for localhost\n- [VSCode] Changed the install link's marketplace confirmation button to \"Trust marketplace and continue\"\n- [Claude Code on the web] Fixed new cloud sessions cutting MCP tool descriptions loaded through tool search at 4,096 characters; the limit is now 16,384, as in other Claude Code sessions\n- [Claude Tag] Fixed artifact links pasted in Slack getting no card when the poster's Claude sign-in had expired in a workspace their organization doesn't manage; a plain card now appears\n","publishedAt":"2026-11-13T20:40:01.000Z","fetchedAt":"2026-11-13T21:33:07.950Z","url":"https://code.claude.com/docs/en/changelog#2-1-325","media":[],"coverageCount":0},{"id":"rel_Bt_9gbwOwzJro_WQRoXV5","version":"v2.1.323","type":"feature","title":"v2.1.323","summary":"Code Mode is available as an experimental opt-in where Claude writes JavaScript to call its tools, and plugins can now declare an `engines[\"claude-code\"]` version range in `plugin.json` so installs pick the newest compatible version. The release also removes the `subagentStatusLine` setting and deprecates a plugin's `agent` setting with startup warnings, alongside a large batch of fixes to credentials, permissions, and background sessions.","titleGenerated":"Claude Code 2.1.323 adds experimental Code Mode and plugin engine version ranges","titleShort":"Code Mode ships as experimental opt-in; plugins get engine version ranges","breaking":"major","importance":4,"content":"## What's changed\n\n- Added Code Mode, an experimental opt-in in which Claude writes JavaScript to call its tools; see https://code.claude.com/docs/en/code-mode to turn it on, or to turn it off for an organization\n- Added `engines[\"claude-code\"]` to `plugin.json`: plugins outside their stated Claude Code version range are not installed, updated or loaded, and install and update pick the newest version that fits\n- Added type checking of a plugin's TypeScript hooks module to `claude plugin validate`; each finding is a warning, and `--strict` fails on it\n- Added to `claude plugin validate` the `$` nouns a mod's `engine.create` hook adds, withholds or replaces\n- Added support for `CLAUDE_CODE_AUTO_MODE_MODEL` in managed settings for Claude Desktop sessions on third-party providers (takes effect once Claude Desktop includes this version)\n- Added an options menu (stop, switch to usage credits, upgrade) when a model needs usage credits that are turned off or used up\n- Added a progress heartbeat every 30 seconds while hooks run long in `claude -p` and SDK sessions, so a host such as Claude Desktop won't end the turn as hung (once it includes this version)\n- Added `.docx` support to Artifact tool asset uploads\n- Added a `/diff to close` hint to the diff panel, and the keys that focus and close a panel to the `?` shortcut guide\n- Added a start-up warning for Bash allow rules with a `*` in a variable's value, such as `Bash(PORT=* npm run dev)`\n- Added a warning when a plugin's language server can't start, naming the server and what to check\n- Added a one-time startup notice naming `CLAUDE_CODE_PERFORCE_MODE` when Perforce settings are found and the mode is off\n- Added `CLAUDE_CODE_DISABLE_AGENT_PROGRESS_SUMMARIES` (available since 2.1.318): set it to `1` to turn off subagent progress summaries and the requests that write them (anthropics/claude-code#86879)\n- Added `CLAUDE_CODE_RESUME_HINT_COMMAND`, which a program that starts Claude Code can set to have the exit hint name its own resume command\n- Added `unsetOptionCount` to the `claude plugin install --json` result: how many `userConfig` options are not yet set\n- Added `isPromptCovered` to the props a mod's pane is drawn with, telling it when a dialog or a field above the prompt covers the prompt\n- Fixed running `/compact` again, or after a command such as `/context`, summarizing away a message that Claude had not answered yet\n- Fixed a resumed session joining your next message to an old interrupted message that was not resumed\n- Fixed a memory spike at the next prompt after Claude changed a very large file\n- Fixed a setting saved in a running session undoing a hand edit made to `settings.json` moments earlier; a settings file you delete during a session now stays deleted\n- Fixed `claude plugin` commands and `claude mcp list` signing you out of a Console profile when they ended while the token was being renewed\n- Fixed Edit and Bash running out of memory on certain `/proc` files, and Edit hanging or running out of memory on devices and named pipes\n- Fixed Claude Code failing to start, or to clear stored credentials at sign-out, when saving its configuration file failed with \"Out of memory\"\n- Fixed requests sent without streaming failing on errors that streamed requests recover from, such as a rejected image, thinking block or beta header\n- Fixed `ExitWorktree`, the worktree exit dialog and `/resume` leaving the session in the wrong folder when the folder the shell was in before entering the worktree had been deleted\n- Fixed accents and other combining marks sometimes landing left of the text, leaving stale copies on screen, or silently disappearing when text wraps\n- Fixed the last key of a key chord, such as Enter in ctrl+x enter, acting on a dialog's text field when the chord's own action isn't available there\n- Fixed `ps` processes piling up when the process listing Claude Code runs before stopping a command takes too long\n- Fixed subagent transcripts and large tool results being saved apart from the session's transcript after entering a worktree created by a `WorktreeCreate` hook or belonging to a nested repository\n- Fixed a session on a named Anthropic profile moving to another sign-in when `/fork`, the Left arrow or `--bg` continued it in the background; reopen any session that was open during the update\n- Fixed a background session sending a conversation under another sign-in when its named Anthropic profile is missing, changed or overridden; it now stops and says so\n- Fixed a session on an API key, auth token or workload identity from its shell (not a setup token) moving to another sign-in in the background; it now stops and says so\n- Fixed MCP tool calls ending in the sign-in server's error, such as `TOKEN_EXPIRED`, instead of asking you to sign in again\n- Fixed an expired claude.ai login showing \"Credit balance too low\" when a saved Console key or `ant auth login` profile without credit was used instead\n- Fixed a session on a named Anthropic profile being handed to an older background service that could not keep its sign-in; it now says to restart the service\n- Fixed a background session's sign-in leaving the previous account's login or other credentials in place; it now clears them for every session on the computer, and a refused sign-in can leave no login\n- Fixed background sessions moving between a profile and the stored login when an update restarted them\n- Fixed sessions opened before a sign-in keeping signed-out feature flags after the sign-in\n- Fixed sessions that kept failing after a Claude apps gateway sign-in expired: once you sign in again with `/login` in another terminal, they carry on at the next message\n- Fixed the billing notice naming your `ant auth login` organization when a custom `X-Api-Key` header is set\n- macOS: Fixed \"Not logged in · Please run /login\" coming back right after a successful `/login` or a token refresh when the keychain was slow to answer\n- Fixed turns failing in Claude Desktop when an organization's gateway answers an expired token with 403: Claude Code now asks Desktop for a fresh token and retries\n- Fixed the error shown when Amazon Bedrock refuses your account a model: it no longer says to refresh your AWS credentials, it says to enable the model\n- Fixed the `awsAuthRefresh` or `gcpAuthRefresh` command sometimes running twice for one request when three or more sessions needed a fresh sign-in at once\n- Fixed the `claude -p` warning about an ended claude.ai login to say that `claude auth logout` also removes saved MCP server logins and plugin secrets\n- Fixed Bash permission prompts offering \"don't ask again\" for commands that write to sensitive paths such as `.git` or `.claude`, where the saved rule did not stop the next prompt\n- Fixed a case where some `lsof` commands ran without a permission prompt\n- Fixed Bash commands that write a file with `cat > file <<EOF` being shown as \"Read 1 file\"\n- Fixed a case where Bash permission checks judged an `rm` from the wrong folder and skipped the dangerous-removal prompt under bypass permissions or a Bash allow rule\n- Fixed `--permission-mode auto` or a settings `defaultMode: auto` starting in a mode that could allow more when auto mode was unavailable. It now starts in Manual mode, which asks first\n- Fixed plan files sometimes being written outside the project when `plansDirectory` is set; a write that can't be shown to stay inside the project is now refused\n- Fixed a permission prompt when Claude read the output of its own background command or agent after entering a worktree\n- Fixed a case where Bash permission checks missed part of a command\n- Fixed a case where the Bash sandbox could switch off during a session\n- Fixed Bash allow rules with an attached wildcard, such as `Bash(ls*)`, skipping the permission prompt in some cases; a rule that names a variable, such as `Bash(NODE_ENV*)`, now asks\n- Fixed auto mode running behind a gateway at `ANTHROPIC_BASE_URL` on a model whose gateway name contains \"haiku\"; such a model, other than `claude-haiku-5-5`, is now refused, as Haiku 4.5 is\n- Fixed Bash deny and ask rules missing some commands run through `sudo`, `env` and similar wrappers\n- Fixed Bash allow rules written with combined `sudo` flags skipping the permission prompt in some cases\n- Fixed `sandbox.excludedCommands` letting a command that it does not name run outside the sandbox in some cases\n- Fixed a case where the Write and Edit tools could write to a path covered by a `Read` deny rule\n- Fixed a case where a Bash command ran without a permission prompt for zsh users with certain non-default shell options set\n- Fixed the Bash sandbox not fully protecting Claude Code's own config on installs still using the legacy config file\n- Fixed a permission prompt from a mod's `tool.check` hook naming every plugin hooked on the tool instead of the one that asked\n- Fixed sandboxed commands being able to read Claude Code's file checkpoint copies under the config directory\n- Fixed a rewind sometimes changing your permission mode\n- Fixed project settings sometimes being read through a symbolic link that leads to a different folder than the one Claude Code checked\n- Fixed HTTP hooks failing with HTTP 500 on Linux when `CLAUDE_CODE_SUBPROCESS_ENV_SCRUB` is set and `sandbox.enabled` is off\n- Fixed `claude -p` and the Agent SDK running Bash unsandboxed when `--settings` or SDK settings that turn the sandbox on fail validation; Claude Code now stops at start-up and names the invalid setting\n- Fixed `/export` silently stopping early after a long run of calls to an MCP tool that is no longer loaded\n- Fixed earlier MCP tool calls missing from a resumed conversation, and lines repeating in scrollback, when the server connects late\n- Fixed `claude mcp add --help` splitting long example commands across lines in narrow terminals and pipes, so a pasted example ran without its header\n- Fixed plugins with `defaultEnabled: false` staying off in a claude.ai chat's cloud session after you turn them on in claude.ai\n- Fixed `/fast on` and `/fast off` on Vertex not asking a plugin's `config.set` hook, as the `/fast` menu does\n- Fixed documents created through the Claude Docs connector not opening in your browser when you work in a terminal\n- Fixed the misleading \"not cached\" error for a plugin only a repository's settings enable: it now says it isn't installed, gives the install command, and shows in text and stream-json `-p` runs\n- Fixed mods being able to reword Claude Code's own confirmation questions\n- Fixed a mod's hot reload sometimes ignoring a file created just after a failed reload\n- Fixed marketplaces named `npm`, `pip`, `uv`, `cargo`, `github` or `gh` being refused as reserved names\n- Fixed a stopped MCP call, or a stopped shell command in fullscreen, reading as \"Called\" or \"Ran\" in a collapsed tool group: it now says no result was recorded and shows its input\n- Fixed load errors of plugins enabled by managed settings going only to the debug log in SDK and Desktop sessions: `init.plugin_errors` now lists them\n- Fixed plugin installs hiding what a plugin's install command wrote to stderr when that command succeeded, such as a note that the plugin has no skills yet\n- Fixed one malformed entry in a plugin marketplace's `renames` map making Claude Code ignore the marketplace's other renames\n- Fixed `claude --bg --safe-mode` sessions stopping at startup to ask about project MCP servers that safe mode never starts\n- Fixed MCP link prompts offering Accept for non-http links without saying the link wouldn't open\n- Fixed `/mcp` and `/plugins` opening an MCP server's page with the cursor on \"Disable\" when the server needs authentication but still holds its sign-in; the page now offers \"Reconnect\" and opens on it\n- Fixed plugins with a version list in `enabledPlugins` staying off with no message: `/plugin` and `claude plugin list` now say why\n- Fixed plugins and marketplaces failing to install on Windows when the path of a cloned file reaches 260 characters\n- Fixed a Stop pressed right after sending a message showing as an error in remote and SDK sessions while a `UserPromptSubmit` or plugin prompt hook was still running\n- Fixed an MCP server staying disconnected with no tools after an Agent SDK app or `claude -p` caller removed and re-added it, when a tool call had tried to reach it in between\n- Fixed text-output `claude -p` runs printing nothing to stderr when a `--plugin-dir` mod's hook hangs or crashes the hooks worker\n- Fixed SDK sessions giving an MCP server named by `--allowedTools` or an `mcp_tool` hook less than 2s before the first turn when settings lowered `CLAUDE_CODE_MCP_STARTUP_WAIT_MS` after start-up\n- Fixed headless sessions ending when an SDK host sent a null entry in an MCP server's tools list\n- Fixed WebFetch dropping empty table cells and shifting later values a column left\n- Fixed temporary files of a session that had changed directory being left behind when the session expired under `cleanupPeriodDays`\n- Fixed managed settings (`managed-settings.json`, `managed-settings.d`) sometimes being missed at start-up while another program was rewriting the file: start-up now reads such a file a second time\n- Fixed `.claude/settings.local.json` not being ignored globally when the folder became a git repository later, or on a new machine\n- Fixed `/diff` waiting for Claude to finish before closing the diff panel in fullscreen\n- Fixed `/memory` showing wrong or roundabout paths for folders whose names start with the working directory's or home folder's name, and losing slashes in the file path on Windows\n- Fixed a broken emoji character in a tool call's parameter name being resent with later requests, which strict JSON parsers reject\n- Fixed a collapsed tool call looking as if it ran after you rejected it with a reason: it now shows \"Not run · rejected\"\n- Fixed a case where part of a saved session could be lost on NFS or FUSE file systems when Claude Code discarded a reply\n- Fixed WebFetch not rechecking organization policy and `CLAUDE_CODE_DISABLE_WEB_FETCH` at each call\n- Fixed file titles, app names and session names in @-mentions from claude.ai and the desktop app sometimes being treated as text you typed\n- Fixed a multi-row status line being pushed off screen by a long prompt in the fullscreen renderer\n- Fixed a case where certificate errors behind a TLS-inspecting proxy persisted when the proxy itself is reached over `https://`\n- Fixed Pro and Max accounts being offered feedback drafts that their organization would refuse\n- Fixed Claude in Chrome's policy message telling people to contact their admin when no admin setting can turn it on\n- Fixed the artifact preview sometimes opening a different artifact than the one Claude just published\n- Fixed certificate name-constraint errors being retried for minutes; they now fail at once with a message that says what to fix\n- Fixed `claude --teleport` opening an empty session when the server could not find the session's history, and `claude -p --teleport` exiting without an error when the resume failed\n- Fixed a misspelled keybinding modifier, such as `ctl+k`, binding the bare key; the binding is now ignored\n- Fixed `/btw` answers and recaps sometimes including earlier replies from the conversation\n- Fixed start-up requests switching from an organization's managed proxy and CA settings to user-set ones when a read of the managed settings file failed\n- Fixed Claude giving a wrong reason when a new message interrupted a file it was sending you\n- Fixed `CLAUDE_CODE_RESUME_INTERRUPTED_TURN` re-running a turn that a hook's `continue: false` had stopped\n- Fixed Claude being told that automatic model switching is off when safeguards stopped a subagent whose model had no other model to switch to\n- Fixed the changed-files list of remote sessions leaving out files added with `git add -N` before the first commit\n- Fixed some Claude in Chrome tool calls, such as `browser_batch`, not showing the tool's name in the fullscreen transcript (ctrl+o) or in a subagent's progress\n- Fixed subagent progress summaries skipping an update when the reply took between 10 and 20 seconds\n- Fixed `claude doctor` and `/status` saying a linked managed settings file is still read when it could not be read\n- Fixed an interactive `--chrome` session giving no reason for starting without Claude in Chrome when your organization's policy turns it off: a line now says so\n- Fixed the `/rewind` list on network paths and hung folders: it named every tracked file as changed, was slow to show labels, and showed none at all when a file read never returned\n- Fixed `--debug` and `--debug-file` replacing a file named `latest` in the debug log's folder\n- Fixed a timed-out feedback send being reported as too large; it now says the request timed out\n- Fixed screen reader mode cutting off an announcement when the system clock moved forward while it was being read\n- Fixed screen reader mode dropping dictated or pasted text, and emoji, typed into an answer to a question or permission prompt\n- Fixed the name label under permission and plan dialogs hiding part of some agent names\n- Fixed a case where pressing Ctrl+C just as `claude update` finished removed the `claude` command\n- Fixed the fullscreen header naming the session's effort level for a viewed subagent on a different model, which that subagent doesn't use\n- Fixed the model picker and `/model` showing a picked effort level that `CLAUDE_CODE_EFFORT_LEVEL` overrides\n- Fixed the tip after a slow worktree creation suggesting a `sparsePaths` list without `.claude`, which left worktrees without your project settings and rules\n- Fixed the newest model's highlight missing from the `/model` picker\n- Fixed `/upgrade` and Ctrl+clicked links in fullscreen failing silently when no browser can open: the address is now shown\n- Fixed the Claude apps gateway's JSON logs escaping newlines and other control characters in a message twice\n- Fixed terminal control sequences in a subagent's message taking effect instead of showing as plain text when the message is expanded\n- Fixed Claude in Chrome setup saying `/chrome` finishes the connection; it now tells you to restart with `claude --chrome`\n- Fixed the Option-key hint appearing while typing ø or π inside words on Danish, Norwegian, Faroese and Greek keyboard layouts\n- Fixed a Claude Desktop session getting stuck as busy after another of your sessions sent it the same message twice or many messages at once\n- Fixed long tool results staying on disk after a session run with `--no-session-persistence`; they are now removed on normal exit, Ctrl+C or SIGTERM\n- Fixed `PreCompact`, `PostCompact` and post-compaction `SessionStart` hooks not receiving `agent_id` and `agent_type` when a subagent compacts\n- Fixed `CLAUDE_CODE_TRANSCRIPT_LOCAL_GC` dropping messages from the saved transcript when another process rewrote the file while it was being trimmed\n- Fixed the \"Now using usage credits\" notice repeating or not showing past the weekly Fable limit when Claude Code made a quick side request on a smaller model, such as naming the session\n- Fixed Fable's weekly limit showing no reset time and no options menu in the terminal\n- Fixed a wait chosen in the usage limit menu silently moving to a later reset when its request was refused; the menu now opens\n- Fixed the spend limit menu offering a lower limit when your current limit couldn't be loaded\n- Fixed `/usage` hiding limit bars for setup tokens and `CLAUDE_CODE_OAUTH_TOKEN`: it keeps the session and weekly bars and says why the rest are missing\n- Fixed `/usage` showing a whole-session percentage beside a subagent type's name, and saying sessions ran 8 hours when it only counts hours with requests\n- Remote Control: Fixed sessions showing as disconnected for a minute or more after a network change such as switching Wi-Fi or reconnecting a VPN\n- Fixed Remote Control showing nothing while Claude Code asks whether to use usage credits: the question now also shows in the Claude app\n- Fixed Remote Control showing a session as running while it waited on a permission prompt that can only be answered in the terminal\n- Remote Control: Fixed `claude remote-control` reconnecting as a new computer when restarted in the same folder more than 4 hours after it stopped; it now keeps its identity for up to 30 days\n- Remote Control: Fixed sessions left running at shutdown, and finished sessions holding their slot, when the disk stops responding\n- Remote Control: Fixed other devices and phone notifications not showing what Claude is asking permission for when `DISABLE_TELEMETRY` or `DO_NOT_TRACK` is set\n- Remote Control: Fixed the host exiting, or stopping reading a session's output, in verbose or debug mode when a tool call's `command` was a number, boolean or object\n- Remote Control: Fixed a host that could stall after sessions failed to start in folders on a filesystem that had stopped responding\n- Remote Control: Fixed sessions from the desktop app showing as idle on your phone while background subagents were still running\n- Fixed the Agent SDK's `managedSettings` ignoring the Remote Control, agent view, Workflows and Artifact off switches\n- Fixed OpenTelemetry export failing silently when an `otlp` exporter had an endpoint but no `OTEL_EXPORTER_OTLP_PROTOCOL`: it now defaults to `http/protobuf`\n- Fixed OpenTelemetry export to the default `localhost:4318` endpoint going through a configured proxy\n- Fixed one malformed `OTEL_RESOURCE_ATTRIBUTES` entry dropping all of that variable's attributes from the OpenTelemetry resource; the others are now kept, and `/status` names the one left off\n- Fixed OpenTelemetry `user_prompt` events and interaction spans recording a background agent's reply when `OTEL_LOG_ASSISTANT_RESPONSES=0`\n- Fixed OpenTelemetry exports failing when a proxy took over 10 seconds to reach the collector; it now has 25 seconds, also the most that `OTEL_EXPORTER_OTLP_TIMEOUT` can allow (previously 30)\n- Fixed OpenTelemetry `agent.name` showing a user-defined subagent's name instead of `custom` when an official plugin's skill launched it\n- Fixed `claude agents` showing a status from an earlier turn after a session resumes\n- Fixed read-only git commands needing approval, or being refused under `claude -p`, when `CLAUDE_CODE_SUBPROCESS_ENV_SCRUB` is set on Linux\n- Fixed `claude agents` losing a typed prompt: Esc now needs a second press to clear it, and ctrl+y restores it\n- Fixed the move-to-background key doing nothing in `claude attach` and `claude agents` sessions; to detach with Ctrl+B d, press d within 800 ms of Ctrl+B\n- Fixed background sessions ignoring a `DISABLE_ERROR_REPORTING` set only in your shell when the background service started without it\n- Fixed false \"needs input\" notifications in `claude agents` after replying to a blocked session, and repeated ones after a background turn ended with no text and only basic tools such as Bash or Read\n- Fixed `claude -p` silently using another model in text output when settings block the requested one; it now warns on stderr\n- Fixed `claude -p` printing notes about `deniedModels` and `availableModels` entries on every run; `/status` and `claude doctor` still list them\n- Windows: Fixed some file-reading Bash commands running without a permission prompt when given a wildcard or a list of files\n- Windows: Fixed Claude Code running a `.cmd` or `.bat` wrapper found beside a program's `.exe` in the same PATH folder (a `git.cmd` next to `git.exe`, say); it now runs the `.exe`, as `cmd.exe` would\n- Windows: Fixed shell commands failing with \"Working directory was deleted\" on drives where a folder's real path can't be resolved\n- Fixed a cloud session's history becoming unreadable if its machine stopped just after the history was rewritten\n- Fixed scheduled tasks from an ended session sometimes not running after another session in the same folder used `/clear`, `/resume` or `/branch`\n- Fixed an approved plan being answered with \"You are not in plan mode\" when the approval arrived after the session's process had restarted, in cloud, self-hosted runner and SDK sessions\n- Cloud sessions: Fixed an answer to Claude's question being lost after a restart in a long conversation\n- Fixed a message dropping out of a cloud session's conversation when the session restarted before Claude replied\n- Cloud sessions: Fixed Claude saying an attached image is missing when the message names its file\n- Fixed a resumed cloud session making a model call nobody asked for when files arrived right after a chat's first reply\n- Fixed streamed replies in cloud sessions missing text from the middle when the session's container shut down partway through a reply\n- Fixed skill files sent from Cowork cloud sessions sometimes failing to upload on the first try\n- Self-hosted runner: Reverted the exit code of an on-demand runner whose work order was superseded to 1; 2.1.321 and 2.1.322 had changed it to 0\n- Improved sign-in errors in `/login` and `claude setup-token` for a rejected, unreachable or timed-out request: they now say what happened and what to check\n- Improved Bash permission checks so the same command more reliably gets the same answer on a slow or busy machine\n- Improved Bash permission checks: some commands that could run differently from how they were checked now ask for permission\n- Improved `/mcp` to show how many of a server's tools were left out because the API would reject their input schema, and why\n- Improved the \"Not now\" choice when Claude Code asks to turn on the mods it writes in a session: it now says when they load\n- Improved the error for an MCP tool call whose connector ID is mistyped by a character or two: for a server that the host application registered, it now names the tool to call\n- Improved Claude Desktop and cloud sessions for claude.ai accounts with more than 2,000 plugins turned on: a shorter wait at startup, and Claude can now say why plugins are missing\n- Improved memory use with many subagents: a finished subagent's transcript is freed sooner instead of staying in memory while later subagents run\n- Improved installing with the install script: the latest version is no longer downloaded twice\n- Improved WebFetch speed and memory use on deeply nested pages\n- Improved the `cleanupPeriodDays` cleanup to count folders it can't read as errors\n- Improved the Claude apps gateway's `/v1/models` to state `max_input_tokens` for models whose context window is certain on every upstream\n- Improved `ToolSearch` to say when a tool Claude searched for by its short name is already available, instead of only reporting no match\n- Improved start-up on Linux when `/home` is an autofs map: Claude Code no longer looks up files in `/home` itself. Set `CLAUDE_CODE_AUTOMOUNT_ROOTS` for a map root deeper than one folder below `/`\n- Improved the Fable limit message in `-p`, SDK, IDE and cloud sessions: it now says when the limit resets, along with how to continue\n- Changed `/security-review` to ask before Claude runs it, and skills' inline git commands to stop in bare-repository worktrees and similar folders until you allow them, outside auto mode and bypass\n- Changed how many messages each of your sessions on the same machine can send the others without a prompt from you, from about 20 to 100,000\n- 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 20 to 100,000\n- Changed organization settings and policy requests to no longer use a proxy set only by `ALL_PROXY`; set `HTTPS_PROXY` if your network needs a proxy for them\n- Changed the context window assumed for a model ID that Claude Code doesn't recognize from 200K to 1M; use `/autocompact` to compact earlier, or `CLAUDE_CODE_DISABLE_1M_CONTEXT=1` to keep 200K\n- Changed sessions on other vendors' models with a smaller window: once full, they can fail every turn until `/clear`, so set `CLAUDE_CODE_DISABLE_1M_CONTEXT=1` before they fill\n- Changed Claude Code's own ripgrep and most internal git commands to run without most of its sign-in tokens and API keys, so a credential helper or hook under such a git command no longer finds them\n- Changed `MCP_SDK_GENERATION=v1` to run the v2 MCP client in a compatibility mode (earlier handshake, relaxed OAuth sign-in checks), because the v1 MCP client is removed\n- Changed the tool Claude uses to finish an MCP server sign-in so that it asks you to confirm, even in bypass and auto modes; where nobody can be asked, such as in `dontAsk` mode, use `/mcp`\n- Self-hosted runner: Changed a session at a connector's approval prompt or sign-in inside a running tool to be released as paused at the idle or lifetime limit; a late answer resumes from a fresh clone\n- Changed artifact updates to ask first, even under an allow rule, when people outside your organization can see the artifact or Claude can't check who can\n- Changed `glab config set` without `--global` or `--host`, and `glab config edit --local`, to fail read-only inside the Bash sandbox; run them outside it\n- Changed the Bash sandbox on macOS, Linux and WSL to ignore write access granted through a symlink in a writable folder that leads elsewhere or to a missing path; grant the real path instead\n- Changed \"don't ask again\" so an allow rule that looks like it holds a secret, or would be saved to a `settings.local.json` that git tracks, lasts for the session only and isn't saved\n- Changed the sandbox on macOS and Linux: `git worktree add`, `move` and `remove` now fail read-only, as do commits in a worktree the session wasn't started in or given, even one inside the project\n- Changed start-up to stop when a `--settings` file isn't valid JSON and has a `sandbox` key, even one that turns the sandbox off\n- Changed what happens when a tool's permission check fails with an error after a `PreToolUse` hook returned \"allow\": Claude Code now decides the call as if no hook had answered\n- Changed what happens when a tool's permission check fails on input that a `PermissionRequest` hook rewrote: Claude Code now asks again, or refuses the call where nobody can answer\n- Changed the GitLab merge request badge to show only for the repository a sandboxed session opened in\n- Changed the `claude mcp add --help` example to a local command\n- Changed mods: a `*`, namespace or negated hook on `engine.create` now only observes. To add or withhold a `$` noun, hook `engine.create` by name\n- Changed `claude plugin validate` to fail a mod whose `engine.create` hook isn't written in the one accepted form; installed mods keep loading\n- Changed marketplace and plugin installs to fail when a git submodule isn't on the commit its repository records\n- Changed project (`.mcp.json`) MCP server approvals to match the server's exact name; an approval entry written in another spelling now asks again\n- Changed self-hosted runners under `--trust-workspace false` and VS Code sessions in an untrusted folder: the repository's settings can no longer disable your hooks or turn plugins on or off\n- Changed `tool.list` plugin hooks so the tools they add or remove apply to the model and every agent\n- Changed `claude plugin install` and `/plugin` to show a warning instead of success when a plugin's install command writes to stderr\n- Changed claude.ai chats and Cowork tasks to check a plugin file for secret keys before sending it, and to ask first if it may hold one\n- Changed `claude rm` and the agents view to say a session is removed from the list, not deleted; `claude rm` help says the conversation is kept and `claude --resume` lists it\n- Changed `/stickers` to say that the sticker pack is all claimed for now, instead of opening a page in the browser\n- Changed a managed settings `policyHelper` to no longer take environment variables from user or `--settings` files; set what it needs in managed `env` or the launching shell\n- Changed a `/skill` or `@file` typed between two files picked in claude.ai or the desktop app, or after a picked file and a link, to no longer take effect\n- Changed split-pane agent teammates to check that requests come from their team lead: one that outlives its lead has its asks refused, and plan approvals must be re-sent after `/clear` or `/resume`\n- Changed worktree creation to refuse checkouts whose `.git` folders are links or hold hard links, such as git-new-workdir layouts and `cp -al` copies; use a plain clone\n- Changed what Claude is told after safeguards stop a forked skill or a workflow agent: retry with a narrower prompt or tell you, instead of doing the work itself\n- Changed browser tool calls (Claude in Chrome, the desktop browser pane) to refuse some escaped or JSON-text inputs and tell Claude how to respell them\n- Changed workflows so that a prompt an app marks as not typed no longer reaches their agents as your request\n- Changed computer use on macOS to refuse key chords that hold Space with Control or Option, and more system shortcuts that need your grant\n- Changed first-run setup in Terminal.app to name the \"Use Option as Meta key\" setting, what Option stops typing, and how to undo it before you agree\n- Changed the maximum of `CLAUDE_CODE_OVERLOADED_RETRY_MAX_DELAY_MS` from 64000 to 300000\n- Changed workflows in the desktop app's Code tab to pause at a usage limit and continue after it resets\n- Changed the spend limit and 1M context errors, and `/usage-credits`, in cloud sessions and under Remote Control: they no longer give a web address\n- Changed which MCP servers are reported by name in OpenTelemetry exports to follow the current connector directory\n- Changed `/logout` in a background session, which had no effect: it now signs out every session on the computer, deletes stored MCP logins and plugin secrets, and ends the session, as in a terminal\n- macOS: Changed the Bash sandbox so `lsof` and similar tools may show less about processes a command did not start\n- Changed cloud scheduled tasks: a `PostToolUse` hook from your own settings can no longer replace the plan in `ExitPlanMode`'s result\n- Changed `CLAUDE_AUTOCOMPACT_PCT_OVERRIDE` in Anthropic-hosted cloud sessions to stop applying to auto-memory extraction and, in some sessions, to `/btw`; the main conversation still follows it\n- Removed the `subagentStatusLine` setting: subagent rows always use the default look, and settings files that still hold the key load as before\n- Deprecated a plugin's `agent` setting: it still applies, and now warns at startup, in `claude plugin list` and in `claude plugin validate`\n- [VSCode] Added a Claude Code trust prompt before the first session in a folder you haven't trusted; declining leaves the folder's settings and MCP servers unused\n- [VSCode] Added the live output of a running Bash or PowerShell command to the chat, under the command's card\n- [VSCode] Fixed a response that was discarded because its tool call could not be parsed still showing after Claude retried\n- [VSCode] Fixed a queued message missing from a conversation or fork when it directly followed a compaction summary\n- [VSCode] Fixed the \"Buy usage credits on claude.ai\" button counting as agreement to pay for Fable while usage credits were off\n- [VSCode] Fixed the \"extensions want to relaunch the terminal\" warning coming back on open terminals after every window reload or restart\n- [VSCode] Fixed \"Couldn't load this conversation's saved history\" appearing in a new chat whose Claude process had restarted before the first prompt\n- [VSCode] Fixed the sign-in screen showing the login choices again after a pasted code; it now shows \"Signing in…\" until the sign-in finishes\n- [VSCode] Fixed the \"You've hit your Fable limit\" line staying in the VS Code and JetBrains panels after you pick another model\n- [VSCode] Fixed the Fable usage warning in the VS Code and JetBrains panels naming usage credits when they're off or used up\n- [VSCode] Fixed the diff tab not opening for multi-line edits on files with CRLF line endings\n- [VSCode] Fixed dictation continuing to record after switching sessions\n- [VSCode] Improved sign-in errors behind TLS-inspecting proxies: a certificate failure now says what to do, as the terminal does\n- [VSCode] Changed opening a Claude Code on the web session in a folder where Claude Code's sandbox is on: the extension no longer switches to the session's branch for you, and the dialog says why\n- [Claude Code on the web] Improved the message when a file fails to upload to a cloud or Remote Control session: it now gives the reason, such as too many pages, a password or a time-out\n- [Claude Code on the web] Improved adding a repository Claude can't access to a cloud session: a short message and a Reconnect GitHub button replace the long error\n- [Claude Code on the web] Fixed self-hosted sessions staying on \"Starting Claude Code\" after it had started, when you reopened a paused session without sending a message\n- [Claude Code on the web] Fixed a paused session in a self-hosted environment sometimes still showing \"Sending\" and a Stop button under the \"Session paused\" banner\n- [Claude Code on the web] Fixed pull request updates from GitHub Enterprise Server linking to github.com in cloud sessions; new updates link to your own server\n- [Claude Code on the web] Improved the message when some network secrets can't be added to a new cloud environment: it lists each host, says why where it can, and stays until you dismiss it\n- [Claude Code on the web] Fixed very large GitHub Actions job logs and run log archives failing to download or being cut off in cloud sessions\n- [Claude Code on the web] Improved the new-session repository picker when GitHub single sign-on hides organizations: it now says so even with other repositories listed, and links to reconnect GitHub\n- [Claude Code on the web] Fixed new cloud sessions started from the iOS app with a repository not getting a branch of their own\n- [Claude Code on the web] Fixed an open cloud or Remote Control session's page staying \"Disconnected\" after a few minutes of failed connections; it now tries to reconnect by itself\n- [Claude Code on the web] Fixed Try again on an open cloud or Remote Control session's \"Connection lost\" card sending \"Try again\" as your message; it now tries to reconnect instead\n- [Claude Code on the web] Fixed the Stop button disappearing from a cloud session five minutes into a reply that had shown nothing yet while Claude was still working\n- [Claude Code on the web] Fixed an approval prompt sometimes disappearing from a tab left open after a self-hosted runner released the session while it waited for your answer\n- [Claude Code on the web] Added a failure notification when a scheduled task's run can't start because its settings require a sandbox that isn't available where it runs\n- [Claude Code on the web] Improved the Git providers page when your GitHub sign-in has expired: it now asks you to reconnect instead of silently hiding organizations you can link\n- [Claude Code on the web] Fixed removing a repository from a scheduled task silently moving its GitHub trigger to another repository; Save now waits until you pick one\n- [Claude Code on the web] Fixed git clones and fetches that take longer than ten minutes in a cloud session sometimes being cut off at the very end\n- [Claude Code on the web] Added the No repository option to the repository picker for new sessions on a self-hosted environment\n- [Claude Code on the web] Added a warning to a cloud session's Delete dialog when deleting it will pause scheduled tasks that run in that session\n- [Remote Control] Fixed your own session saying its environment was deleted when you deleted nothing; it now says your computer is no longer connected\n- [Remote Control] Fixed some sessions still showing as running in the session list after the computer disconnected before Claude started; they now show an error mark or as idle\n- [Remote Control] Added a Try again button to the error shown when the computer running your session is signed out of Claude Code\n- [Claude Tag] Added Claude Tag sessions started in Slack to the Compliance API on Enterprise plans, in beta: list the sessions and read their transcripts\n- [Claude Tag] Added a confirmation step when an admin turns off Claude Tag for the organization, with a dialog that says what stops\n- [Claude Tag] Fixed scheduled tasks in a Slack channel moved between Enterprise Grid workspaces failing to use Slack until someone posted there\n- [Claude Tag] Fixed the Approve and post card in a DM with Claude appearing for private or Slack Connect channels, where Claude can't post; Claude now gets the reason and shows no card\n- [Claude Tag] Added a Retry button to a Slack channel's Configure page for when its settings fail to load\n- [Claude Tag] Added a Scope filter to the Network events tab of Claude Tag's Activity page, so admins can download one scope's events instead of the whole organization's\n- [Claude Tag] Fixed Remove connector in Claude Tag admin settings being refused when access bundles also held its credentials; it now removes those set on the connector and leaves the bundles' own\n- [Claude Tag] Fixed the Channels tab in Claude Tag admin settings splitting a workspace from its channels across pages; channels now appear right under their workspace\n- [Claude Tag] Fixed the Set for Claude Tag link on the Cloud environments admin page not leading to the environment setting; it now opens the Slack page's Advanced tab, where you set it\n- [Code Review] Added a Download CSV button to Code Review admin settings: export review costs per repository by month, week or day, for one repository, one GitHub organization or all\n","publishedAt":"2026-11-12T22:12:13.000Z","fetchedAt":"2026-11-13T01:19:29.681Z","url":"https://code.claude.com/docs/en/changelog#2-1-323","media":[],"coverageCount":0},{"id":"rel_OG9-TdKFhSsrl4sGQODME","version":"v2.1.322","type":"feature","title":"v2.1.322","summary":"Fast mode is available on Vertex in the global, us, and eu regions, falling back to standard speed if Vertex refuses it.","titleGenerated":"Claude Code 2.1.322 adds fast mode on Vertex in global, US, and EU regions","titleShort":"Fast mode now available on Vertex","breaking":"none","importance":2,"content":"## What's changed\n\n- Added fast mode on Vertex in the global, us and eu regions, falling back to standard speed if Vertex refuses it\n","publishedAt":"2026-11-10T02:29:43.000Z","fetchedAt":"2026-11-10T04:33:02.487Z","url":"https://code.claude.com/docs/en/changelog#2-1-322","media":[],"coverageCount":0},{"id":"rel_Ls8uVUyDJFEXbMguUWLEP","version":"v2.1.321","type":"feature","title":"v2.1.321","summary":"MCP OAuth requests no longer send secrets such as refresh tokens to another host when a server redirects them, and a read-only Bash command can no longer run outside the sandbox without a permission prompt. Claude Fable 5.2 is now the default Fable model with 1M context, and saved Fable 5.1 model picks move to fable.","titleGenerated":"Claude Code v2.1.321 fixes MCP OAuth token leak and adds Claude Fable 5.2","titleShort":"MCP OAuth no longer leaks refresh tokens; Fable 5.2 default","breaking":"major","importance":4,"content":"## What's changed\n\n- Added Claude Fable 5.2 (`claude-fable-5-2`), now the default Fable model — 1M context, $8/$40 per Mtok with $0.20/Mtok cache reads\n- 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`\n- Fixed MCP OAuth requests sending secrets such as refresh tokens to another host when a server redirected them\n- Fixed a case where a read-only Bash command ran outside the sandbox without a permission prompt\n- Fixed a case where Bash permission checks judged a file write from the wrong folder and didn't prompt\n- Fixed deny rules and protected-file prompts not applying to the file that a `cp` or `mv` into a directory writes\n- Fixed the \"Draws from usage credits\" note showing after a switch to Fable on Enterprise seats whose plan includes Fable\n- Fixed the fallback model sometimes declining a request or mentioning a stopped response when the first model was blocked before replying\n- Fixed `claude mcp login` printing a SEP-2352 issuer warning each time you sign in again to a server that Claude Code registered with\n- 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)\n- 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)\n- 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\n- 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\n- Fixed `--plugin-dir .` and `--plugin-dir ./builtin` writing generated type files for Claude Code's built-in plugins into a `builtin` folder\n- Fixed context usage estimates overcounting tool definitions when exact token counts are unavailable\n- Fixed `/compact` showing twice in Remote Control on web and mobile\n- Fixed plugin auto-updates and `/plugin` marketplace updates downloading a plugin again for each worktree or folder of a repository\n- Fixed auto mode losing a background subagent's report that had started a turn when the conversation was compacted before Claude answered it\n- 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\n- Fixed `claude -p` sometimes telling Claude that background commands get 30 minutes when they are stopped after 10\n- Fixed `pkill`, `pgrep` and `pidwait` in cloud sessions matching the session's own parent processes\n- Fixed Claude in Chrome and computer use server processes staying alive after their parent exits behind a proxy that stalls connections\n- Fixed a large hook output pointing Claude to a file that held another hook's output from the same tool call\n- Fixed apps that embed Claude Code getting their own PreToolUse hook's `ask` reason instead of another hook's when both asked\n- Fixed a PreToolUse hook's `ask` being lost when another hook's `defer` finished first and could not be honored\n- Fixed a rare crash on permission prompts and background tasks\n- Fixed project skills, commands and agents disappearing from the menu in SDK and desktop sessions after Claude entered a worktree\n- Fixed `/desktop` on macOS saying Claude Desktop is not installed when it is in `~/Applications`\n- Fixed OpenTelemetry repository attributes (`vcs.*`) missing from Claude Desktop sessions that start on a pre-started process\n- 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\n- Fixed screen reader mode delaying the prompt at startup when the system clock moved back while Claude Code was starting\n- Fixed a command's panel sometimes not coming back after a dialog closed, leaving no panel or prompt on screen until the next redraw\n- 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\n- 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\n- Fixed high memory use in long sessions that repeatedly read single lines from large files\n- 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\n- Fixed one request carrying far more background agent reports than intended after Claude was stopped while several were waiting\n- Fixed messages from a Claude Desktop Code tab session to your other sessions staying paused after you typed while Claude was still working\n- 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\n- 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\n- Fixed a Remote Control message sent with \"Queue for later\" reaching Claude in the middle of the running turn\n- Fixed the git status given to Claude at session start reading as clean when git status failed; it now reads as unavailable\n- Fixed the git status given to Claude at session start showing the first listed file as staged when its changes were not staged\n- Fixed a hang or out-of-memory crash when a connector in a cloud session lists resources with very long names\n- Fixed Remote Control showing as connected after its connection closed for good while the session's history was still uploading\n- Fixed screen reader mode turning off when you return to a conversation from agent view\n- Fixed screen reader mode y/n questions taking as yes a stray \"y\" at the end of text typed as the question opened\n- 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\n- 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\n- Fixed Bash commands sometimes running in the wrong folder when Claude sent several at once and an earlier one changed folders\n- 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\n- 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\n- Fixed sign-in, token refresh and logout waiting indefinitely when something on the network accepts the connection but never answers the TLS handshake\n- 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\n- Fixed sub-agent progress summaries running a request to its end after the model replied with a tool call\n- 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\n- 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\n- Fixed Bash commands coming back empty for zsh users whose `.zshrc` sets `KSH_OPTION_PRINT`\n- 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\n- Fixed agent view erasing another agent view's saved input text when it restarted for an update with nothing typed\n- 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\n- Fixed `claude remote-control` accepting a `y` typed without Enter at its first-run question when the terminal's input ended\n- Fixed Claude Desktop not showing spend limit warnings behind a Claude apps gateway whose policy has no `code:` rules\n- 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\n- Fixed prompt suggestions making extra API requests when the model replied with a tool call\n- Fixed `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\n- Fixed skill and command names holding a Bamum letter that some fonts draw as blank: they are now refused\n- 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\n- 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\n- Fixed NotebookEdit reporting success to hooks and telemetry when the edit failed, for example when the notebook changed while the permission prompt was open\n- Fixed a blank screen or no answer at startup, and empty `claude mcp list` output, when a stored MCP sign-in record was null\n- 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\n- Fixed a restarted cloud session dropping a message with a large image when the restart came before Claude's first reply to it\n- Fixed a restarted cloud session leaving a subagent's message to the main conversation unanswered\n- Fixed Claude not being told when you stop or pause a running workflow, so it no longer says the workflow is still running\n- Improved the `/model` picker: \"More models\" is now an unnumbered row below every other, and older models open under it without renumbering the rest\n- Gateway: Improved spend limits on a busy replica, which now skips fewer checks and records spend sooner\n- 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\n- Improved `claude install` and the install scripts: they finish 2 to 4 seconds sooner\n- Improved how soon a Remote Control session whose process crashed can pick up a new message\n- Windows: Improved startup and first-command time by finding programs on PATH without starting a helper process\n- Improved auto mode in headless sessions, the desktop app and VS Code: reports from background subagents that wait together are answered in one model call, as in the terminal\n- 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\n- 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\n- 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\n- Improved `claude remote-control`'s messages for an unanswered first-run question: each now says what to do next\n- Improved start-up time by loading trusted TLS certificates faster\n- Improved Claude apps gateway error messages to show the gateway's address without any user name or password\n- Improved the Claude apps gateway's spend limit error: the reset now shows in your local time, such as \"resets 4pm (America/Los_Angeles)\"\n- Windows: Improved startup time with a standard Git for Windows install by no longer checking which git.exe to run on every launch\n- 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\n- Updated the bundled dataviz skill to suggest a list of meters, one row per item, for several ratios that each have their own limit\n- Updated the \"Upgrade to Max\" options to say \"Upgrade to Max or Ultra\"\n- 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\n- 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\n- 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\n- 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\n- Changed nested skill loading: `.claude/skills` folders inside `~/.claude/plugins` no longer load when Claude works on files there\n- 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\n- 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\n- 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\n- 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\n- Changed `AskUserQuestion` in `claude -p` and SDK sessions to deny a host's permission answer that sets the undocumented `afkTimeoutMs` input field\n- 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\n- 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\n- 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\n- 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\n- Changed headless sessions to reload mods in `.claude/skills` when their files change; set `CLAUDE_CODE_PLUGIN_DIR_WATCH=0` to turn this off\n- Changed the Cyber Verification Program sentence on a cyber safeguard block to use the same words as claude.ai\n- 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\n- Changed `cp` and `mv` spelled by a path, such as `/bin/cp`, to get the same path checks as `cp` and `mv`\n- 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\n- Changed artifact design guidance to treat Inter and Space Grotesk as good default typefaces\n- 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\n- Changed Remote Control sessions to keep the folder they were started in: a connected client can no longer move a session to another folder\n- 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`\n- [VSCode] Added a warning above the prompt box when both an auth token and an API key are set, matching the terminal's startup warning\n- [VSCode] Fixed a plugin's permission request showing a name with hidden characters: it now reads \"a plugin\"\n","publishedAt":"2026-11-09T19:22:30.000Z","fetchedAt":"2026-11-09T20:27:33.563Z","url":"https://code.claude.com/docs/en/changelog#2-1-321","media":[],"coverageCount":0},{"id":"rel_hkpRaOLC0K0hHwthFBQ1b","version":"v2.1.319","type":"feature","title":"v2.1.319","summary":"Auto mode with telemetry or feature flags off now asks about or denies calls the server can't check instead of falling back to the built-in classifier, with CLAUDE_CODE_AUTO_MODE_SERVER=0 restoring the old behavior. The release also fixes a sandbox escape when creating a worktree and a Bash permission bypass where a crafted environment prefix could hide part of a command from permission checks.","titleGenerated":"Claude Code 2.1.319 changes auto mode fallback and fixes sandbox escape","titleShort":"Auto mode no longer falls back to built-in classifier; sandbox escape fixed","breaking":"major","importance":4,"content":"## What's changed\n\n- Changed auto mode with telemetry or feature flags off, including behind LLM gateways: calls the server can't check are asked about or denied, with no fallback to the built-in classifier; set `CLAUDE_CODE_AUTO_MODE_SERVER=0` to keep using it (Bedrock, Vertex and Foundry are unchanged)\n- Changed auto mode behind an LLM gateway: `CLAUDE_CODE_AUTO_MODE_SERVER=1` turns on the server-side check even with `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS` set; if managed settings set the latter, they must set the `1` too\n- Added support for MCP tasks in the terminal: when an MCP server runs a tool call as a task, Claude Code moves it to the background right away and tells Claude the result when it finishes\n- Added the `CLAUDE_CODE_DISABLE_MCP_TASK_BACKGROUND` environment variable to make Claude Code wait for an MCP task inside its tool call instead of moving it to the background right away\n- Added `permissions.denyInsteadOfPrompting` setting: a tool call that would wait on a permission prompt is denied and the turn continues, in terminal sessions, subagents and teammates\n- Added `/claude-api latency-optimize` to find where your Claude API code makes people wait and try faster settings against your eval\n- Added the `CLAUDE_CODE_DISABLE_FEATURE_FLAGS` environment variable, which stops feature-flag fetching without turning off telemetry; every flag then uses its built-in default\n- Added spend for today, this week and this month to `/usage` for developers behind a Claude apps gateway who have no spend limit, when the gateway runs this version or later\n- Added `admin.pooled_conflict_rule` and `admin.pooled_priority_order` to the Claude apps gateway, to set which pooled group budget pays first for a developer in several groups\n- Added an optional `login_notice` setting to the Claude apps gateway that shows an operator-defined message on the gateway's browser verification page and on Claude Code's sign-in screen\n- Added `CLAUDE_GATEWAY_LOG_FORMAT=json` to the Claude apps gateway: operational log lines become JSON with `severity`, `time` and `message`, for log collectors that read the level from a field\n- Added a Claude apps gateway boot warning for a case where another policy's `availableModels` also allows a custom model id that one policy lists by name\n- Added start-up warnings to the Claude apps gateway for policies with no `desktopHome` that rely on `coworkTabEnabled: false` or `chatSessionRetentionDays`, ahead of Claude Desktop's November 10 change\n- Added a warning when your claude.ai login ends and Claude Code continues on the sign-in from `ant auth login`, which bills usage to your Claude Console organization\n- Added a line at startup naming the Claude Console organization that is billed in a case where Claude Code uses the sign-in from `ant auth login`\n- Added an \"Earlier in this session\" row to `/tasks` that lists finished subagents, so you can reopen one and message it even after it leaves the list or you restart\n- Added a note to `/loop` confirmations in cloud sessions that a timer stops with its container, with an offer of a loop that survives\n- Fixed MCP servers on the older handshake failing to connect, or returning empty results, because Claude Code declared an extension that it only uses with newer servers\n- Fixed a long-running MCP tool call on a claude.ai connector sometimes never reporting its result after `--resume`\n- Fixed a session's diff failing to load on claude.ai and the mobile apps when the folder Claude Code runs in has a file or folder named `HEAD`\n- Fixed `/diff` showing \"Diff unavailable\" when a git repository has a file or folder named `HEAD` at its top level\n- Fixed the `/diff` pane opening blank in sessions where 64 or more turns edited files\n- Fixed slash commands typed after a leading space, tab or newline being sent to Claude as plain text instead of running\n- Fixed a request that kept failing with a 401 refreshing your login on every retry and causing 401s in your other sessions\n- Fixed `/context` sending its full visual grid to Claude on every later turn; Claude now reads only the table\n- Fixed `claude -p` exiting 0 in auto mode when its actions were refused because no safety check was available: text and JSON output now exit 1\n- Fixed a case where Claude Code listed every project file at start on Linux for projects on a network file system (NFS, SMB, Lustre and similar)\n- Fixed a case where auto mode was wrongly refused when a client sent a model change and a request for auto mode back to back\n- Fixed `/status` showing \"Auto mode server: Enabled\" when the server-side review isn't reaching the session, for example behind an LLM gateway that drops it\n- Fixed Claude-drafted feedback, surveys and `/feedback` continuing to appear after the server refused feedback for your organization\n- Fixed Claude drafting feedback on third-party gateways, where it can't be sent\n- Fixed auto mode refusing actions without saying why when permission rules or `autoMode` settings are very large: it now explains and, unless `autoMode` deny rules are set, asks you in the terminal\n- Fixed a case where Bash deny and ask rules missed commands written with quoted words\n- Fixed subagent hand-back reports showing an internal marker and provenance note when expanded outside verbose mode\n- Fixed `claude -p --resume` with a prompt losing a deferred tool call when its MCP server is missing: the run now exits 1, keeps the call and refuses later input\n- Fixed a case where text was dropped from informational messages sent to SDK clients\n- Fixed a case where Agent SDK hosts got a generic exit-code error instead of a failed run's own error message when the disk was full\n- Fixed `CLAUDE_CODE_RESUME_INTERRUPTED_TURN` re-running a turn that was interrupted in plan mode when the session resumes in another permission mode\n- Fixed edits to a settings file being ignored when made within a few seconds of Claude Code writing that file\n- Fixed a config file left at the legacy name hiding your setup with no warning: Claude Code now says on stderr when a home holds two config files, and which is in use\n- Fixed `~/.claude.json`, settings and edited files ending up empty or cut off when the disk fills during a save in a read-only folder or to a bind-mounted file (one session saving at a time)\n- Fixed Write ending the turn as a refusal when it couldn't read the file to overwrite\n- Fixed background Bash commands that finish almost instantly losing their output while being reported as completed\n- Fixed the Bash tool's diff of changed files missing a file that the command wrote just before it ended\n- Fixed a mistyped model name, such as `opus-5-5`, telling you to update Claude Code instead of suggesting the model you meant\n- Fixed a case where `NotebookEdit` changed the wrong cell when the notebook changed while a permission prompt was open\n- Fixed a ~0.5s hang after pressing Enter on the `/login` success screen: the dialog now dismisses immediately and the post-login refresh runs in the background\n- Fixed Ctrl+G not opening the external editor while typing an \"Other\" answer to a question in screen reader mode\n- Fixed Monitor's WebSocket watch failing with \"could not resolve\" behind a proxy when `CLAUDE_CODE_PROXY_RESOLVES_HOSTS` is set and the machine has no DNS of its own\n- Fixed `claude plugin eval` warnings suggesting `max: 0` for a tool that must not be called, which could never pass\n- Fixed background plugin auto-update not installing a dependency added by the updated plugin version\n- Fixed Cmd+click not opening links in Ghostty over tmux in sessions opened from the agents view\n- Fixed a mod's hot reload in `.claude/skills` missing a file that was deleted just after it appeared, and missing its return\n- Fixed `/rewind` failing with \"No files were restored\" when restoring code to a checkpoint from before `/branch`\n- Fixed the `/claude-api` skill saying Console web settings don't apply to Managed Agents sessions, and tightened its credential and permission examples\n- Fixed needless permission prompts for read-only commands wrapped in a plain `timeout DURATION`, such as `timeout 30 rg foo`\n- Fixed WebFetch describing claude.ai shared-chat links as empty pages; it now says shared chats can't be read and to paste the conversation instead\n- Fixed Artifact database calls being re-sent to artifacts without a database; errors now say whether republishing helps\n- Fixed a case where artifact database reads that save to a folder (`out_dir`) were refused\n- Fixed detailed beta tracing not presenting `CLAUDE_CODE_CLIENT_CERT` to collectors that require a client certificate\n- Fixed the OpenTelemetry `claude_code.session.count` metric counting Agent SDK `prewarm()` processes that were never claimed; a pre-started process now counts once, when it is claimed\n- Fixed `/clear` killing running commands of background agents, teammates and workflows that keep running, and foreground subagents of background agents and teammates\n- Fixed a terminal whose Claude apps gateway sign-in had expired sending telemetry and managed-settings requests that the gateway refused and logged as `invalid_token`\n- Fixed the agents view appearing to send a reply to a background session waiting on a permission prompt; the reply now stays in the box with a hint to open the session and answer\n- Fixed agent view removing a draft saved by another agent view in the same directory when exiting\n- Fixed a case where agent view came back with another agent view's input text after restarting itself for an update\n- Fixed the agents view preview cutting off long commands, results and status lines without an ellipsis, and dropping `**`, backticks and tags from commands waiting for approval\n- Fixed Claude in Chrome reporting the extension as not connected when the connection dropped mid-action; Claude now checks the page before repeating an action that may have run\n- Fixed Remote Control sessions still showing \"Limit reached\" on other devices after a usage limit was lifted outside the session\n- Fixed the file card sometimes missing in the Claude iOS app when Claude sent a single file\n- Fixed self-hosted runner sessions failing with no explanation when the server keeps refusing to renew model access: the runner now retries sooner and posts a warning in the session\n- Fixed `/permissions` in cloud sessions replying only that it isn't available; it now says where the session's rules live and how to change them\n- Fixed cloud sessions losing session-scoped deny and ask permission rules, such as those a PermissionRequest hook adds, when the session moves to a new worker\n- Fixed mod hot-reloading turning off for the session when you typed three messages instead of answering its question in a cloud session\n- Fixed background tasks that keep running after `/clear` disappearing from the task list in Claude Code on the web until another task started or ended\n- Fixed hidden commands appearing in the slash command menu of cloud sessions on mobile and web\n- Fixed a case where `/rewind` could restore or delete a file through a directory symlink\n- Fixed a case where a `sed` command could skip the permission check in accept-edits and auto mode\n- Fixed a case where `CLAUDE_CODE_AUTO_MODE_SERVER=1` in managed settings was ignored behind an LLM gateway when the gateway's URL is set only there and server-managed settings can't be fetched\n- Fixed a Bash permission bypass where a crafted environment prefix could hide part of a command from permission checks\n- Fixed a case where deny and ask rules, hook matchers and `disallowedTools` entries written with a claude.ai connector's terminal name (`mcp__claude_ai_Slack`) were ignored in Claude Desktop (takes effect once Claude Desktop includes this version)\n- Fixed a sandbox escape when creating a worktree\n- Fixed self-hosted runners: `--trust-workspace false` now also stops hooks in the repository's settings files and skills from running\n- Fixed a case where idle background sessions started through a shell-exported LLM gateway left the gateway after an update\n- Fixed help links in `/status` and the auto mode billing notice opening only part of the address when they wrap\n- Fixed a case where certificate errors behind a TLS-inspecting proxy persisted when `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC` was set. With it set, Claude Code now makes a TLS handshake with the configured API host after a certificate failure, through your http:// proxy if one is set; it sends no API request\n- Fixed `claude remote-control` exiting silently with status 0 when a first-run question gets no answer (scripts, services, batch jobs, Ctrl+C); it now says why and exits 1\n- Fixed `claude remote-control` ignoring a piped `y` that has no newline after it at the first-run question; it now counts as a yes\n- Fixed `/rename` reporting plain success when the new name could not be saved to the session transcript; it now warns that the new name may not last\n- Fixed `${CLAUDE_PLUGIN_ROOT}` not being substituted in a plugin-provided `subagentStatusLine` command, which can now reference the plugin's own scripts\n- Fixed `claude plugin update` and `/plugin` \"Update now\" not installing dependencies added by the new plugin version\n- Fixed subagents being told after compaction to expect a completion notice for background agents they neither started nor took over\n- Fixed a case where a background agent's later report was lost silently: the notification now says it was refused and how to ask for it\n- Improved config saves on Windows: Claude Code waits longer for a file another program has open, so a kill during a save is less likely to leave `~/.claude.json` empty\n- Improved Claude in Chrome after a brief disconnect: it waits up to 6 seconds for the session's own browser to return before picking another\n- Improved Claude in Chrome: `javascript_tool` results of up to 60,000 characters now stay in the conversation instead of being saved to a file\n- Improved `claude purge` and the clean-up of old sessions to retry a folder that could not be fully deleted, and to report it as an error if it remains\n- Improved the refusal for a path outside the working directories on macOS and Windows: when only its letter case differs from a path inside them, the message names that spelling\n- Improved the contrast of faint gray secondary text, such as hints, details in `/stats`, `/model` and `/sandbox`, and rejected-action messages\n- Improved the artifact database error when an update targets a document that does not exist, and when the service pauses repeated refused writes\n- Improved the Claude apps gateway's warnings about a failed upstream call: each now names the upstream, not only its provider\n- Improved the Claude apps gateway's spend limit error: it reads \"spend limit reached · resets Dec 1\", then `admin.blocked_message` if one is set\n- Improved the `/model` picker: older models are now grouped under a \"More models\" row instead of being listed after the current ones\n- Improved `/context` at the terminal so the table Claude reads stays short: each list keeps its 10 largest rows, and MCP tools that load on demand share one row per server until they're called\n- Improved `/usage` in cloud sessions: it now links to Settings → Usage on claude.ai in a case where plan limits can't be shown\n- [VSCode] Improved the confirmation for removing a plugin marketplace: it now checks which of its plugins are installed at that moment\n- Changed MCP tasks to work in the terminal without feature-flag fetching (Bedrock, Vertex, Foundry, LLM gateways, `DISABLE_TELEMETRY`); set `CLAUDE_CODE_DISABLE_MCP_TASKS=1` to turn them off\n- Changed auto mode so that a project's settings files or a plugin can no longer switch its server-side check back on once you have switched it off outside the project\n- Changed Monitor's WebSocket watch to follow a rule on every Bash command: a deny refuses it, and under an ask rule an allow on Monitor, bypass mode or auto mode no longer skips the prompt\n- Changed `CLAUDE_CODE_PROXY_RESOLVES_HOSTS` so that settings files can no longer set it where the host manages the network; set it in the process environment\n- Changed a session's exact Bash or PowerShell ask rule to be refused, not asked about, when a broader deny rule covers the command\n- Changed the Bash tool to ask for permission before running `which` with flags outside a lookup/display set\n- Changed `--add-dir` directories to register plugin marketplaces only once that directory's own trust dialog has been accepted\n- Changed nested skill loading: with the Bash sandbox on, skills in `.claude/skills` folders below the starting folder load only if you start Claude Code there or `/cd` there\n- Changed self-paced `/loop` to end when a turn finishes without rescheduling, instead of scheduling a fallback wakeup; set `CLAUDE_CODE_LOOP_KEEPALIVE=1` to keep the old behavior\n- Changed PostToolUse hooks to run, with up to 10 seconds to finish, on tool calls that finish after a new message interrupts the turn\n- Changed what Claude is told after safeguards stop a subagent: retry with a narrower prompt or tell you, instead of doing the work itself\n- Changed cloud sessions to drop session-scoped allow rules at a move to a new worker when their deny and ask rules can't all be kept, so you may be asked again\n- Changed the automatic certificate repair behind a TLS-inspecting proxy: it no longer runs when `ALL_PROXY` is the only proxy variable set. Set `HTTPS_PROXY` to the proxy's http:// URL instead\n- Changed artifact database saves of rows others may have written: each file gets a hidden marker file, and existing files without one, earlier versions' exports among them, are left alone and reported\n- Updated `/claude-api managed-agents-onboard` templates to match the Claude Console: removed Performance tuner, renamed two, and turned off web search and fetch in five whose prompts don't use them\n- Removed `shortcuts_list` and `shortcuts_execute`, which listed and started shortcuts saved in the extension, from Claude Code's built-in Claude in Chrome tools\n- Gateway: Added startup warnings for upstreams that mix AWS GovCloud with other upstreams, and for base URL variables in the gateway's environment\n- Gateway: Added a start-up log warning when managed settings name an Amazon Bedrock guardrail in a header, which the gateway doesn't forward, and no `bedrock` upstream sets a `guardrail`\n- Gateway: Fixed Claude Desktop offering a web search tool that always failed when every upstream is Amazon Bedrock\n- Self-hosted runner: Fixed a runner started with `--use-anthropic-git-proxy` or `--configure-git` registering before a failed startup check made it exit; it now exits first\n- Self-hosted runner: Changed `CLAUDE_RUNNER_CHECKOUT_PATH` so that each repository in a session gets its own folder: a folder name already taken in the session gets `-2`, `-3`, … at the end\n- Self-hosted runner: Removed the end-of-session upload of a session's debug log; a failed session's log stays on the runner host unless `--remove-session-state` is set\n- [VSCode] Fixed prompts over 50,000 characters being cut off silently: the prompt now stays in the chat box with a notice, and a paste over the limit is refused with a notice\n- [VSCode] Fixed a click on the survey or \"Report a problem\" buttons sending the unfinished prompt\n- [VSCode] Fixed cloud sessions opened from the extension's Web tab bringing back turns that `/clear`, or Retry or Edit of a typed prompt, had removed\n- [VSCode] Fixed a deleted session's file coming back when that session was renamed in the extension\n- [VSCode] Fixed Claude carrying out an unapproved plan after a window reload: a step interrupted in plan mode now stops with a notice when the session comes back in another permission mode\n- [VSCode] Improved the \"not signed in\" message behind a Claude apps gateway: it now looks like an error and has a Sign in link\n- [VSCode] Changed links in Claude's replies that point at Claude sign-in, connector or API addresses to show as text with the address, instead of opening in the browser\n- [VSCode] Fixed a possible chat panel freeze when a dropped item's path contains a very long run of \"/\" characters\n- [VSCode] Fixed the `claude-vscode.toggleDictation` command starting dictation in every open Claude view at once; it now acts only in the view you are using, and a second press stops it\n- [VSCode] Fixed the chat panel slowing down while a reply with many unclosed `$$` math fences, each shorter than the last, streams in\n- [VSCode] Improved the chat panel's diagnostics tool for large workspaces: it leaves out files with no problems, returns compact JSON, and accepts an optional minimum severity\n- [Cloud sessions] Fixed scheduled tasks silently dropping a connector named memory: runs now keep it, mounted under that name plus a short suffix\n- [Cloud sessions] Fixed Send now on a waiting message in cloud sessions ending running background agents; they now keep running\n- [Cloud sessions] Improved the stalled-setup warning on self-hosted environments: its link now prefers your organization's default or an environment with a live runner, and names it\n- [Code Review] Changed unlinking a GitHub organization to also turn off its All repositories setting in Code Review; it stays off after you relink until you turn it back on\n- [Claude Tag] Fixed Claude Tag setup blocking launch when the channel list couldn't fully load; admins can now launch without picking a channel\n- [Claude Tag] Added a Claude Tag row to the Models admin page for Enterprise orgs with Claude Tag connected, linking to where its model defaults are managed\n- [Claude Tag] Fixed saving auto mode allow rules in Claude Tag admin settings overwriting another admin's changes; the save is now refused with a message\n","publishedAt":"2026-11-06T17:04:01.000Z","fetchedAt":"2026-11-06T19:38:13.884Z","url":"https://code.claude.com/docs/en/changelog#2-1-319","media":[],"coverageCount":0},{"id":"rel_2RHL6lTHakEm6erqNrDGL","version":"v2.1.318","type":"feature","title":"v2.1.318","summary":"Stored claude.ai sign-in end dates now appear in claude auth status and claude -p JSON results, with a stderr warning starting 10 days before expiry. Fixed config and login loss on shared network home directories, and a regression in 2.1.314 where --continue could lose the earlier conversation.","titleGenerated":"Claude Code 2.1.318 adds sign-in expiry dates and fixes data-loss bugs","titleShort":"Sign-in expiry surfaced; shared-home config no longer silently dropped","breaking":"minor","importance":3,"content":"## What's changed\n\n- Added a record of your answer when Claude Code asks to turn on the mods it writes in a session\n- Added `assume_role` on `mantle` upstreams of the Claude apps gateway, for a role in another AWS account and for per-developer AWS cost attribution\n- Added an `effort` option to a mod's `$.model.fork`, so a quick side job can run at `low` whatever effort the session uses; the result reports the effort sent\n- Added the date a stored claude.ai sign-in ends to `claude auth status` (`loginExpiresAt`) and to `claude -p` JSON results (`credential_expires_at`)\n- Added a one-line stderr warning to `claude -p` text output, starting 10 days before a stored claude.ai sign-in ends\n- Fixed `--continue` sometimes losing the earlier conversation after a malformed tool call's retries ran out (regression in 2.1.314)\n- Fixed macOS sessions with many running Bash commands stalling while memory grew\n- Fixed permission prompts being answered by keys typed just after sending a message, or before a slow prompt appeared\n- Fixed `/logout`, `claude auth logout` and `claude mcp remove` reporting success, and changes to trust, consent or MCP settings being dropped with no word, when the config file could not be updated on a shared network home directory\n- Fixed jobs exiting with `Not logged in` when another terminal or host signs in again to the same account and organization on the same home\n- Fixed `claude mcp add` and `remove` writing the config without its lock: they now wait for it or exit 1, though another job's save can still replace a saved server\n- Fixed `claude plugin` commands exiting before a slow token refresh was saved, which could sign you out on the next run\n- Fixed `claude plugin validate` passing a custom-named marketplace file that `claude plugin marketplace add` rejects\n- Fixed stdio MCP servers that exit on the version probe being re-probed on every start; a server that dies in under 3s for its own reason is also pinned to the legacy handshake for 7 days\n- Fixed sessions being refused at startup by an `allowedProviders` list that another login on the same machine had stored\n- Fixed idle compaction replacing a conversation with a reply that was not a summary\n- Fixed the \"Now using usage credits\" notice reappearing about every 30 seconds while a background subagent is running\n- Fixed custom slash commands that fail, or that only Claude can invoke, printing nothing in `claude -p` and SDK sessions\n- Fixed `claude -p` and Agent SDK sessions reporting success when an API error cut off a reply while a mod hooked `turn.step`\n- Fixed a message sent right after a model change still running after Stop in headless sessions: Stop now cancels or stops it as it would a queued message\n- Fixed a deleted session transcript being re-created with its last prompt when an idle non-interactive session exited\n- Fixed `claude remote-control` hosts, the background daemon and self-hosted runners failing to start new sessions (`spawn ENOENT`) after an update or version cleanup removed the binary of a still-running process\n- Fixed a running Claude Code process owned by another user (for example one left by `sudo claude`) being treated as dead by version cleanup, which could delete its binary\n- Fixed `/remote-control` showing a session as available after a failed connection\n- Fixed `/design revoke` reporting success when the request failed with a 404\n- Fixed a shell command approved after a self-hosted runner or cloud session restart running in the session's start folder, not the folder it was asked in; if that folder can't be entered, it is not run\n- Fixed Claude not being told when an approved connector tool was unavailable after a cloud session restart\n- Fixed cloud sessions being stopped in the middle of a turn when the service refuses uploads for several minutes\n- Fixed variables a `Setup` hook writes to `CLAUDE_ENV_FILE` not reaching the Bash tool in some cloud sessions\n- Fixed the iPhone app showing a cloud session's turn as finished while it kept running, after a response was cut off at a tool call or a refused image or document was removed\n- Fixed a restarted session re-sending a message that failed with an API error, ahead of the newer message waiting behind it\n- Fixed a restarted session dropping its interrupted turn when the restart came right after Claude read a large image or retried a tool call\n- Fixed every connection failing when the `NODE_EXTRA_CA_CERTS` file is empty, not PEM, or starts with a damaged certificate: readable certificates are now used, with a warning on stderr\n- Fixed workflows re-running finished agents when resumed after a usage limit\n- Fixed `/mcp` in cloud sessions counting tools the model can't use; a server's row now shows how many of its tools are available\n- Fixed a duplicate notification when Claude left the status field out of a PushNotification call\n- Fixed the error for a too-large image in a request with more than 20 images always saying the limit was 2000px; it now shows the API's limit\n- Fixed the runtime in background shell details not updating while the dialog is open\n- Fixed a mod's `turn.step` hook making a stopped reply finish as a success in headless runs\n- Fixed retention cleanup leaving a home that only runs `claude -p` with no `~/.claude.json` backup: it now takes a fresh backup before deleting the last good one\n- Fixed the stored login being replaced by an empty file when a network home directory has no room left for data but still takes new files\n- Fixed tokens from `setup-token` being refused at start on machines with `forceLoginOrgUUID`, when the token belongs to a permitted organization\n- Fixed non-interactive launches replacing a `~/.claude.json` that isn't valid JSON with defaults; the start now fails, says how to restore the file, and leaves it as is (anthropics/claude-code#87551)\n- Fixed an empty `~/.claude.json` with no backup being reported as corrupted; Claude Code now starts as it does when the file is missing\n- Fixed the restore command printed for a damaged or missing `~/.claude.json` naming an empty or cut-off backup, or failing on a read-only file\n- Fixed the sign-in failure when a stored claude.ai login reaches its time limit: it now says so, and in `claude -p` names `claude auth login` and `claude setup-token`\n- Improved how `~/.claude.json`, settings files and the stored login survive saves on shared network home directories: they are less often deleted, briefly emptied or replaced by a near-empty or older copy\n- Improved sessions on network file systems to look again when the configuration file is wrongly reported as missing, so fewer start without your MCP servers and trusted folders\n- Improved MCP tool errors: a call made before the tool is loaded now returns its input schema\n- Improved the reliability of git commits and pull request creation when the Bash tool runs macOS's built-in `/bin/bash`\n- Improved \"Ask your admin for more usage\" in the usage limit menu to send the request in one step\n- Improved how fast remote sessions reconnect after a pause or sleep\n- Improved You should know toasts to say \"try again\" instead of \"press 2\", which only works in the terminal\n- Improved auto-update reliability: a brief network failure while verifying a release is now retried right away instead of waiting for the next check\n- Improved unattended runs (`claude -p`, SDK apps) to stop rewriting `~/.claude.json` on every run when nothing in it changed\n- Changed Claude Code to no longer install or update the IDE extension when started in an IDE terminal; use `/ide` from an outside terminal or your editor's extensions view\n- Changed subagents to stop instead of switching models when \"Switch models when a message is flagged\" is set to \"Ask each time\"\n- Changed background subagents to move a shell command to the background after 3 minutes on all providers, so the prompt cache doesn't expire while it runs\n- Changed subagent progress summaries on Bedrock, Vertex, Foundry, or with telemetry off: on a cold prompt cache, one can now wait up to 5 minutes for the subagent's pending response\n- Changed what a mod attaches at `prompt.submit` to stay in the user's turn rather than go out as a system message\n- Changed `deniedMcpServers` URL entries to also block Anthropic-provided servers whose upstream URL matches, and connectors in cloud sessions and on self-hosted runners, where the session is given the connector's own URL; otherwise deny a connector by name\n- Changed the `/permissions` rule details, built-in rules and delete confirmation views to use the standard dialog; pressing Ctrl+C twice there now goes back or cancels instead of quitting\n- Changed `/claude-api managed-agents-onboard` to give a new agent's built-in tools, and a quickstart template's MCP tools, the `auto` permission policy (\"Judge risky actions\") by default, matching the Claude Console; say \"run without asking\" to opt out\n- Removed the 2.1.288 fix that kept the model's earlier thinking when resuming a conversation started on 2.1.286 or earlier; that thinking can now be dropped\n- Gateway: Added a log warning when the Google access token from a federation (`external_account`) credentials file is about to expire\n- Gateway: Improved the log hints about `listen.trusted_proxies`: they now say to list only a proxy that adds the connecting address to `X-Forwarded-For` itself\n- Gateway: Improved the warning when AWS CountTokens fails on Amazon Bedrock: it says Claude Code, not the gateway, counts after the 501, and points at IAM only when AWS denies access\n- Self-hosted runner: Added `CLAUDE_RUNNER_SPAWN_REASON` to the orchestrator's spawn-runner hook environment: why Anthropic sent the request (e.g. `runner_lost`, `resumed`); unset when none was sent\n- Self-hosted runner: Added a startup warning when `--release-idle-session-min` is the last argument with no value (still ignored) or has a value of only spaces (still read as `0`)\n- Self-hosted runner: Changed the orchestrator's log lines: `ok session=…` now prints at every log level on stdout (was debug only, stderr) with `reason=…` after `jti=…`, and `nack session=…` ends its fields with `reason=…` after `retryable=…`; update scripts that match `retryable=true:` or `retryable=false:`\n- Remote Control: Fixed a misleading \"spawn error: ENOENT\" when a session's working folder is missing; the error now names the folder\n- Remote Control: Improved the title a fork of a terminal fork is created with when the app doesn't set one: \"Name (fork 2)\"\n- [VSCode] Fixed a window whose sign-in the organization's settings refuse starting a new Claude Code process in the background every minute\n- [VSCode] Fixed a long word in the chat panel's error banner pushing its close button out of view\n- [VSCode] Improved the sign-in screen to show what Claude Code printed when its sign-in status check fails\n- [Cloud sessions] Added a No repository option: a cloud session on an Anthropic-hosted environment can now start without picking a repository\n- [Cloud sessions] Added the usual Chrome names to cloud sessions, so tools other than Playwright can find the built-in Chromium through google-chrome and CHROME_PATH\n- [Cloud sessions] Added a New session button to the card shown when a cloud session's environment was deleted, where the message box is hidden\n- [Cloud sessions] Fixed reading GitHub Actions logs in cloud sessions: `gh run view --log` and `gh api` log requests now get the log in most environments, not an unreachable redirect\n- [Cloud sessions] Improved most Create PR failure messages in cloud and Remote Control sessions: they now give the reason or a link to GitHub instead of saying Try again\n- [Cloud sessions] Changed connector search results to tell Claude that the search covers only part of the connector directory\n- [Remote Control] Fixed side-by-side checkouts of one repository looking identical in the environment menu's Remote Control rows; each now shows its own folder name\n- [Claude Tag] Fixed keyboard focus and screen reader announcements in Claude Tag admin setup; pressing Escape in the connect form no longer also closes the Add dialog\n- [Code Review] Improved the Code Review analytics page to retry once after a server or network failure, then show a Try again banner instead of empty cards\n","publishedAt":"2026-11-04T00:01:03.000Z","fetchedAt":"2026-11-04T02:54:30.956Z","url":"https://code.claude.com/docs/en/changelog#2-1-318","media":[],"coverageCount":0},{"id":"rel_O-C6_gk8wVnAo7EiAQZWU","version":"v2.1.317","type":"feature","title":"v2.1.317","summary":"CLAUDE_CODE_DISABLE_NESTED_SESSION_ID=1 makes claude -p inside a self-hosted runner session use that session's ID and transcript as in earlier versions, and project and local MCP servers no longer disconnect after the first turn in some Claude desktop sessions. The release also adds a warning when /schedule assigns unrequested connectors and a check that asks before a web or shell tool call sends your account email to the network.","titleGenerated":"Claude Code 2.1.317 restores nested session IDs and fixes MCP disconnects","titleShort":"Nested session IDs restorable; desktop MCP servers stay connected","breaking":"none","importance":3,"content":"## What's changed\n\n- Added a warning to `/schedule` when a new cloud scheduled task gets connectors you did not ask for, with how to remove them\n- Added a check that asks before a web or shell tool call sends your account email address to the network\n- Added `onFailure: \"block\"` for Agent SDK hook callbacks, so a callback that fails blocks the action instead of letting it run\n- Added a notice in Claude Desktop's Code tab when your organization's remote managed settings haven't loaded for the session, and another once they do\n- Added a notice when auto mode's check has no answer for a subagent's tool call, which used to leave nothing in the main view\n- Added `anthropic.plan`, `anthropic.seat_type`, `anthropic.provider` and `anthropic.custom_base_url` to OTel metrics, events and spans\n- Added a cost report to the Claude apps gateway admin API, `GET /v1/organizations/analytics/cost`, for reading past spend over a date range in total, per developer, per day or per IdP group\n- Added a daily record of each developer's IdP groups on Claude apps gateways with `admin:` set, kept for `admin.identity_retention_days`, which the cost report groups by\n- Added `CLAUDE_CODE_DISABLE_NESTED_SESSION_ID`: set it to `1` on a self-hosted runner to have `claude -p` run inside a session use that session's ID and transcript, as in earlier versions\n- Fixed project and local MCP servers disconnecting after the first turn in some Claude desktop sessions\n- Fixed auto mode failing every turn when an LLM gateway set as the base URL answers 403 or 422 to its server-side safety check (a 403 with an apiKeyHelper is unchanged)\n- Fixed `claude auth status` exiting before a slow token refresh was saved, which could sign you out on the next run\n- Fixed a project's `effortLevel` setting being ignored by sessions that an app starts ahead of time and then opens in the project, such as new sessions in the desktop app's Code tab\n- Fixed `claude --from-pr` reporting \"No conversations found\" in projects with many sessions when the linked session was older than the first page\n- Fixed the second of two simultaneous permission prompts showing no file path or command on claude.ai and the mobile apps\n- Fixed Remote Control and SDK hosts offering \"don't ask again\" on a mod's tool calls\n- Fixed session titles being lost in cloud and SDK sessions when a new message or Stop arrived while the title was generating\n- Fixed self-paced `/loop` wakeups sometimes not being scheduled when Claude left a field out of its request\n- Fixed auto mode in cloud sessions stopping for approval when the Artifact tool uploaded, wrote to a database or saved a file through the outputs folder\n- Fixed skills synced from claude.ai being lost or downloaded again every ten minutes when they can't be written to disk\n- Fixed `/schedule` leaving a cloud scheduled task's next run time and link out of its results after creating or updating it\n- Fixed screen readers speaking a running subagent's elapsed time and token count on every change in screen reader mode\n- Fixed a region or network block (HTTP 403) being shown as a sign-in failure; the error now explains the block\n- Fixed Claude saying a working claude.ai connector needed authorization or had failed in cloud sessions when a synced plugin included the same server\n- Fixed `claude -p` exiting 0 when a full disk or quota lost its transcript or ~/.claude.json writes; it now prints an error and, in text and json output, exits 1. Full NFS exports may be missed\n- Fixed edits to settings files made in the first second or so after launch not taking effect until a settings file changed again\n- Fixed old transcripts not being cleaned up on machines where some runs skip cleanup, such as runs using `--setting-sources`\n- Fixed deleted agent team inboxes being re-created by a teammate's late message\n- Fixed a re-created agent teammate being removed from its team by an earlier shutdown approval\n- Fixed stored scheduled tasks being impossible to list or delete after scheduled tasks were turned off\n- Fixed Claude deleting your most recently updated `/team-onboarding` guide when asked to delete one without naming which\n- Fixed Claude saying a `/team-onboarding` guide was deleted when the server refused the delete\n- Fixed the Claude apps gateway hiding Amazon Bedrock's \"Input is too long.\" error, which left sessions failing every turn with \"upstream rejected the request\" instead of compacting\n- Fixed self-hosted runner sessions losing recent turns or the whole conversation when restarted while a plan waited for approval\n- Fixed \"Prompt is too long\" after automatic compaction in sessions whose very large CLAUDE.md changed since it was last sent\n- Fixed high memory use and slow startup in headless sessions with many plugins that register commands\n- Fixed stdio MCP servers receiving an empty `AWS_BEARER_TOKEN_BEDROCK` when their env config sets it to empty; the variable is now unset, so the server can use the machine's own AWS credentials\n- Fixed API requests failing with \"unable to get local issuer certificate\" behind some TLS-inspecting proxies after a root certificate renewal\n- Fixed very long conversations losing their earlier summary when compaction had to shorten its own request\n- Fixed a PDF's own metadata being able to make Read and @-mentions under-count its pages\n- Fixed `DISABLE_PROMPT_CACHING_OPUS` and `DISABLE_PROMPT_CACHING_SONNET` having no effect on `[1m]`, other-region and undated spellings of the default model\n- Fixed `/artifacts` and the Artifact tool's shared list showing no shared artifacts when you have 200 or more of your own\n- Fixed Claude treating partial Grep and Glob results as complete when a search hit its time limit\n- Fixed OpenTelemetry exports to an `https://` endpoint through a proxy never being sent when they started while another export was still in progress\n- Fixed a flood of connection attempts to a proxy that blocks (HTTP 403) OpenTelemetry exports to an `http://` endpoint\n- Fixed OpenTelemetry exports hanging indefinitely behind a proxy that never answers; they now time out after 10 seconds, or `OTEL_EXPORTER_OTLP_TIMEOUT` up to 30 seconds\n- Fixed resumed sessions telling Claude again about a file change it already knew about\n- Fixed Claude sometimes not using a built-in tool in a resumed session after an earlier call failed because the tool was switched off\n- Fixed `/remote-control` answering \"Unknown command\" instead of saying that Remote Control is turned off by the `disableRemoteControl` setting\n- Fixed agent team teammates in tmux or iTerm2 panes starting in auto mode when the lead session was in manual mode\n- Fixed a background session sometimes losing a just-made change of permission mode when it was upgraded or retired\n- Fixed Ctrl+C stopping all running background agents when pressed to clear typed text at the prompt\n- Fixed /feedback showing \"Enter to confirm\" while a report is being sent\n- Fixed child processes of a stopped shell command sometimes being sent SIGTERM twice, which could make some programs skip their cleanup\n- Fixed `claude -p` run inside a cloud or self-hosted runner session sharing that session's ID: it wrote into the session's transcript, and a restart could resume the wrong conversation\n- Fixed `claude -p` run inside a cloud or self-hosted runner session generating a title of its own and sending it to that session\n- Improved the error shown on Microsoft Foundry when Azure credentials can't be loaded: it now says to check or refresh them, and reports as `cloud_credential_error`\n- Improved `/design-sync` to say up front when your design system has moved to Claude, and stop before building instead of failing at upload\n- Improved the startup error when a required Claude apps gateway can't be reached: it points to the network instead of a sign-in, and says when the gateway's TLS certificate can't be verified\n- Improved the \"Resume this session with\" hint shown at exit to repeat the flags you launched with (for example `--chrome --debug`) ahead of `--resume`\n- Improved how self-hosted runners recover from server errors: after a 5xx other than 501 or 529, a runner's next poll now opens a new connection instead of reusing the one that failed\n- Improved how self-hosted runner sessions recover from server errors: after a 5xx other than 501 or 529, a session now opens new connections instead of reusing the one that failed\n- Improved how self-hosted runner sessions recover when connections stop answering: after a heartbeat gets no answer, a session now opens new connections instead of trying each idle one in turn\n- Changed SendMessage so a message to a running background agent arrives earlier: its running command moves to the background, without being stopped\n- Changed subagents and teammates on a different model from the main conversation to use that model's saved effort level or its default, instead of the session's effort level\n- Changed exit to skip the \"Background work is running\" prompt for plugin monitors\n- Changed the built-in `plugin-authoring` skill to offer once, after a mod you built works, to help share the mod\n- Removed the Read tool's \"file unchanged since your last Read\" stub: reading an unchanged file again returns its content\n- Windows: Fixed every tool call being refused on some volumes where the native path lookup fails; the working-directory check now compares the folder's identity instead\n- Windows: Fixed the session scratchpad and other temporary paths growing with your project's path length in sessions not started by Claude Desktop, which made repositories cloned into them hit the 260-character path limit\n- Gateway: Fixed the status line and `/usage` showing a developer's own spend limit as 100% used after their group's pooled budget ran out\n- Self-hosted runner: Fixed Claude not seeing the interrupted tool call when a session moves to a new runner and the turn is older than `CLAUDE_CODE_RESUME_INTERRUPTED_TURN_MAX_AGE_MS`\n- Self-hosted runner: Fixed an MCP server that was switched off coming back on after the session moved to a new machine, when the session service had refused to save the switch\n- Remote Control: Improved fork names when the app doesn't set one: a long name keeps its \"(fork)\" ending, and in sessions started with `claude remote-control` a fork of \"Name (fork)\" is \"Name (fork 2)\"\n- [VSCode] Fixed forked conversations dropping messages: only one of the tool calls Claude ran in parallel was kept, and the last messages a compaction kept were lost, which could leave Claude's answer without the question it answered\n- [VSCode] Fixed a message being lost when Claude did not finish starting in time: the extension now starts Claude once more before showing the error\n- [VSCode] Fixed the sign-in screen opening when a request is blocked for its region or network\n- [VSCode] Fixed the Account panel showing the plan name with a lower-cased second word\n- [VSCode] Fixed file links in a Claude tab opening on top of the chat instead of in another editor group\n- [VSCode] Fixed attachment chips, menus, banners and dialogs in the Claude panel staying small when `chat.fontSize` is raised\n- [VSCode] Fixed `ultrathink` not raising effort for its turn or showing its rainbow when telemetry is off\n- [Cloud sessions] Added a link card to cloud sessions, so you can open or decline a link an MCP server asks you to visit instead of the request going unseen\n- [Cloud sessions] Changed the organization environment dialog in admin settings to the personal dialogs' short layout: each section opens from a row, and unsaved input asks before closing\n- [Cloud sessions] Fixed adding a network secret for a host that already has one silently replacing it; the Add secret form now refuses and points you to the existing secret\n- [Cloud sessions] Fixed new scheduled tasks ignoring your organization's default environment: the form now starts on the one an admin chose instead of the Anthropic-hosted default\n- [Remote Control] Fixed the next message to an idle session going unanswered after you restart `claude remote-control`, when you sent it within a minute of your previous message\n- [Claude Tag] Changed newly pasted artifact links in Slack to show a simpler card with an Open button that opens the artifact in Claude, instead of Slack's side panel\n","publishedAt":"2026-11-03T01:25:46.000Z","fetchedAt":"2026-11-03T02:38:06.668Z","url":"https://code.claude.com/docs/en/changelog#2-1-317","media":[],"coverageCount":0},{"id":"rel_RtkLb4jV3d50qor9FjLYd","version":"v2.1.316","type":"feature","title":"v2.1.316","summary":"Fast mode for Opus 4.8 is deprecated and will be removed on November 30, 2026, after which requests return an error without falling back to standard Opus 4.8; migrating to Opus 5.5 keeps the speedup. Also added CLAUDE_CODE_SUBAGENT_EFFORT to set subagent effort instead of inheriting the session's, plus pooled group budgets and layered policy mode for the Claude apps gateway, and fixes for orphaned shell commands on exit and Bash tool alias handling.","titleGenerated":"Claude Code v2.1.316 deprecates Opus 4.8 fast mode, adds subagent effort control","titleShort":"Opus 4.8 fast mode deprecated; subagent effort env var added","breaking":"minor","importance":4,"content":"## What's changed\n\n- Added `CLAUDE_CODE_DISABLE_ULTRATHINK`, which turns the `ultrathink` keyword off\n- Added pooled group budgets to the Claude apps gateway: one monthly budget that a group's members share, on top of each member's own limit, set with `pooled_budget` on the spend limits admin API\n- Added `CLAUDE_CODE_DISABLE_LOOPED_CROSS_CERT_DROP` to keep system cross-certificates that certify each other in a loop, for servers whose private CA needs one\n- Added `CLAUDE_CODE_SUBAGENT_EFFORT` to set the effort level subagents run at when nothing else names one, instead of inheriting the session's\n- Added `managed.policy_mode: layered` to the Claude apps gateway: every policy that matches a developer applies, each merged over the ones listed below it, not only the first\n- Added `guardrail.input_tagging` on Claude apps gateway Bedrock upstreams, which wraps prompts in the input tags that an Amazon Bedrock guardrail's prompt attacks filter needs; `all_messages_and_system` also tags the system prompt\n- Deprecated fast mode for Opus 4.8 as of October 30, 2026; it will be removed on November 30, 2026. After removal, fast mode requests on Opus 4.8 return an error and do not fall back to standard Opus 4.8. Migrate to Opus 5.5 to keep the speedup\n- Fixed shell commands being left running on macOS and Linux when Claude Code exits by itself, for example when credentials are refused or a cloud session is archived\n- Fixed bash aliases with line breaks in their value (except in Git Bash), and functions whose names start with a dash, being garbled or lost in the Bash tool's shell\n- Fixed bash scripts started from the Bash tool stopping after their first command when `SHELLOPTS` is exported\n- Fixed PowerShell hooks not blocking on exit code 2 when the command begins with a type cast, `param`, `using` or an attribute\n- Fixed some network requests hanging when a compressed response was cut off or corrupted\n- Fixed `sonnet`, `opus` and `haiku` on Vertex AI and Amazon Bedrock showing an error mid-session instead of switching to an older model when the newest is not yet available to the account or region\n- Fixed the Claude apps gateway not checking every model a request names against `availableModels`, `deniedModels` and `models`; a request that names a blocked model now gets a 400 naming the field\n- Fixed `claude auth login` refusing to run where managed settings require a Claude apps gateway; in a terminal it now shows the gateway sign-in\n- Fixed `--restricted` sessions uploading files from outside the working directories when a Read allow rule covered them\n- Fixed prompts containing invisible characters needing a second Enter to send; plain prompts now send cleaned on the first Enter, and commands still wait for review\n- Fixed sub-agents that inherited a parent sub-agent's model moving to the main model when a background job or message woke them\n- Fixed a managed `autoMode.allow` list without `\"$defaults\"` getting the built-in exceptions back from a host app's `--settings`, a first rule added in `/permissions`, or `/auto-mode-setup`\n- Fixed cloud sessions and cloud scheduled tasks not being marked as failed when auto mode stopped a turn because the server returned no safety verdict\n- Fixed `claude -p` text output sometimes not saying why auto mode stopped a turn when a background agent was still running\n- Fixed `CLAUDE_CODE_AUTO_MODE_SERVER=1` being ignored in sessions started by a host application (such as Claude Desktop) behind a proxy or LLM gateway with experimental betas turned off\n- Fixed plugin MCP `headersHelper` scripts receiving a corrupted `CLAUDE_PLUGIN_ROOT` and server URL when a credential-named environment variable was set to a switch value such as `1`\n- Fixed the SDK `register_repo_root` control request not loading the CLAUDE.md of a directory outside the working directory (with `CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1`)\n- Fixed the Artifact tool showing `?` for artifact database access rules at the `self` level, and not quoting those rules back when it refuses a republish that would drop them\n- Fixed `/model <id>` not warning when the model isn't recognized by this version and auto-compact will assume a 200K context window\n- Fixed the \"hold space to speak\" hint using up its three showings in one session, and showing an empty key when push-to-talk is unbound\n- Fixed Grep and Glob never returning when the search process is killed at the timeout but does not exit, seen on Windows\n- Fixed a rare case where resuming a conversation did not restore its plan, when `/clear` or `/resume` had been run while part of that conversation was still being saved\n- Fixed first-run setup dropping words from the security notes and the terminal-setup explanation in narrow terminals\n- Fixed the Enterprise survey card taking a typed 1 or 0 after a short pause; a typed answer now needs Enter\n- Self-hosted runner: Fixed a message sent while Claude was working being answered after a Stop, once the session's process had been replaced\n- Self-hosted runner: Fixed an MCP server's on/off switch sometimes being lost when a session moved to a new machine\n- Windows: Fixed the Bash tool losing your aliases and functions in Git Bash when the rc file sets errexit and pipefail and only winpty aliases, or none, are defined\n- Improved `/code-review` started from the Code tab in Claude Desktop or on the web: cloud and `claude remote-control` sessions now show the findings card, not a plain-text list\n- Improved `/diff` on large repositories: each refresh usually runs one fewer full-repository git command, and refreshes are spaced out when git is slow\n- Improved the messages shown when `/ultrareview`, a cloud session or a cloud agent finds no GitHub remote: they now say to create a repository at https://github.com/new, and `/ultrareview` no longer suggests a `gh` command that `gh` rejects\n- Improved SSL certificate errors to name the likely causes, among them a server that leaves out an intermediate certificate, and the fix for each\n- Changed auto mode so that an `autoMode.allow` list without `\"$defaults\"` no longer removes the built-in rule on scheduled-task prompts\n- Changed server-managed settings for Team and Enterprise logins: settings fetched from Anthropic's API now apply without an approval prompt, and a stored copy that doesn't verify is refused\n- [VSCode] Added sign-in to a Claude apps gateway from the chat panel, with no terminal needed\n- [VSCode] Changed `ultrathink` to run its turn at `xhigh` effort and show in a rainbow in the chat box, as in the terminal\n- [VSCode] Fixed the missing Effort slider, Ultracode switch and model name when the model setting has the `[1m]` tag, such as `opus[1m]`\n- [VSCode] Fixed saved `/btw` side-question history being deleted too early while a settings file has errors\n- [VSCode] Fixed a conversation coming back in Auto mode after a reload or restart when its plan was approved with \"Yes, and manually approve edits\"\n- [Cloud sessions] Added a form card to cloud sessions, so you can fill in or decline a form an MCP server asks for instead of the request being canceled without you seeing it\n- [Cloud sessions] Added a \"Catching up…\" line that shows while a cloud or Remote Control session you switch back to is still bringing its messages up to date\n- [Cloud sessions] Changed network secrets so Enterprise members can add and delete them on their own cloud environments, instead of only viewing them\n- [Cloud sessions] Improved the Changes pane for cloud sessions: the note about what the list leaves out is now a badge beside the branch name, not a row above the diff\n- [Cloud sessions] Fixed file search in a cloud session's Files pane showing folders as ../.. paths after Claude changed directory; results now show their path from the project's root\n- [Claude Tag] Fixed Claude refusing to edit or append to a canvas it made in a Slack channel, from that channel, once the canvas was also shared to another channel or a DM\n- [Claude Tag] Fixed untagged replies going unanswered in a Slack Connect thread when both organizations have installed Claude and Slack delivered the reply to the other organization's app\n- [Claude Tag] Fixed a cloud scheduled task's live status checklist in Slack staying half-done; a run can now edit the top-level messages it posted during that run\n- [Claude Tag] Changed a Slack channel's Respond automatically switch to a \"How Claude participates\" choice on the channel's Configure page and in Claude Tag admin settings\n- [Code Review] Added a findings-by-severity card to the Code Review analytics page, showing the period's total findings, a daily chart stacked by severity, and each severity's count\n","publishedAt":"2026-10-31T01:01:35.000Z","fetchedAt":"2026-10-31T01:48:50.079Z","url":"https://code.claude.com/docs/en/changelog#2-1-316","media":[],"coverageCount":0},{"id":"rel_GVnwz7Q4leUVNJVfqLNdq","version":"v2.1.314","type":"feature","title":"v2.1.314","summary":"Added permissions.blockWritesOutsideWorkingDirectories, which makes Write, Edit and NotebookEdit refuse paths outside the working directories in every permission mode, plus idle compaction for 200K+ token conversations left idle in VS Code and Claude Desktop's Code tab. Fixed sessions failing turn after turn with \"Invalid effort level\" when an admin caps a model's effort, claude -p exiting 0 with no output on startup failure, and Bedrock sessions with many images failing compaction with 400 Input is too long.","titleGenerated":"Claude Code v2.1.314 adds write-block setting and fixes turn after turn effort errors","titleShort":"Writes outside working directories now blockable; effort-cap error fixed","breaking":"none","importance":3,"content":"## What's changed\n\n- Added `permissions.blockWritesOutsideWorkingDirectories`: Write, Edit and NotebookEdit refuse paths outside the working directories in every permission mode\n- Added an optional `groups` list to Claude apps gateway upstreams: only developers in the listed IdP groups are sent to that upstream, on failover too\n- Added `guardrail_trace` to Claude apps gateway Bedrock upstreams, to log which Amazon Bedrock guardrail filter fired on a request, without the matched text\n- Added idle compaction to VS Code and Claude Desktop's Code tab: a 200K+ token conversation left idle is summarized, and the summary is used on your next message if the prompt cache has expired\n- Fixed sessions failing turn after turn with \"Invalid effort level\" when an organization's admin caps a model's effort: Claude Code now lowers the effort, retries, and says which level it uses\n- Fixed `claude -p` exiting with code 0 and no output when startup failed; it now prints the error and exits 1\n- Fixed Bedrock sessions with many images failing every turn and `/compact` with \"400 Input is too long.\" instead of removing older images and compacting\n- Fixed Microsoft Foundry sessions failing when the resource has no deployment for the default Sonnet, Opus or Haiku model: Claude Code now falls back to an older model of that family\n- Fixed repeated refused requests behind a gateway that doesn't accept tool search's `tool_reference` blocks: the request is sent again once without tool search, and the session remembers\n- Fixed TLS connections failing with `UNABLE_TO_GET_ISSUER_CERT` when the system certificate store holds a renewed root certificate that is cross-signed with its predecessor\n- Fixed MCP servers that also work without sign-in showing \"needs authentication\" after their stored sign-in stopped working; they are now used signed out\n- Fixed `claude mcp logout` reporting success when the MCP sign-in was not removed, for example while the macOS keychain is locked over SSH\n- Fixed extra sign-in renewals when several sessions share an Anthropic profile login and the credentials lock is busy\n- Fixed `claude ultrareview` and `claude remote-control` asking the workspace trust question in a CI job that has a terminal; with `CI` set they now refuse an untrusted folder at once\n- Fixed an administrator's TLS trust settings in managed settings not being applied in Claude Desktop sessions\n- Fixed model switching being refused on machines with managed plugin settings when one of your own `--plugin-dir` or `--plugin-url` plugins had a broken manifest or failed to download\n- Fixed the Bedrock and Vertex model upgrade offer coming back at every start when managed settings pin the model\n- Fixed `forceRemoteSettingsRefresh` set only in remote managed settings no longer applying once the stored copy had expired, which let sessions start without policy when the server could not be reached\n- Fixed \"Continue automatically at usage limit\" continuing more than three times without a prompt from you when a task restarted itself between waits (a /goal check-in, a background or scheduled task)\n- Fixed background and cloud sessions started with Ultracode turning it back on at their next start after it had been switched off\n- Fixed the last broken reply and its retry prompts staying in the conversation after Claude gave up on a tool call that could not be parsed\n- Fixed a PreCompact hook with very large output making compaction fail or repeat; each hook's output is now cut at 10,000 characters\n- Fixed background diff refreshes briefly taking the repository's index lock, which could make a git command run at the same moment fail with \"Unable to create '.git/index.lock': File exists\"\n- Fixed markdown tables jumping between grid and stacked layouts on half-typed rows while streaming, and long tables falling back to raw pipes\n- Fixed the /resume picker showing the text of CLAUDE.md or CLAUDE.local.md as a session's title when the session only ran slash commands\n- Fixed `/btw` answering from the main conversation when asked from a subagent's open transcript\n- Fixed pasting an image over SSH to a Linux machine doing nothing silently; the prompt now says to copy the file over and paste its path\n- Fixed `claude remote-control` sessions that failed to start when the session service was slow for a moment: the worker's registration is now retried with growing waits\n- Fixed model switches over Remote Control that take over 8 seconds getting a \"still pending\" error instead of the final result\n- Fixed the `sandbox.excludedCommands` setting description saying listed commands always run outside the sandbox; it now says every command in a call must match an entry\n- Fixed shell and MCP calls that auto mode, a deny rule or a PreToolUse hook blocked, or that otherwise did not run, reading as if they ran in a collapsed tool group\n- Fixed sandbox protections not applying to a worktree entered by `claude -p --resume` or `--continue`\n- Fixed hooks matching Bash not running on shell commands started by the Monitor tool; such hooks can block or ask, but not approve or rewrite those commands\n- Fixed a settings file excluded with `--setting-sources` still supplying some settings, such as auto mode rules and saved permission consents\n- Fixed auto mode reporting a call as blocked, and listing it in `/permissions`, when its safety check gave no verdict\n- Fixed a Claude apps gateway Vertex upstream hanging for minutes when a Google token endpoint accepts the connection and never answers\n- Fixed the Claude apps gateway's start-up warning for a deprecated Claude Desktop key that it still serves as written saying \"stops being accepted\" after the key's end date\n- Fixed `/upgrade` opening the browser and starting a new login for plans billed through the App Store or Google Play; it now says where to change the plan\n- Fixed `--resume`, and rewinding past a switch to another model after a refusal, moving a session on a model name Claude Code doesn't recognize to a different model\n- Fixed requests failing when a gateway or cloud provider refuses one of Claude Code's beta headers by name: the request is sent again without it\n- Fixed model IDs spelled with a dot behind an LLM gateway, such as `claude-opus-4.8`, being treated as an older model with the wrong context window, output limit, effort, thinking mode and cost\n- Fixed Remote Control clients showing a stray \"Compacted (ctrl+o to see full summary)\" row after /compact\n- Fixed the cloud sessions tip and the Remote Control idle notice showing when an organization's policy turns those features off\n- Fixed a resumed claude.ai cloud session making a model call nobody asked for when the previous worker stopped just after receiving a new chat's files\n- Fixed WebFetch in Cowork cloud sessions silently dropping a PDF's text past 100,000 characters; it now says how much was unread and where to read on\n- Fixed Claude being told only that a command was declined after you stopped it with Esc or Stop in a cloud session or the IDE; it now also hears that the command may have run\n- Fixed blank lines at the top of a code block disappearing when the code block opens a quote\n- Fixed a blank line appearing above some `⎿` rows in the transcript view (Ctrl+O)\n- Fixed vim mode `dw`, `dW`, `yw` and `yW` carrying on past the end of the line, and made `w` and `W` stop on empty lines\n- Fixed x and Enter in the agent panel sometimes acting on the wrong agent when another agent started or finished\n- Fixed pasted text and dragged-in files being lost in Claude's question dialog unless the \"Other\" row had focus\n- Fixed pasted text being lost when a row in the prompt footer was selected\n- Fixed vim mode `p` over a character selection ignoring whether the yanked text was whole lines\n- Fixed vim mode `aw` and `iw` taking the line break at the end of a line\n- Fixed ↓ selecting footer items hidden by the ? help menu, the command list or a paste hint\n- Fixed Esc on a command's panel, such as `/btw`, also interrupting the subagent whose transcript is open\n- Fixed the welcome header in fullscreen mode showing the session's model while viewing a subagent or teammate that runs on a different model\n- Fixed the fullscreen transcript being left mid-conversation after \"Restore conversation\" in the rewind menu while scrolled up\n- Fixed a rare case where a deleted session could reappear in the resume list as an almost empty entry\n- Fixed prompts containing tags such as `<bash-stdout>` or `<local-command-stdout>` not being acknowledged to a local stream-json host\n- Fixed sessions staying untitled when the first prompt failed (for example on a usage limit) and the conversation continued afterward\n- Fixed the skill file watcher checking every removed skill kept in `~/.claude/skills/.trash`, which added file system load that grew with each removal\n- Fixed the session name from /rename disappearing while Claude asks for permission or a plan decision, so side-by-side sessions stay identifiable\n- Fixed background sessions that fail to start being reported as a deleted daemon binary when the binary is still on disk\n- Fixed the tip and idle reminder about push notifications appearing when the push settings in `/config` are hidden or your organization has turned off Remote Control\n- Fixed `~/.claude/skills/.trash` and `~/.claude/plugins/.trash` growing without limit: each now holds about 1024 entries, whatever `cleanupPeriodDays` says\n- Fixed the Bash edit diff listing a whole file as changed when a CRLF file under git's line-ending conversion was re-saved unchanged or had one line edited\n- Self-hosted runner: Fixed an idle session moving to another runner instead of pausing when the runner's request to pause it lost its connection before Anthropic answered\n- Improved memory use in long headless and SDK turns that run many subagents: the parent session no longer keeps each subagent's tool results until the turn ends\n- Improved prompt history: Ctrl+R now matches text inside collapsed `[Pasted text]` pills\n- Improved the `/design-sync` upload prompt to show the folder, the project and the files it will upload or delete\n- Improved `/permissions` review tab, now called Recently blocked: the footer says what Esc will do with marked rows\n- Improved auto mode's blocked calls: messages name the command, say Claude may retry and auto mode checks again, and point to `/permissions` to review\n- Improved permission prompts: a long reason takes fewer rows, so it pushes the command being approved off the screen on fewer terminals\n- Improved the cloud session error shown while a request refused for your claude.ai login is still being retried: it now also suggests `claude auth login`\n- Improved the behavior of stored claude.ai, Team, and Enterprise logins when MDM and remote settings differ on base URL or `apiKeyHelper`\n- Improved `/config` in agent view to say how to stop agent view opening by default\n- Improved session auto-titles: a session that starts with a skill or a prompt-based command such as `/security-review` now gets a title from it instead of staying untitled\n- Improved `--resume` speed and memory use on damaged transcripts\n- Improved `/context` speed in sessions with ten thousand or more MCP tools\n- Self-hosted runner: Improved the error shown when a session's branch isn't on the git remote: it now covers a never-pushed branch, says how to fix each case, and no longer suggests retrying\n- Updated the `/claude-api` skill: its Datadog setup for API tracing now sends the `dd-llmobs-enabled: true` header, as the Console does, and its `ant` install instructions drop the outdated macOS `xattr` step\n- Updated the bundled dataviz skill to advise printing y-axis tick labels in full under 10,000, and horizontal ones in full under a million where they fit\n- Updated the bundled dataviz skill to answer donut and pie chart requests with ranked bars or a segmented bar that prints its values\n- Changed Claude's file tools to ask before editing the local copy of your organization's server-managed settings, as they do for your settings files\n- Changed `/ultrareview` to check workspace trust before it runs git, as `claude ultrareview` does\n- Changed the Bash tool's shell to clear any inherited `which` alias or function and to stop passing exported `which`/`unsetenv` functions to commands, so `which` is the system command or zsh builtin\n- Changed `/permissions` Recently blocked to refuse approving a call you can't read in full, including one the terminal is too short to show\n- Changed auto mode in Cowork cloud sessions: a hook's allow of an MCP tool call now counts as no answer, so the classifier or a permission prompt decides\n- Changed hook `if` conditions such as `Monitor(tail *)` to match the Monitor tool's shell commands as `Bash(tail *)` matches Bash's\n- Changed the `claude-api` skill to ask to read a Managed Agents workflow run's workflow before changing the agent to improve that run\n- Changed flagged requests to retry on another model, not end in a refusal, when `ANTHROPIC_DEFAULT_OPUS_MODEL`, `availableModels` or `deniedModels` leaves only an older Opus model, or none\n- Changed idle compaction to also run for subscribers with telemetry turned off, and on Bedrock, Vertex, Foundry and LLM gateways once the one-hour prompt cache is on\n- Changed `FORCE_HYPERLINK=false` (also `no` and `off`) to turn hyperlinks off instead of on\n- macOS: Changed `git grep` in most commands that Claude runs to use the C locale, to avoid a reported memory leak in macOS's regex library on files that aren't valid UTF-8\n- [VSCode] Fixed \"Clear authentication\" for MCP servers reporting success when the saved sign-in could not be removed\n- [VSCode] Fixed missing spinner and Stop button while a queued message's hooks run\n- [VSCode] Fixed session group changes that cannot be saved, such as a delete, failing silently; an error notification now appears\n- [VSCode] Fixed Cmd+F / Ctrl+F not opening the find widget in the plan preview tab\n- [VSCode] Fixed the voice dictation shortcut (Ctrl+D) doing nothing under non-Latin keyboard layouts such as Russian, Greek or Hebrew\n- [VSCode] Fixed missed one-time saved scheduled tasks being deleted without asking when the extension opened without a chat\n- [VSCode] Changed `CLAUDE_CODE_DISABLE_CFC_PROMPT` to also stop the extension adding the Claude in Chrome instructions to a session's messages; the browser tools stay available\n- [Cloud sessions] Added an Include completed button to the Scheduled tasks page when finished one-off tasks are the only ones hidden, so you can show them without opening the Sort menu\n- [Cloud sessions] Fixed recently used self-hosted environments never appearing among the recent environments on the new session card\n- [Cloud sessions] Fixed cloud sessions reporting uncommitted changes every turn after Claude moved a branch to a commit it had pushed from another worktree\n- [Cloud sessions] Fixed the self-hosted environment details dialog in admin settings squeezing its content out of view in very short windows or at high zoom; the dialog now scrolls\n- [Cloud sessions] Fixed the permission mode picker staying on Auto after a running cloud or Remote Control session refused the switch; it now shows the mode in use and tells you\n- [Cloud sessions] Improved scheduled tasks: runs in Anthropic's cloud now load the plugins enabled on your claude.ai account\n- [Cloud sessions] Improved Test connection for GitHub Enterprise in admin settings to also connect the way cloud sessions do, and to warn you when that second test fails\n- [Cloud sessions] Improved cloud session transcripts: a rename row now names the new title, and repository and environment lists show as a table instead of one line of JSON\n- [Cloud sessions] Changed cloud sessions from the web, desktop and mobile apps to start even when a setup script exits with an error, instead of failing to start\n- [Remote Control] Fixed the permission mode button in Remote Control sessions showing the wrong mode until you typed; while your computer is online it now shows the real mode on open\n- [Claude Tag] Added a dismissible link to the Claude Tag API docs on the Bundles and Channels tabs in admin settings, for Enterprise organizations with Slack connected\n- [Claude Tag] Fixed `@Claude !restart` giving up after one failed workspace check in channels shared across Enterprise Grid workspaces; it now retries, then asks you to run it again\n- [Claude Tag] Fixed a second confirmation DM and channel note when you connect your Claude account from Slack twice from the same link within a few minutes\n- [Claude Tag] Fixed follow-up messages in Slack showing no sign of progress while Claude's session waits for a self-hosted runner; Claude now posts a wait notice in the thread\n- [Claude Tag] Fixed Claude sometimes not answering a tag in its own thread in Slack Connect channels where both organizations have installed Claude\n- [Claude Tag] Fixed Retry on plugin and skill pages in Claude Tag admin settings discarding unsaved access changes after a failed load; your edits now stay in place\n- [Claude Tag] Fixed the repositories control in Claude Tag admin settings showing the same action twice; its menu now lists only what the button doesn't already do\n","publishedAt":"2026-10-30T01:06:26.000Z","fetchedAt":"2026-10-30T01:32:25.811Z","url":"https://code.claude.com/docs/en/changelog#2-1-314","media":[],"coverageCount":0},{"id":"rel_mapQ4E72ZAef5Bh48WVjA","version":"v2.1.312","type":"feature","title":"v2.1.312","summary":"The Write tool previously could overwrite a very large CLAUDE.md, rules file, or @-included file in a subdirectory after seeing only its start; it now asks Claude to read the whole file or edit in place. Also fixed per-directory and paths: skills not being found after a Bash cd deeper into a git repository, invalid command names blocking all command loading and leaving claude -p with no output, and managed settings not being fetched for the rest of the session after signing in from VS Code.","titleGenerated":"Claude Code v2.1.312 stops Write from clobbering large CLAUDE.md files","titleShort":"Write no longer clobbers large CLAUDE.md; per-directory skills fixed","breaking":"minor","importance":3,"content":"## What's changed\n\n- Added `oidc.user_id_claim` to the Claude apps gateway, to identify people by an id_token claim other than `sub`, such as Microsoft Entra ID's `oid`\n- Added click to send: in fullscreen, clicking the \"to send now\" hint under a queued message sends it right away, as its key does\n- Fixed managed settings and organization policy not being fetched for the rest of the session after signing in from the VS Code extension\n- Fixed Claude overwriting a very large CLAUDE.md, rules file or @-included file in a subdirectory after seeing only its start; Write now asks Claude to read it all or edit in place\n- Fixed per-directory and `paths:` skills not being found after a Bash `cd` deeper into a git repository\n- Fixed a rare case where the first reply after resuming a session in the terminal could be missing the next time the session was resumed\n- Fixed a command with an invalid name stopping all commands from loading, which could leave `claude -p` with no output\n- Fixed the Read tool sometimes failing on PDF pages with \"pdftoppm failed: Could not write image\" when the clean-up of old session files ran at the same time\n- Fixed the spinner, OpenTelemetry `effort` values and transcripts naming a newly picked effort level during a task before the model applied it; they now show the level in force\n- Fixed the `/model` and `/effort` cache warning saying the re-read waits for your next message; a change made while Claude is working can trigger it sooner\n- Fixed some sessions showing as waiting on you when the last message only mentioned items waiting on you mid-sentence\n- Fixed agent view restoring input text on the next launch after it was cleared with Ctrl+C just before exiting\n- Fixed agent view reopening in shell mode after exiting with Ctrl+D from an empty `!` input\n- Fixed vim mode `d`, `c` and `y` with a `gj` or `gk` count larger than the lines left deleting or copying text; like Vim, they now only move the cursor\n- Fixed `/verify` showing a long description meant for Claude in `/help` and the `/` command menu instead of a short one\n- Improved the prompt's effort indicator: a change made mid-task now says it may wait for your next message, and the note clears when it no longer applies\n- Improved WebFetch errors in cloud sessions: a host that does not resolve is now explained as possibly outside the allowed network domains, so Claude names the host to allow\n- Changed Bedrock, Vertex, Foundry and Mantle to use the shorter system prompt for model names Claude Code doesn't recognize, such as custom deployment names and inference profile ARNs, even when they front Sonnet or Haiku; set `CLAUDE_CODE_SIMPLE_SYSTEM_PROMPT=0` to keep the full prompt\n- Changed automatic model switches after a flagged message to also cover model names Claude Code doesn't recognize, such as custom deployment names; set `CLAUDE_CODE_DISABLE_REFUSAL_FALLBACK=1` to turn them off\n- Changed usage limit messages to call the 5-hour window the \"5-hour limit\" instead of the \"session limit\"\n- Changed cloud sessions started from the CLI to get their names and branch names from the server\n- Changed prompt and agent hooks to run at most 24 at a time per session, sub-agents included, plus at most 12 that have waited longer than their event's timeout; a `timeout` under 1 second now counts as 1 second\n- Changed a subagent's `autoCompactWindow` to lower the window set by `CLAUDE_CODE_AUTO_COMPACT_WINDOW`, where it had no effect before; it still never raises that window\n- Changed OpenTelemetry `mcp_server.name` to report the real name of fixed-URL remote MCP servers declared by plugins from the official Anthropic marketplace, instead of `custom`\n- Changed Claude Code to stop sending the `claude_code.safety_event` OpenTelemetry event\n- [VSCode] Fixed sessions ending on a bare exit code when your organization's settings couldn't be loaded or were refused at startup; the tab now says why and offers Retry or Sign in again\n- [VSCode] Changed usage limit resets to show a clock time (\"Resets at 3:10 PM\") and to say \"5-hour limit\"\n- [Cloud sessions] Added an effort level to the model picker when you create or edit a routine that runs in the cloud\n- [Cloud sessions] Added a Bash editor with colors, line numbers and a Copy button to the setup script box in cloud environment settings\n- [Cloud sessions] Improved the step shown when Claude adds a repository to a cloud session: opening it now shows a short table of what happened instead of raw JSON\n- [Cloud sessions] Fixed resending an unchanged message after a failed send in a cloud session sometimes running it twice\n- [Cloud sessions] Fixed a failed cloud environment creation showing its error only as a toast outside the open dialog; it now appears inside the dialog\n- [Cloud sessions] Fixed Edit in a routine's menu on the Routines page going to the routine's page without opening its edit dialog\n- [Remote Control] Fixed an unarchived Remote Control session showing as working until a computer reconnected; it now shows as idle\n- [Claude Tag] Improved Claude Tag's retry notice in Slack to name the model in use, or say rate limits were hit, instead of a generic API error\n- [Claude Tag] Fixed a Claude Tag channel's on/off switch and agent settings falling back to workspace defaults when an older private Slack channel goes public and gets a new ID\n- [Claude Tag] Fixed Claude in Slack not replying in channels whose environment has network access set to None\n- [Claude Tag] Fixed Claude in Slack sometimes posting a file in the wrong thread when it was answering in another thread of the same channel\n- [Code Review] Fixed Code Review reviewing an outdated commit, or none at all, when GitHub delivered a push event late; it now reviews the pull request's current commit\n","publishedAt":"2026-10-27T20:24:20.000Z","fetchedAt":"2026-10-27T20:56:52.406Z","url":"https://code.claude.com/docs/en/changelog#2-1-312","media":[],"coverageCount":0},{"id":"rel_SUgZu2mcsNgA4M9QNDQ0a","version":"v2.1.311","type":"feature","title":"v2.1.311","summary":"Fixed long startup and post-edit freezes caused by very long CLAUDE.md list items, and standard-context sessions that hit \"Usage credits required for 1M context\" over 200K tokens now compact and continue instead of stopping. Also fixed interrupt latency during stalled MCP reads, out-of-order transcript messages in claude -p and SDK sessions, a 5-second Write/Edit stall on thousands of lines, and subagents losing worktrees or missing late-connecting MCP servers, plus dozens of smaller fixes.","titleGenerated":"Claude Code v2.1.311 fixes startup freezes and 1M-context billing stops","titleShort":"Long CLAUDE.md freezes fixed; 200K sessions now compact, not stop","breaking":"none","importance":3,"content":"## What's changed\n\n- Added `--hold-idle-file` and `--max-hold-min` to `self-hosted-runner`, to keep an idle session while your host reports processes it left running\n- Added a tip for the Explanatory output style, shown when you ask Claude for explanations\n- Fixed long freezes at startup and after reading or editing files when a CLAUDE.md file holds a very long list item\n- Fixed interrupting Claude not taking effect until a timeout while an MCP server had not answered a resource read or a skill was still downloading\n- Fixed standard-context sessions stopping with \"Usage credits required for 1M context\" when a request ran over 200K tokens; Claude Code now compacts the conversation and continues\n- Fixed Write and Edit stalling for up to 5 seconds after they ran, and reporting 0 lines changed, when changing thousands of lines in a file\n- Fixed subagents being unable to use MCP tools from servers that finish connecting after the subagent starts\n- Fixed Claude seeing stale MEMORY.md and CLAUDE.md content right after automatic compaction when those files were edited earlier in the session\n- Fixed `claude -p` and SDK sessions saving transcript messages out of order or keeping messages that were taken back\n- Fixed `SubagentStop` hooks running about twice a minute for each running subagent instead of once when it finishes\n- Fixed the @ file picker ignoring `.ignore`/`.rgignore` negation rules that re-include git-ignored files\n- Fixed a subagent's isolated worktree being removed while a background subagent it started was still working in it\n- Fixed background agents in auto mode that wait on their own shells starting a new main conversation turn on every wake\n- Fixed the Bash tool description claiming the working directory always persists between calls\n- Fixed the environment block naming a shell commands don't run in, such as fish when the Bash tool uses zsh\n- Fixed `claude mcp add`, `claude mcp remove` and `/mcp` naming the current folder instead of the project entry in `~/.claude.json` that holds a local-scope server when run below a repository root\n- Fixed `claude mcp list` and `claude mcp get` showing the error code twice on failed-connection lines (for example `ENOENT: ENOENT: …`)\n- Fixed Remote Control sessions staying on \"Connecting\" for about two minutes when a network proxy blocks a startup request\n- Fixed messages that Claude sends to a cloud or Remote Control session appearing as your own message in the Claude mobile apps\n- Fixed Claude not being told the cause when a file attached to a message sent over Remote Control or to a cloud session could not be saved because the disk was full, or that a message's attachments past the first 20 were left out\n- Fixed WebFetch and WebSearch hanging on a stalled response in Cowork cloud sessions; an unfinished request now fails after 300 seconds\n- Fixed a background session that crashed between turns not being restarted while it was attached, pinned, or had a scheduled task or wake-up set\n- Fixed the \"background service did not respond\" error suggesting the disabled `claude daemon restart`; it now suggests `claude daemon stop --any` and says it ends background sessions\n- Fixed `claude respawn <id>` starting the background service before reporting that no background session matches the id\n- Fixed `/effort` not warning about a prompt cache miss in plan mode when the plan model differs from the session model\n- Fixed `/ide` saying the VS Code extension was installed before the install finished or when it failed\n- Fixed the Write tool's file preview skipping rows and miscounting hidden lines when a line is wider than the terminal\n- Fixed ↑ in an agent's chat tab recalling a message that was still waiting to be delivered, so resending it sent a second copy\n- Fixed Cowork cloud sessions losing their connectors and most tools after a restart when a plugin or your settings named an `agent`; Cowork cloud sessions now ignore that setting\n- Fixed agent teams sessions showing an empty task list when coming back with `--resume`, `--continue`, `/resume` or `--session-id`, including after `/clear` or a crash, and `/branch` writing to the original session's list\n- Fixed pasting text that mentions an image file, such as `! open /tmp/shot.png` or `/review look at /tmp/shot.png`, attaching the image and dropping its path from the prompt\n- Fixed the feedback draft card showing its 1/2/0 and Esc shortcuts while the prompt has text, when those keys act on the prompt instead of the card\n- Fixed a lone backslash in quotes or brackets, like `\"\\\"` or `(\\)`, disappearing from Claude's responses\n- Fixed agent teammates compacting their history when `DISABLE_COMPACT` is set\n- Fixed the rewind menu offering \"Summarize from here\" and \"Summarize up to here\" when `DISABLE_COMPACT` is set\n- Fixed a finished `/code-review`, or another skill that runs in a sub-agent, running a second time when a cloud or background session restarts\n- Fixed long hidden thinking being cut off as idle on Google Vertex AI, Microsoft Foundry and Claude Platform on AWS, where the endpoint sends thinking progress updates\n- Fixed `/ide` saying it had installed the JetBrains plugin when nothing was installed\n- Fixed Claude Pro and Max subscribers being told to ask their admin when their subscription can't be used; the message now points to billing settings\n- Fixed the diff pane opening by itself after an edit in wide fullscreen terminals when it had nothing to show\n- Fixed `/mcp` showing a remote server as failed after \"Clear authentication\" when the server works without sign-in\n- Fixed `/mcp` showing \"authenticated\" for a remote server that Claude Code sends no credentials to\n- Fixed a `--plugin-dir` folder of plugins loading as one plugin when an empty `.mcp.json` was at its top\n- Fixed `/usage` never showing large prompt-cache misses under \"What's contributing to your limits usage?\"\n- Fixed status line links staying plain text in background sessions after you open them from a terminal that supports links\n- Fixed the /rewind confirmation saying the conversation \"will be forked\" when the later messages are removed from the conversation\n- Fixed the /rewind warning that wrongly said manual and Bash edits are never affected by a code restore\n- Fixed plan approval in screen reader mode sending feedback you had dismissed with Escape\n- Fixed Claude Code retrying requests the API marked not to retry, which could hold a turn for minutes or, in unattended retry mode (`CLAUDE_CODE_RETRY_WATCHDOG`), indefinitely\n- Linux: Fixed sandboxed Bash commands failing with `bwrap: Can't create file` when a `--plugin-dir` folder was also an added directory and `permissions.blockReadsOutsideWorkingDirectories` was on\n- Self-hosted runner: Fixed a session failing to start when a rate limit (HTTP 429, or a git wrapper that exits 75) blocks an added read-only repository; it now starts without that repository\n- Improved startup time for Pro and Max accounts by skipping two network requests on launches within an hour of the last check\n- Improved `/teleport` and `claude --teleport` to warn when a compacted cloud session loads from its last compaction, so the messages it summarized aren't in the local copy\n- Improved the messages shown when a background session's process exits: they no longer promise a restart that isn't coming, and they say how to resume the session\n- Improved artifacts with several views: Claude now routes each view on the link's `#name` and no longer says the address bar stays the same as you switch views\n- Improved shell commands that reach their timeout: the result now shows the output so far and says the command is still running in the background\n- Improved compaction: when the model's reply has no summary in it, Claude Code now usually asks for the summary once more before it replaces the conversation\n- Improved the Artifact tool's choice between a page and a document: explainers and visual or interactive work become artifact pages; documents are for writing people keep editing or pass around\n- Improved startup time for launches whose only arguments are options and their values, such as `claude --model opus`\n- Improved the \"Connection dropped\" API error: it now says where to look first (the proxy or gateway in use, else VPN or firewall; Anthropic API only, Bedrock, Vertex and Foundry are unchanged)\n- Improved screen reader mode: deleting a rule in /permissions says \"Deleted allow rule …\" right away\n- Improved the message shown when a pasted sign-in code is refused: it now says the code may have expired or been used, and to sign in again\n- Improved Slack connector sends: Claude now checks a channel's topic for posting rules before messaging a channel you didn't name\n- Improved screen reader mode: Claude's questions are listed as \"Questions: Color (answered), Pages\", without arrows or symbols\n- Changed `ultrathink` in the interactive terminal to run its turn at xhigh effort on Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.1 through Anthropic's API; typed while Claude is working, it raises the rest of that turn\n- Changed the `ultrathink` rainbow to appear only where the keyword can raise effort, and not when the word is quoted or part of a file path\n- Changed Ctrl+D in agent view to match the prompt: on an empty input, pressing it twice exits\n- Changed `--output-format json` and `stream-json` output to write DEL, C1 control and Unicode format characters inside strings as `\\uXXXX` escapes; parsed values are unchanged\n- Changed model names and IDs to drop the `[1m]` suffix and its label on models where the suffix has no effect; the suffix is still accepted\n- Changed startup warnings about permission rules in settings files to show once per user until the rule changes\n- Changed `/code-review` to tell Claude to put the effort level first and use max only when asked\n- Changed `DISABLE_COMPACT` in settings files and in a mod's `$.env.set` so they can only turn compaction off, not back on\n- Changed Claude Code on Bedrock, Mantle, Vertex, and Foundry to load model settings from a signed file on downloads.claude.ai; set CLAUDE_CODE_MODEL_CATALOG=off to turn this off\n- Changed sessions that set `CLAUDE_CODE_SYNC_SKILLS` with feature-flag fetching off to reuse a claude.ai skills list up to 10 minutes old that another session on the machine fetched\n- Changed clicks on scrollbar arrows and bars in fullscreen lists to act when the mouse button is released, so moving off the scrollbar before letting go cancels the click\n- Changed `claude purge` to say it deletes data only on this computer, not anything stored by Anthropic\n- [VSCode] Added syntax coloring to code blocks in Claude's replies, in light, dark and high contrast dark themes\n- [VSCode] Added a \"Running UserPromptSubmit hooks\" spinner status and screen reader announcement for slow UserPromptSubmit hooks\n- [VSCode] Fixed custom skills and custom slash commands sent from the VS Code or JetBrains panel leaving out the open file or selected lines that ordinary messages include (anthropics/claude-code#102799)\n- [VSCode] Fixed the extension host freezing while the login shell ran its start-up files to read PATH on macOS and Linux\n- [VSCode] Fixed clicking an image in a tool result, or closing its preview, also opening the result's text in an editor tab\n- [VSCode] Fixed voice dictation continuing to record after a message was sent with the Send button; sending now stops dictation first, as pressing Enter already did\n- [VSCode] Changed the extension to start Claude Code sessions with `DISABLE_TELEMETRY` set while VS Code's telemetry setting is off, so they send no usage metrics or error reports and fetch no feature flags\n- [VSCode] Changed the session list side bar to load Web sessions when the Web tab is first opened, not at startup\n- [Cloud sessions] Fixed routines in Anthropic-hosted environments staying locked after their owner left the organization and rejoined; the owner can now turn them back on\n- [Cloud sessions] Fixed a failed routine API token delete or regenerate showing nothing if the edit dialog had already closed; it now shows an error toast\n- [Cloud sessions] Fixed cloud environment dialogs, including the admin dialog, accepting duplicate names; they now refuse a name another listed environment already has\n- [Claude Tag] Added a Skills view to Claude Tag analytics showing how often Claude runs each attached skill in Slack channels and which go unused, with Usage links in admin settings\n- [Claude Tag] Improved Claude Tag routines that run in a new session: a run with nothing to report can now end without posting a throwaway message to the channel\n- [Claude Tag] Changed Claude Tag admin settings to show a lock instead of # for private Slack channels that your own Slack account can see\n- [Code Review] Improved Code Review's check run to say no further automatic review will start when the model still can't analyze a change on the automatic retry\n- [Code Review] Improved Code Review reliability: a review whose session fails to start because of a transient server fault is now retried once automatically, and its check run says so\n","publishedAt":"2026-10-26T21:51:23.000Z","fetchedAt":"2026-10-27T00:43:13.296Z","url":"https://code.claude.com/docs/en/changelog#2-1-311","media":[],"coverageCount":0},{"id":"rel_oxCEzfwGzsmjnvFzgNs_T","version":"v2.1.310","type":"feature","title":"v2.1.310","summary":"Fixed background cleanup deleting recent files such as pasted images when cleanupPeriodDays is set to a very large value, and requests failing behind gateways that reject the effort-2025-11-24 beta value or mid-conversation system messages now retry without them. Added the Monitor tool and plugin monitors on Bedrock, Vertex, and Foundry, plus Enterprise custom roles for cloud sessions and Remote Control.","titleGenerated":"Claude Code v2.1.310 fixes background cleanup deleting recent files","titleShort":"Cleanup no longer deletes recent files; gateway retries hardened","breaking":"none","importance":3,"content":"## What's changed\n\n- Added the Monitor tool and plugin monitors on Bedrock, Vertex, and Foundry, and when telemetry or nonessential traffic is disabled; long foreground sleeps are now blocked there too\n- Fixed requests failing behind some gateways and compatible servers that reject the `effort-2025-11-24` beta value or mid-conversation system messages; Claude Code now retries without them\n- Fixed background cleanup deleting some recent files, such as pasted images, when `cleanupPeriodDays` is set to a very large value\n- Fixed compaction after a \"Prompt is too long\" error keeping too much of a conversation with dense text, such as base64 or SVG, so the next request could fail again\n- Fixed file read permission prompts not showing where a symbolic link leads\n- Fixed a question's auto-continue timeout firing the moment the external editor closed after a long edit\n- Windows: Fixed PreToolUse and PermissionRequest hooks letting a tool call through when Git Bash is missing and the hook's command is only quoted strings, such as a quoted path\n- Improved auto mode's prompt before the first read outside the working directories: it now says when blocking those reads makes shell commands ask for permission\n- Improved memory use of non-interactive sessions (`-p`, Agent SDK) during compaction when the transcript file is very large\n- Improved a mod's `Image` to draw inside tmux with `allow-passthrough` and `focus-events` on, in kitty, Ghostty, and iTerm2 3.5+ with fullscreen rendering, where it showed only its `alt` text\n- Linux: Improved `/sandbox` and the sandbox startup warning to name Ubuntu's AppArmor restriction on user namespaces, and how to lift it, when it keeps bubblewrap from starting the sandbox\n- Changed the error for an MCP tool call that times out in a non-interactive session while the server's elicitation is unanswered: it now tells Claude that no answer came and not to repeat the call\n- [VSCode] Fixed pressing Enter on the plan review card's \"Send feedback and keep planning\" button approving the plan instead of sending your comments back\n- [VSCode] Fixed the panel showing \"Something went wrong\" after \"Switch branch & continue\" on a cloud session opened from Session history's Web tab\n- [Cloud sessions] Added Claude Code cloud sessions and Remote Control permissions to Enterprise custom roles, so Owners can control access to each by role\n- [Cloud sessions] Added a Choose file button to load a cloud environment's setup script from a file\n- [Cloud sessions] Fixed failed saves and archives in the cloud environment edit dialog showing no error inside the dialog\n- [Cloud sessions] Fixed failed routine API token actions showing only a toast; the error now appears in the routine edit dialog\n- [Cloud sessions] Improved the session start checklist to show a cloud environment's setup script output in a scrolling box\n- [Cloud sessions] Changed turning on Cloud sessions or Remote Control in admin settings to ask for confirmation and list the custom roles that have that permission\n- [Cloud sessions] Changed the cloud environment dialog to open network access, setup script and environment variables from rows instead of listing every editor at once\n- [Claude Tag] Fixed Claude Tag treating `!` commands such as `!restart` and `!help` as ordinary messages when sent in bold, italic or strikethrough\n- [Claude Tag] Fixed Claude Tag admin channel and workspace pages being replaced by a load error when a background refresh of the bundle list fails\n- [Claude Tag] Fixed Claude Tag's spend limits page in admin settings reporting that no channels match a filter while channel names are still loading\n- [Code Review] Fixed the Code Review admin section hiding its repositories and switches after a failed refresh of the repository list\n- [Code Review] Improved the Code Review admin section to retry the first load and show one Try again banner when it still fails\n","publishedAt":"2026-10-23T23:39:00.000Z","fetchedAt":"2026-10-23T23:53:58.698Z","url":"https://code.claude.com/docs/en/changelog#2-1-310","media":[],"coverageCount":0},{"id":"rel_ADlLUlVUnJirwpM4j8dgB","version":"v2.1.309","type":"feature","title":"v2.1.309","summary":"Sandboxed Bash commands on macOS and Linux no longer run the shell's startup files outside the sandbox, symlinked macOS denyWrite globs now apply at real paths, and sandboxed commands and auto-accepted edits can no longer modify the MCP config passed with --mcp-config. Also fixes a compaction loop from large nested CLAUDE.md files, lost Claude Desktop session messages on connection refresh, and ctrl+enter queuing behavior.","titleGenerated":"Claude Code v2.1.309 fixes sandbox escapes and macOS startup-file leak","titleShort":"Sandbox escapes closed; MCP config now write-protected","breaking":"major","importance":4,"content":"## What's changed\n\n- Fixed sandboxed Bash commands on macOS and Linux running the shell's startup files outside the sandbox before each command\n- Fixed macOS sandbox `denyWrite` glob patterns not applying at a path's real location when the pattern's folder is reached through a symlink\n- Fixed sandboxed commands and auto-accepted edits being able to modify the MCP config file passed with `--mcp-config`\n- Fixed a disabled MCP server being reconnected by a later permission prompt in print and SDK sessions\n- Fixed the Bash permission check trusting a variable set by a command whose redirect can fail to open\n- Fixed Bash permission checks not prompting for some compound commands that change directory before running git when the `cd` is wrapped or prefixed, as they do for a plain `cd`\n- Fixed an `Edit(path)` deny rule not blocking a Bash redirect to that path in some commands that also held a subshell\n- Fixed worktree-isolated sessions accepting some locale-prefixed commands the isolation guard should refuse\n- Fixed worktree-isolated sessions on zsh accepting some arithmetic commands that the isolation guard should refuse\n- Fixed worktrees and branch checkouts that Claude Code creates so they ignore git replace refs, as its other git operations already do\n- Fixed profile store write protection not applying when the config directory is reached through a symlink\n- Fixed sessions compacting repeatedly after reading files under a very large nested CLAUDE.md; a nested CLAUDE.md over the size limit now loads only its start\n- Fixed ctrl+enter (send now) only queuing a message when pressed before the response had started, and restarting a prompt's hooks while leaving the message queued when pressed while those hooks were running\n- Fixed key-bound slash commands running in the shell from `!` mode\n- Fixed ctrl+o not showing messages from other agents and sessions that are still waiting to be delivered; they now appear in full under the transcript, marked as waiting\n- Fixed long tool output previews being cut through an emoji, leaving session transcripts that `jq` could not parse\n- Fixed Read reporting \"File does not exist\" for PDF page ranges in Linux sandboxes that share the parent's `/proc`\n- Fixed messages from a Claude Desktop-hosted session sometimes being lost when its connection refreshed\n- Fixed features missing from Claude Code sessions in the Claude desktop app when the terminal on the same computer was signed in to a different account or organization\n- Fixed the `claude remote-control` status line showing `HEAD` as the branch for a folder that is not in a git repository\n- Fixed split-pane agent teammates not inheriting the tool deny and `--tools` rules given on the leader's command line\n- Fixed sessions renamed with `/rename` appearing to other sessions under a generated name after a VS Code window reload or a non-interactive `--resume`\n- Fixed the `/rewind` list showing \"No code changes\" for a turn that changed checkpointed files through `/code-review --fix`, a subagent or a shell command\n- Fixed `/privacy-settings` showing an out-of-date \"Help improve our AI models\" value; it now also says when the setting can't be fetched\n- Fixed feedback saying \"try again later\" when your organization does not accept feedback; it now shows the server's reason and where to report a bug\n- Fixed oversized /feedback reports failing instead of being resent without the transcript\n- Fixed a skill in an `--add-dir` folder being unreachable when another skill already held its name; it is now offered as `<folder>:<name>`\n- Fixed `/verify` and `/code-review` not finding a nested package's own skill when its files were changed from the shell\n- Fixed the /claude-api skill saying that `ant beta:worker poll --on-work` passes trace context to the command\n- Fixed the /claude-api skill's C# tracing setup missing spans from the cloud platform clients\n- Fixed the /claude-api skill saying that API spans for requests to Claude Platform on AWS can't join the caller's trace\n- Fixed hooks that run in PowerShell not blocking when the hook's program exits with code 2, and on Windows `PreToolUse` and `PermissionRequest` hooks with `\"shell\": \"powershell\"` failing open when Git for Windows isn't installed\n- Fixed a mod's `turn.step` hook that names a model by alias, such as `haiku`, failing every request with a model error: the alias now resolves to its model\n- Fixed plugins and skills synced from claude.ai being removed, sometimes mid-session, after a disk read error\n- Fixed artifact publishes and updates being refused for some ordinary pages, such as pages with certain minified scripts or artifacts Claude could not read first\n- Fixed Claude asking for permission on each artifact update for the rest of a turn after the conversation was compacted in that turn\n- Fixed fine-grained tool streaming staying off on Amazon Bedrock when `ANTHROPIC_BEDROCK_BASE_URL` is set to the default `bedrock-runtime.<region>.amazonaws.com` endpoint\n- Fixed the \"Waiting for API response\" hint flickering behind gateways that send only keepalive pings: routes off Anthropic's backend now wait 45 seconds\n- Fixed Claude notifying unrelated GitHub users by writing role names and placeholders as mentions in PRs, issues and comments\n- Linux: Fixed sandbox write denies lifting when a background command outlived the session\n- Windows: Fixed PowerShell permission checks auto-approving some `Set-Location` or `cd` commands that target the current directory and carry extra arguments; these now prompt for approval\n- macOS: Fixed the \"Install Command Line Developer Tools\" dialog opening each time Claude Code starts on a Mac without Xcode or the Command Line Tools\n- Self-hosted runner: Fixed `--startup-timeout-min` stopping a session whose first prompt, such as `/code-review`, was still working in a subagent; a frozen process is still stopped\n- Improved the speed of many git commands Claude Code runs for itself on Windows; on every platform, errors from those commands now show in English even when git is set to another language\n- Improved the error when a hooks module is nested too deep to load: it now names the file instead of showing a bare stack-overflow message\n- Improved disk-full handling: a Bash command cut off by a full disk now says so and names the folder to free up, and starting on a full temp volume prints a clear message, not a raw ENOSPC error\n- Improved the /claude-api skill to offer API tracing alongside SDK tracing when asked to add OpenTelemetry\n- Improved the /claude-api skill to check the live docs before saying a feature is unsupported\n- Self-hosted runner: Improved sessions that move to a new runner: Claude now sees the tool call that was interrupted, marked as outcome unknown\n- Updated the bundled dataviz skill to accept APCA as a text-contrast measure (keeping WCAG 2 ratios where AA conformance is required), add an `apca(text, background)` function to its palette validator scripts, and say whether a label inside a colored chart fill should be white or near-black\n- Updated the /claude-api skill: service account keys and Workload Identity Federation tokens can now manage OpenTelemetry destinations\n- Changed `sed -i` to get the standard Bash prompt and run as written, not a simulated file-edit preview; like other Bash edits, it is no longer restored by `/rewind`\n- Changed `claude ultrareview` to require workspace trust: it asks on a terminal, else refuses; scripts pass `--trust-workspace`, which takes no value (put a PR number or base branch before it), or set `CLAUDE_CODE_DISABLE_ULTRAREVIEW_TRUST_CHECK=1`\n- Changed the dangerous `rm` and `rmdir` prompt in bypass permissions mode in the Desktop app and VS Code to deny the command at once with a rewrite hint, so unattended sessions keep going\n- Changed commands the settings exclude from the sandbox to ask before writing to its write-list paths while the sandbox is enabled but cannot run; the unsandboxed fallback is refused there\n- Changed worktree creation and reuse so git no longer fetches objects on demand, in ordinary clones and in partial clones with a Bash sandbox configured; such a partial clone fetches them first\n- Changed the `/claude-api` skill to build Managed Agents subagent teams on `multiagent_20261001` with dynamic workflows on by default, instead of `coordinator`\n- [VSCode] Added a log warning that says what to do when dictation or the extension's own requests fail certificate verification, such as behind a TLS-inspecting proxy\n- [VSCode] Fixed the Claude panel staying blank after the editor cached an incomplete copy of its files; it now recovers by itself, or says what went wrong and how to fix it\n- [VSCode] Fixed Session history filling up with empty \"Untitled\" sessions each time you start a new session without sending a message\n- [VSCode] Fixed long slash-command messages such as `/goal` filling the chat panel; they now fold behind Show more like other prompts (anthropics/claude-code#97469)\n- [VSCode] Fixed long messages not folding behind Show more after the chat panel is made narrower\n- [VSCode] Fixed Ctrl+J in the chat input always inserting a newline: it is now left to VS Code and the OS when \"ctrl+j\" is set to null in the Chat context of ~/.claude/keybindings.json\n- [VSCode] Fixed a mod's \"From the … plugin\" label above the prompt being drawn over the slash command and @-mention menus (anthropics/claude-code#102981)\n- [VSCode] Fixed an empty list item, paragraph or code block, and list items renumbered from 1, when pasting copied LaTeX math\n- [VSCode] Fixed the feedback error showing only an HTTP status when your organization does not accept feedback\n- [VSCode] Improved Edit and Write speed in the chat panel: the wait for new problems after each edit is now 0.75 seconds in most cases, down from 1 to 1.5 seconds\n- [Cloud sessions] Fixed the Add connector list in routine forms saying no connectors were available while they were still loading\n- [Cloud sessions] Fixed the cloud environment editor shuffling environment variables and adding backslashes to values that contain both kinds of quote\n- [Cloud sessions] Fixed the self-hosted environment details dialog in admin settings so its title, tabs and buttons stay in view while only the content scrolls\n- [Cloud sessions] Fixed the \"read documentation\" tool row in cloud sessions, which showed escaped JSON instead of a readable page\n- [Cloud sessions] Improved a cloud session's Changes pane, which now says when it lists only pushed commits or only Claude's edits\n- [Claude Tag] Fixed Claude in Slack posting a draft you approved for a channel into your DM by mistake; a post with no channel is now refused so it can go to the approved channel\n- [Claude Tag] Fixed Allow and Deny on Claude's approval cards in Slack doing nothing when the session had stopped; your answer now restarts the session\n- [Claude Tag] Fixed Claude ignoring untagged messages in a Slack channel after its workspace moved to a different Claude organization, including unshared channels on Enterprise Grid\n- [Claude Tag] Fixed Claude in a Slack channel posting a threaded reply before reading a newer message someone had just added to that thread\n- [Claude Tag] Fixed the Slack page in Claude Tag admin settings saying no workspaces were connected when the list failed to load; it now shows an error with a Retry button\n- [Code Review] Fixed a finished review showing no check on the pull request when earlier tries to create it failed; a final try now adds a completed check\n- [Code Review] Fixed a push queued behind a running review still starting a review after `@claude review stop`; its check now closes and says why\n","publishedAt":"2026-10-22T22:18:57.000Z","fetchedAt":"2026-10-22T23:37:32.348Z","url":"https://code.claude.com/docs/en/changelog#2-1-309","media":[],"coverageCount":0},{"id":"rel_OAVc-fTuUrMlzRWOzCauG","version":"v2.1.308","type":"feature","title":"v2.1.308","summary":"Bug fixes and reliability improvements.","titleGenerated":"Claude Code v2.1.308 bug fixes and reliability improvements","titleShort":"Bug fixes and reliability improvements","breaking":"none","importance":1,"content":"## What's changed\n\n- Bug fixes and reliability improvements\n","publishedAt":"2026-10-22T04:34:55.000Z","fetchedAt":"2026-10-22T07:26:37.175Z","url":"https://code.claude.com/docs/en/changelog#2-1-308","media":[],"coverageCount":0},{"id":"rel_02w8hKiA8kGD49zHtSjtY","version":"v2.1.307","type":"feature","title":"v2.1.307","summary":"Added /claude-api api-version to move Claude API code to API version 2026-10-21, and a claude_code.safety_event OpenTelemetry event for each stopped, prompted, or sent-back tool call. Fixed HTTP MCP tool calls being sent twice when a connection closed mid-call by retrying only read-only or idempotent tools, plus dozens of smaller fixes across MCP, cloud sessions, VSCode, and the self-hosted runner.","titleGenerated":"Claude Code v2.1.307 adds /claude-api api-version and safety_event telemetry","titleShort":"MCP retries now idempotent-only; safety_event telemetry added","breaking":"minor","importance":3,"content":"## What's changed\n\n- Added `/claude-api api-version` to move Claude API code to API version `2026-10-21`\n- Added the `claude_code.safety_event` OpenTelemetry event for each tool call a permission rule, auto mode, a safety check, a hook or the sandbox stopped, prompted or sent back\n- Fixed requests failing with a model-not-found error when `CLAUDE_CODE_DISABLE_1M_CONTEXT` is set and the model is an alias with the `[1m]` suffix, such as `opus[1m]`\n- Fixed the `prompt-caching-scope` beta header being sent behind custom `ANTHROPIC_BASE_URL` gateways and on Foundry, where it has no effect\n- Fixed MCP servers showing \"needs authentication\" in `claude -p` on macOS while the login keychain was locked; they now say why, and the next run after an unlock connects without a new sign-in\n- Fixed MCP servers being reported as timed out, and staying failed in headless runs, when a macOS or Linux computer went to sleep while they were connecting\n- Fixed HTTP MCP tool calls being sent twice when the connection closed mid-call; only read-only or idempotent tools are now retried\n- Fixed a turn sometimes starting without an SDK MCP server's latest tools when a background task was running\n- Fixed some \"Autocompact is thrashing\" errors in turns that read many PDF pages, images or other large tool results\n- Fixed `/autocompact` in cloud sessions going back to the default window after the session restarted\n- Fixed a conversation cleared with `/clear` sometimes coming back when a cloud or self-hosted runner session restarted\n- Fixed `CLAUDE_STREAM_FIRST_BYTE_TIMEOUT_MS` being ignored in Claude apps gateway sessions\n- Fixed telemetry from Claude Desktop sessions behind a Claude apps gateway being rejected when `OTEL_EXPORTER_OTLP_HEADERS` is set\n- Fixed `/permissions` > Recently denied not showing the denied shell command, and auto mode's local classifier ignoring an approval from there when it checks a retried command; approving a row now names the command, not its description\n- Fixed `/help` and the `/` command menu showing long model-facing descriptions for bundled skills such as `/claude-api` instead of their short descriptions\n- Fixed subagent transcripts recording placeholder token usage and an empty stop reason for streamed responses\n- Fixed model switches through the Agent SDK's `setModel()`, and apps built on it, being refused when the API took more than 5 seconds to confirm the model; Claude Code now waits up to 15 seconds\n- Fixed a background command stopped through a connected app (such as the desktop app archiving an idle session) being reported to Claude as stopped by the user; the notice now names its output file\n- Fixed WebFetch in Cowork cloud sessions reporting a tool error when a site answered with a 4xx status or a robots.txt refusal; it is now a normal result, as in local sessions\n- Fixed write tools on a directly configured GitHub MCP server showing a form that closed without creating anything in the desktop app's Code tab\n- Fixed Artifact tool reads failing with a connection reset when the server had closed a reused connection\n- Self-hosted runner: Fixed sessions with `--use-anthropic-git-proxy` committing a `.claude/settings.local.json` left in the clone by an earlier session\n- Self-hosted runner: Fixed archiving or requeueing a session, or stopping the runner, during its checkout hook leaving the hook's processes running and the runner's slot held\n- Improved recovery from unparseable tool calls: Claude Code now retries the call up to twice before ending the turn with an error\n- Improved fork subagents launched together: each is now told which launch is its own, so forks are much less likely to try to start each other's work\n- Improved browser sign-in: the success page now appears as soon as sign-in completes, without waiting for your account profile to load\n- Improved web search errors in cloud and self-hosted runner sessions: a search refused with a 403 that gives a reason now shows that reason instead of a generic \"Authentication error\"\n- Improved the time to the first reply in cloud sessions when one of the desktop app's MCP servers can't connect\n- Improved Remote Control connect time for sessions whose transcript contains large pasted images or PDF reads\n- Changed the notice Claude gets when the WebSearch budget is used up to also tell it not to query search engines in a browser or through computer use\n- Changed the `axScreenReader` setting and `CLAUDE_AX_SCREEN_READER` in a settings `env` block to apply only from user, managed or `--settings` settings (since 2.1.305)\n- Self-hosted runner: Changed checkout hooks to be stopped after 30 minutes by default, failing the session's setup; if yours needs longer, raise `--checkout-hook-timeout-min` or set it to `0`\n- Self-hosted runner: Changed the certificate file built for Anthropic-managed git to filter its trusted certificates as the CLI does; a `[runner:warn]` line gives the number left out\n- [VSCode] Added a notice on the microphone button when voice dictation stops by itself (15 seconds without recognized speech, or the 2-minute limit), saying why and that the text so far is kept\n- [VSCode] Fixed a long prompt expanded with Show more covering the whole chat panel: a pinned prompt is now capped at 40% of the panel and scrolls\n- [VSCode] Fixed part of a list item or table cell with LaTeX math pasting as a new list or table in rich text editors\n- [VSCode] Fixed dictation failing behind a TLS-inspecting proxy when `NODE_EXTRA_CA_CERTS` is set only in `~/.claude/settings.json`; it now also follows `CLAUDE_CODE_CERT_STORE`\n- [VSCode] Changed the \"still open somewhere else\" notice to also cover a Claude process left over from an extension restart, which usually ends on its own within about 10 minutes\n- [VSCode] Changed voice dictation to filter its trusted certificates as the CLI does\n- [VSCode] Changed the extension's own requests to filter their trusted certificates as the CLI does, and to trust the `NODE_EXTRA_CA_CERTS` file named in the environment or in user settings\n- [Cloud sessions] Fixed the repository picker saying \"No repositories match\" while a search was still running; it now shows \"Searching…\" until the results arrive\n- [Cloud sessions] Fixed the \"Waiting for a runner\" status on self-hosted environments being cut off; with on-demand runners it now counts up from your last message and shows failed attempts\n- [Code Review] Added `@claude review stop` (also `off` and `pause`) to Code Review, ending an `@claude review always` subscription on a pull request without starting a new review\n","publishedAt":"2026-10-21T22:17:02.000Z","fetchedAt":"2026-10-21T23:21:08.116Z","url":"https://code.claude.com/docs/en/changelog#2-1-307","media":[],"coverageCount":0}],"pagination":{"nextCursor":"2026-10-21T22:17:02.000Z|2026-10-21T23:21:08.116Z|rel_02w8hKiA8kGD49zHtSjtY","limit":20},"summaries":{"rolling":{"windowDays":90,"summary":"Claude Code has shifted toward giving users more granular control over how Claude works — tuning effort levels, customizing interface behavior, and enabling powerful features like background agents and comprehensive code review. The `/effort` command now spans from low to max with interactive tuning, auto mode matches terminal themes, and `/ultrareview` runs cloud-based multi-agent code critique. Teams and enterprise customers gained OS certificate store trust by default, subprocess sandboxing with PID namespace isolation on Linux, and the ability to create managed policy fragments in `managed-settings.d/`. The product also expanded automation: `/loop` handles recurring intervals with cron scheduling, `/powerup` teaches features via animated lessons, and background agents (spawned as tasks with dependency tracking) run independently while the main session continues. Memory and session handling improved significantly—auto-memory now persists useful context, `/recap` provides context when returning to sessions, and prompt caching got longer TTLs (1-hour option available via environment variable). Plugin infrastructure matured with monitor support for background events, conditional hook execution via `if` fields, and persistent state via `${CLAUDE_PLUGIN_DATA}`. A batch of security fixes hardened bash tool permission handling against environment-variable and network-redirect bypasses, while terminal rendering shifted to flicker-free mode and pull-based status updates reduced CPU churn.","releaseCount":77,"generatedAt":"2026-04-17T21:07:53.656Z"},"monthly":[{"year":2026,"month":3,"summary":"March shipped steady stability work and infrastructure improvements. Prompt cache reliability saw multiple fixes—invalidation with MCP instructions, mid-session schema drift, and nested CLAUDE.md re-injection in long sessions. The PermissionDenied hook and conditional `if` field for rules expanded control over auto mode behavior, while flicker-free alt-screen rendering via `CLAUDE_CODE_NO_FLICKER` addressed rendering issues. Memory leaks from LRU cache keys, render caches, and streaming buffers were plugged to sustain long-running sessions, and the structured-output schema cache bug that halved workflow reliability was resolved. Windows got attention with CRLF handling on Edit/Write tools and line-ending preservation for Markdown hard breaks.","releaseCount":21,"generatedAt":"2026-04-07T14:38:02.666Z"},{"year":2026,"month":2,"summary":"February pushed toward reliability and multi-agent workflows. Auto-memory shipped to capture and recall context across sessions, while agent teams and background agents enabled coordinated multi-agent collaboration—both preview features. The month prioritized stability through sweeping memory leak fixes (bridge polling, MCP OAuth, file caching, git detection), Windows crash fixes (concurrent config writes, process spawning, ARM64 stability), and terminal rendering corrections. Sonnet 4.6 arrived with fast mode support and full 1M context, and HTTP hooks plus manual OAuth fallback improved MCP server connectivity.","releaseCount":28,"generatedAt":"2026-03-31T14:03:22.292Z"}]}}