{"slug":"coding-agents","name":"Coding Agents","description":"AI pair programmers and autonomous coding agents — IDE-integrated assistants and cloud agents that read, edit, and ship code.","isFeatured":true,"dailySummaryEnabled":true,"weeklyDigestEnabled":true,"members":[{"kind":"product","slug":"claude-code","name":"Claude Code","description":"AI coding agent for the terminal and IDE","org":{"slug":"anthropic","name":"Anthropic","domain":"anthropic.com","avatarUrl":"https://media.releases.sh/orgs/anthropic.png","githubHandle":"anthropics"}},{"kind":"product","slug":"grok-build","name":"Grok Build","description":"AI coding agent for the terminal (Grok CLI).","org":{"slug":"xai","name":"xAI","domain":"x.ai","avatarUrl":"https://media.releases.sh/orgs/xai.png","githubHandle":"xai-org"}},{"kind":"product","slug":"codex","name":"Codex","description":null,"org":{"slug":"openai","name":"OpenAI","domain":"openai.com","avatarUrl":"https://media.releases.sh/orgs/openai.png","githubHandle":"openai"}},{"kind":"org","slug":"cursor","name":"Cursor","domain":"cursor.com","avatarUrl":"https://media.releases.sh/orgs/cursor.png","githubHandle":"getcursor","description":"AI-powered code editor built on VS Code that integrates large language models directly into the development workflow."},{"kind":"product","slug":"github-copilot","name":"GitHub Copilot","description":null,"org":{"slug":"github","name":"GitHub","domain":"github.com","avatarUrl":"https://media.releases.sh/orgs/github.png","githubHandle":"github"}},{"kind":"product","slug":"devin","name":"Devin","description":null,"org":{"slug":"cognition","name":"Cognition","domain":"cognition.ai","avatarUrl":"https://media.releases.sh/orgs/cognition.png","githubHandle":"Cognition-ai"}},{"kind":"org","slug":"windsurf","name":"Windsurf","domain":"windsurf.com","avatarUrl":"https://media.releases.sh/orgs/windsurf.png","githubHandle":"Exafunction","description":"AI-powered code editor that keeps developers in flow state with local and cloud coding agents."}],"orgs":[{"slug":"cursor","name":"Cursor","domain":"cursor.com","avatarUrl":"https://media.releases.sh/orgs/cursor.png","githubHandle":"getcursor","description":"AI-powered code editor built on VS Code that integrates large language models directly into the development workflow."},{"slug":"windsurf","name":"Windsurf","domain":"windsurf.com","avatarUrl":"https://media.releases.sh/orgs/windsurf.png","githubHandle":"Exafunction","description":"AI-powered code editor that keeps developers in flow state with local and cloud coding agents."}],"releases":[{"id":"rel_p7wr5WVE-FeHvSq2EF5D_","version":null,"type":"feature","title":"November 13, 2026","summary":"The session file editor no longer saves automatically; a dot marks unsaved files and Cmd/Ctrl+S writes to Devin's machine, telling Devin which files you saved. Deutsch is now selectable in Preferences, the Agent/Ask switch moved into the composer, and workspace tabs no longer open on their own, collectible instead in an artifacts rail beside the chat.","titleGenerated":"Devin adds German UI and Cmd/Ctrl+S saving in the session file editor","titleShort":"File editor saves on Cmd+S; Devin UI speaks German","importance":3,"content":"**German UI Language**\n\nDeutsch can now be selected as the display language in Preferences, using the terms engineers use for session, merge, push, prompt, logs, service user and secrets.\n\n**Session File Editor Saves on Cmd/Ctrl+S**\n\nThe session file editor no longer saves automatically: press Cmd/Ctrl+S to save, a dot on the tab marks unsaved files, a cloud icon in the editor header confirms each save to Devin's machine, and closing a tab or leaving a session with unsaved edits asks first. Devin is told which files you saved so it can take your edits into account.\n\n**Sidebar Filter Menu Simplified**\n\nThe sessions sidebar Filter menu no longer has the Machine type and Skill filters; both remain available on the Sessions page. The session title menu now lists the sessions above the current session's parent.\n\n**Artifacts Rail Beside the Chat**\n\nWorkspace tabs no longer open on their own as Devin works. What a session produces — pull requests, Changes, previews, the Desktop and Agents views, Plan, citations and attachments — is listed in a compact rail to the right of the chat; hover to see the labels, click a row to open it. The workspace pane remembers per session whether it was open, and a session starts with it closed. Tabs that are not artifacts (Findings, Inbox, Todo…) stay in the add-tab menu, and `?tab=` links still open the tab they name.\n\n**Agent / Ask Moves Into the Composer**\n\nThe Agent / Ask switch moves from the header into a dropdown inside the composer. The same dropdown holds the modes for each side — Agent modes (Code, Data) or Ask modes (Auto, Q&A, Plan) — and switching Agent/Ask flips the composer immediately with that side's default mode selected.\n\n**Review Analytics**\n\nAnalytics > Review now shows what became of Review's findings (fixed, rejected, ignored) for the built-in review and for each custom reviewer, and how each version of a custom reviewer's instructions performed.\n\n**GitLab Merge Trains**\n\nDevin Review can now add GitLab merge requests to their project's merge train. Editing a pending inline comment on a GitLab merge request keeps it at its original location.\n\n**Find Skips Collapsed Files**\n\nFind in Devin Review now skips collapsed files by default. Turn off \"Skip collapsed files\" in the find filter menu to search them; the choice is remembered.\n\n**Chapter Guide File Cards**\n\nReview's chapter guide lists changed files in chapter order with clearer filenames and directory paths; mark individual files or entire chapters as viewed without leaving the guide. Selecting \"View as sections\" restores the guide's chapters even if you previously saved File tree view.\n\n**Keyboard: Auto-merge Shortcut and Shortcuts Dialog**\n\nPressing `m` in standalone Devin Review now enables auto-merge when the button offers it, and the Review keyboard shortcuts dialog no longer repeats identical key combinations.\n\n**Clearer Status When a PR Updates or a Code Host Is Slow**\n\nDevin Review distinguishes a newly pushed commit from an existing commit that is ready to view and from analysis updates that leave the code unchanged. When GitHub will not accept a Review comment because the file or line is no longer in the diff, Review says so instead of a generic error. When a code host is slow or not answering, Review and the session's pull request tab say the code host did not respond and keep retrying (after one attempt, not three) instead of saying you have no access to the repository.\n\n**Unlink Devin from a Finding**\n\nYou can unlink Devin from a security finding from the findings table or the finding's details, so it can be fixed again.\n\n**Copy as Markdown Includes Validation and Review Reasons**\n\nCopy as Markdown on a finding now includes the sandbox validation result and evidence, and the reason the finding was reviewed or dismissed. A board-backed sealed finding can be edited through the content tool, with the changes shown in Activity.\n\n**Security Sidebar**\n\nFix sessions are listed in their own Sessions section below Scans instead of nested under each scan or connection, with Scans showing 5 rows at a time; the empty Sessions list matches the main sidebar's wording; and the \"Introducing Devin Security\" dialog no longer appears over a session page.\n\n**Turn Off Automatic CI Follow-up for your Organization**\n\nOrg admins can turn off Devin's automatic CI follow-up under Settings > Customization > Pull requests, and reach the switch from the CI checks tile on Cost Breakdown. It stays on by default, and asking Devin directly still works.\n\n**Session Details in PR Descriptions Can Be Turned Off**\n\nAdmins can turn off the automatic session links and requester details in Devin-generated PR descriptions from Devin Sessions settings. The setting is on by default and takes effect when sessions start or resume.\n\n**Organization Picture for Every Organization**\n\nAdmins of non-enterprise organizations can now upload or remove an organization picture in Settings > General.\n\n**Plugin MCP Server URLs: Region Choices and Readable Settings**\n\nA plugin can now offer a choice of regions for its MCP server (Datadog uses this for its sites), and that choice is shown before the first sign-in instead of defaulting to one region. The Server URL field shows the part you type as its own input and lists long query parameters one per row. When a plugin moves its server to a new URL, existing installations connect at the new URL without reconnecting.\n\n**Enterprise MCP Editor Asks for a Required OAuth Client**\n\nWhen adding an MCP server from the catalog that needs your own OAuth client (such as Google Calendar) for the whole enterprise, the form opens the OAuth credentials section, shows the setup steps and asks for the client ID before the server can be created, instead of failing silently.","publishedAt":"2026-11-14T13:08:54.000Z","url":"https://docs.devin.ai/release-notes/overview#november-13-2026","media":[],"prerelease":false,"source":{"slug":"release-notes","name":"Release Notes","type":"feed"},"org":{"slug":"cognition","name":"Cognition"},"product":{"slug":"devin","name":"Devin"},"groupSlug":"devin","groupName":"Devin","coverageCount":0,"composition":{"bugs":0,"features":8,"enhancements":12}},{"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","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","url":"https://code.claude.com/docs/en/changelog#2-1-325","media":[],"prerelease":false,"source":{"slug":"claude-code-releases","name":"Claude Code","type":"scrape"},"org":{"slug":"anthropic","name":"Anthropic"},"product":{"slug":"claude-code","name":"Claude Code"},"groupSlug":"claude-code","groupName":"Claude Code","coverageCount":0,"composition":{"bugs":45,"features":4,"enhancements":40}},{"id":"rel_pcUfMj9PpW1o7WnzxlTo8","version":"0.169.0","type":"feature","title":"Codex CLI Release: 0.169.0","summary":"The opt-in terminal workspace now supports multiple side-chat tabs that persist across restarts, and custom model providers can be added or edited from the terminal, including device-code gateway sign-in over SSH. `codex exec` now keeps project configuration and MCP servers disabled in untrusted directories even with `--sandbox workspace-write`, requiring `--trust` to save project trust.","titleGenerated":"Codex CLI 0.169.0 adds multiple side-chat tabs and custom model providers","titleShort":"Side-chat tabs multiply; custom model providers from terminal","importance":4,"content":"## New Features\n\n- The opt-in terminal workspace supports multiple side-chat tabs. Switch between them with the keyboard or mouse, and reopen saved local side chats after restarting Codex. ([#59784](https://github.com/openai/codex/pull/59784), [#60152](https://github.com/openai/codex/pull/60152), [#60181](https://github.com/openai/codex/pull/60181))\n- Search older tasks from the agent command center by name, prompt preview, or working directory. ([#60286](https://github.com/openai/codex/pull/60286))\n- Turn on **Collapse turn details** in the full-screen transcript to fold completed activity beneath each prompt while keeping answers visible and tool output searchable. ([#60554](https://github.com/openai/codex/pull/60554), [#60555](https://github.com/openai/codex/pull/60555), [#60556](https://github.com/openai/codex/pull/60556), [#60557](https://github.com/openai/codex/pull/60557))\n- Add or edit custom model providers from the terminal. With custom gateway onboarding enabled, you can create a provider during sign-in and use a device code for supported gateways over SSH or without a local browser. ([#59779](https://github.com/openai/codex/pull/59779), [#60009](https://github.com/openai/codex/pull/60009), [#60186](https://github.com/openai/codex/pull/60186), [#60486](https://github.com/openai/codex/pull/60486), [#60662](https://github.com/openai/codex/pull/60662))\n- Press `Ctrl+S` in the terminal report editor to save a diagnostic ZIP locally, even when uploads are disabled. ([#60542](https://github.com/openai/codex/pull/60542))\n- With the opt-in Cloud connection on macOS and Linux, send images, choose a model and reasoning effort, and archive or restore Cloud conversations from the terminal. ([#59828](https://github.com/openai/codex/pull/59828), [#60061](https://github.com/openai/codex/pull/60061), [#60635](https://github.com/openai/codex/pull/60635), [#60651](https://github.com/openai/codex/pull/60651))\n\n## Bug Fixes\n\n- `codex exec` keeps project configuration and MCP servers disabled in untrusted directories, even with `--sandbox workspace-write`. Use `--trust` with `codex exec` or `codex review` to explicitly save project trust. ([#60041](https://github.com/openai/codex/pull/60041))\n- **Yes, clear context and implement** now hands the full approved plan to the fresh session and keeps the current session open if the handoff fails. ([#60763](https://github.com/openai/codex/pull/60763))\n- Fixed Windows sandbox handling for quoted `cmd.exe` paths and for PowerShell sessions whose working directory has an unreadable parent. ([#60624](https://github.com/openai/codex/pull/60624), [#60647](https://github.com/openai/codex/pull/60647))\n- Pasted images on macOS retain their intended orientation and use native PNG or JPEG clipboard data when available. ([#60455](https://github.com/openai/codex/pull/60455))\n- The terminal no longer stays busy waiting for MCP servers that a shared app server did not report. ([#60280](https://github.com/openai/codex/pull/60280), [#60622](https://github.com/openai/codex/pull/60622))\n- Interrupted turns now close cleanly, and delayed aborts or file approval cancellations no longer interrupt a later turn. ([#59910](https://github.com/openai/codex/pull/59910), [#60700](https://github.com/openai/codex/pull/60700))\n\n## Documentation\n\n- Updated the `openai-docs` skill to focus on Codex and ChatGPT Work questions and first-time API setup, with dedicated guidance for configuring Codex across a team. ([#59944](https://github.com/openai/codex/pull/59944), [#59949](https://github.com/openai/codex/pull/59949))\n- Documented how app-server clients can list models or authorize a gateway for a configured provider without changing the default provider. ([#59776](https://github.com/openai/codex/pull/59776), [#59819](https://github.com/openai/codex/pull/59819))\n\n## Chores\n\n- Updated bundled SQLite to 3.54 to pick up write-ahead log fixes. ([#60432](https://github.com/openai/codex/pull/60432))\n- Plugin marketplace upgrades can reuse Git objects already on disk, reducing repeat downloads. ([#60528](https://github.com/openai/codex/pull/60528))\n\n## Changelog\n\nFull Changelog: [rust-v0.168.0...rust-v0.169.0](https://github.com/openai/codex/compare/rust-v0.168.0...rust-v0.169.0)\n\n- [#59739](https://github.com/openai/codex/pull/59739) Add a sidebar chooser and floating workspace shortcut panels [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59743](https://github.com/openai/codex/pull/59743) Forward explicit cyber access programs with header authentication [@jamy-OAI](https://github.com/jamy-OAI)\n- [#59745](https://github.com/openai/codex/pull/59745) Repair legacy Windows sandbox bin permissions for SYSTEM [@zm-oai](https://github.com/zm-oai)\n- [#59746](https://github.com/openai/codex/pull/59746) Add typed Cloud user-message submission receipts [@bc-openai](https://github.com/bc-openai)\n- [#59749](https://github.com/openai/codex/pull/59749) Use runtime authentication for OpenAI MCP and app requests [@cooper-oai](https://github.com/cooper-oai)\n- [#59750](https://github.com/openai/codex/pull/59750) Stop Noise relay recovery on rejected executor credentials [@richardopenai](https://github.com/richardopenai)\n- [#59752](https://github.com/openai/codex/pull/59752) Configure the Opus Bazel patch and isolate auth tests [@bkotsopoulos](https://github.com/bkotsopoulos)\n- [#59754](https://github.com/openai/codex/pull/59754) Enforce host-required cyber access programs per thread [@jamy-OAI](https://github.com/jamy-OAI)\n- [#59759](https://github.com/openai/codex/pull/59759) Update WebSocket dependency forks [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#59767](https://github.com/openai/codex/pull/59767) Add sidebar tabs and nonblocking side-chat creation [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59768](https://github.com/openai/codex/pull/59768) Stop prefetching plugin discovery at app-server startup [@zamoshchin-openai](https://github.com/zamoshchin-openai)\n- [#59769](https://github.com/openai/codex/pull/59769) Reuse decoded WebSocket event kinds in telemetry [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#59770](https://github.com/openai/codex/pull/59770) Add shared service-tier support for host-owned agents [@rhan-oai](https://github.com/rhan-oai)\n- [#59772](https://github.com/openai/codex/pull/59772) Remove the legacy sandbox policy constructor [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59774](https://github.com/openai/codex/pull/59774) Remove the legacy Seatbelt command wrapper [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59776](https://github.com/openai/codex/pull/59776) Allow selecting a configured provider in `model/list` [@abhinav-oai](https://github.com/abhinav-oai)\n- [#59778](https://github.com/openai/codex/pull/59778) Expose acceptance metadata and code-cell links in rollout traces [@tz-openai](https://github.com/tz-openai)\n- [#59779](https://github.com/openai/codex/pull/59779) Add provider configuration editing to the TUI [@alexsong-oai](https://github.com/alexsong-oai)\n- [#59780](https://github.com/openai/codex/pull/59780) Enable safe identical code folding for x86-64 musl releases [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59783](https://github.com/openai/codex/pull/59783) Tie gRPC code-mode connection lifetime to live sessions [@frantic-openai](https://github.com/frantic-openai)\n- [#59784](https://github.com/openai/codex/pull/59784) Support multiple side chats in the TUI sidebar [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59785](https://github.com/openai/codex/pull/59785) Decouple message lifecycle state from model-visible item IDs [@euroelessar](https://github.com/euroelessar)\n- [#59787](https://github.com/openai/codex/pull/59787) Support credential export commands for gateway authentication [@alexsong-oai](https://github.com/alexsong-oai)\n- [#59789](https://github.com/openai/codex/pull/59789) Render delegated thread messages as Markdown in the TUI [@bc-openai](https://github.com/bc-openai)\n- [#59792](https://github.com/openai/codex/pull/59792) Centralize managed proxy preparation and sandbox routing [@viyatb-oai](https://github.com/viyatb-oai)\n- [#59793](https://github.com/openai/codex/pull/59793) Support update-boundary paging across forked thread histories [@owenlin0](https://github.com/owenlin0)\n- [#59794](https://github.com/openai/codex/pull/59794) Check extension budgets before model generation requests [@rhan-oai](https://github.com/rhan-oai)\n- [#59795](https://github.com/openai/codex/pull/59795) Support optional thread IDs in hosted file uploads [@lt-oai](https://github.com/lt-oai)\n- [#59805](https://github.com/openai/codex/pull/59805) Pack relative relocations in x86-64 musl release builds [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59816](https://github.com/openai/codex/pull/59816) Retry gateway credential exports after unauthorized responses [@alexsong-oai](https://github.com/alexsong-oai)\n- [#59817](https://github.com/openai/codex/pull/59817) Handle TUI message submission receipts and preserve uncertain drafts [@bc-openai](https://github.com/bc-openai)\n- [#59819](https://github.com/openai/codex/pull/59819) Allow provider selection for gateway OAuth reads and logins [@abhinav-oai](https://github.com/abhinav-oai)\n- [#59822](https://github.com/openai/codex/pull/59822) Add device-code authorization for gateway OAuth [@smurching-oai](https://github.com/smurching-oai)\n- [#59823](https://github.com/openai/codex/pull/59823) Bind runtime auth to the provider's selected credential source [@cooper-oai](https://github.com/cooper-oai)\n- [#59824](https://github.com/openai/codex/pull/59824) Enable resizing the sidebar chooser [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59826](https://github.com/openai/codex/pull/59826) Preserve client metadata through queued submissions [@eddie-openai](https://github.com/eddie-openai)\n- [#59828](https://github.com/openai/codex/pull/59828) Add a gated Cloud backend to the TUI [@bc-openai](https://github.com/bc-openai)\n- [#59832](https://github.com/openai/codex/pull/59832) Add opt-in incremental tools for regular OpenAI Responses [@akira-oai](https://github.com/akira-oai)\n- [#59842](https://github.com/openai/codex/pull/59842) Carry targeted moderation decisions on final-answer messages [@psachdeva-oai](https://github.com/psachdeva-oai)\n- [#59887](https://github.com/openai/codex/pull/59887) Remove incomplete prereleases after asset upload failures [@jif-oai](https://github.com/jif-oai)\n- [#59891](https://github.com/openai/codex/pull/59891) Bound REPL provenance costs in elastic context preparation [@jif-oai](https://github.com/jif-oai)\n- [#59893](https://github.com/openai/codex/pull/59893) Recompute the approval reviewer when switching models [@jif-oai](https://github.com/jif-oai)\n- [#59902](https://github.com/openai/codex/pull/59902) Keep verified question/answer records whole under elastic budgets [@jif-oai](https://github.com/jif-oai)\n- [#59903](https://github.com/openai/codex/pull/59903) Delete incomplete prereleases by release ID [@jif-oai](https://github.com/jif-oai)\n- [#59905](https://github.com/openai/codex/pull/59905) Validate implicit API keys for Guardian Decisions [@jif-oai](https://github.com/jif-oai)\n- [#59908](https://github.com/openai/codex/pull/59908) Avoid full image decoding for original-detail context estimates [@jif-oai](https://github.com/jif-oai)\n- [#59910](https://github.com/openai/codex/pull/59910) Scope file change approval cancellation to the requesting turn [@jif-oai](https://github.com/jif-oai)\n- [#59913](https://github.com/openai/codex/pull/59913) Resolve Guardian policy from the reviewed thread's catalog [@jif-oai](https://github.com/jif-oai)\n- [#59914](https://github.com/openai/codex/pull/59914) Bind Guardian Decisions requests to their credential source [@jif-oai](https://github.com/jif-oai)\n- [#59928](https://github.com/openai/codex/pull/59928) Improve sidebar chat selection and tab hover highlighting [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59933](https://github.com/openai/codex/pull/59933) Use permission profiles for protected-model startup downgrades [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59944](https://github.com/openai/codex/pull/59944) Narrow openai-docs to product Q&A and first-time API setup [@provencher-oai](https://github.com/provencher-oai)\n- [#59949](https://github.com/openai/codex/pull/59949) Route broad Codex setup questions through self-knowledge guidance [@vb-openai](https://github.com/vb-openai)\n- [#59954](https://github.com/openai/codex/pull/59954) Resolve skill display names in read and installations tools [@jif-oai](https://github.com/jif-oai)\n- [#59955](https://github.com/openai/codex/pull/59955) Bump `ignore` to `0.4.30` [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#59958](https://github.com/openai/codex/pull/59958) Expose model metadata from each thread's catalog and settings [@seanh-oai](https://github.com/seanh-oai)\n- [#59960](https://github.com/openai/codex/pull/59960) Skip the network denial grace period when remote decisions drain [@jif-oai](https://github.com/jif-oai)\n- [#59963](https://github.com/openai/codex/pull/59963) Normalize turn context permissions to a required profile [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59969](https://github.com/openai/codex/pull/59969) Reduce contention in thread writer lock coordination [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59972](https://github.com/openai/codex/pull/59972) Fix Opus header installation with relative symlink inputs [@bc-openai](https://github.com/bc-openai)\n- [#59973](https://github.com/openai/codex/pull/59973) Enable gRPC and host output truncation by default for Code Mode [@seanh-oai](https://github.com/seanh-oai)\n- [#59974](https://github.com/openai/codex/pull/59974) Test command completion before network denial grace expires [@jif-oai](https://github.com/jif-oai)\n- [#59980](https://github.com/openai/codex/pull/59980) Test rejection of PNGs with readable dimensions but corrupt pixels [@fjord-oai](https://github.com/fjord-oai)\n- [#59983](https://github.com/openai/codex/pull/59983) Render typed prompts as Markdown in the TUI [@bc-openai](https://github.com/bc-openai)\n- [#59987](https://github.com/openai/codex/pull/59987) Resolve startup models using thread creation options [@seanh-oai](https://github.com/seanh-oai)\n- [#59988](https://github.com/openai/codex/pull/59988) Add inherited styling to TUI display trees [@joesavona](https://github.com/joesavona)\n- [#59991](https://github.com/openai/codex/pull/59991) Preserve app invocation context across MCP tool refreshes [@fahad-OAI](https://github.com/fahad-OAI)\n- [#59993](https://github.com/openai/codex/pull/59993) Add fixed read views for local message boards [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59994](https://github.com/openai/codex/pull/59994) Add read-only evidence retrieval tools for Guardian reviewers [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59995](https://github.com/openai/codex/pull/59995) Add subscription tools for Guardian reviewers [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59998](https://github.com/openai/codex/pull/59998) Add atomic bulk channel subscription updates [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59999](https://github.com/openai/codex/pull/59999) Add model-driven Guardian channel selection [@felixxia-oai](https://github.com/felixxia-oai)\n- [#60000](https://github.com/openai/codex/pull/60000) Plan Guardian evidence delivery in source order [@felixxia-oai](https://github.com/felixxia-oai)\n- [#60001](https://github.com/openai/codex/pull/60001) Add bounded original-evidence delivery for Guardian reviewers [@felixxia-oai](https://github.com/felixxia-oai)\n- [#60002](https://github.com/openai/codex/pull/60002) Decouple Guardian retrieval from local message-board handles [@felixxia-oai](https://github.com/felixxia-oai)\n- [#60003](https://github.com/openai/codex/pull/60003) Apply runtime authentication to realtime connections [@cooper-oai](https://github.com/cooper-oai)\n- [#60004](https://github.com/openai/codex/pull/60004) Expose ChatGPT user IDs in `account/read` responses [@dand-oai](https://github.com/dand-oai)\n- [#60006](https://github.com/openai/codex/pull/60006) Reduce SQLite write overhead for thread activity and history [@jonluca](https://github.com/jonluca)\n- [#60007](https://github.com/openai/codex/pull/60007) Start cloud threads with the local GitHub project [@bc-openai](https://github.com/bc-openai)\n- [#60008](https://github.com/openai/codex/pull/60008) Refactor TUI clipboard and composer input handling [@bc-openai](https://github.com/bc-openai)\n- [#60009](https://github.com/openai/codex/pull/60009) Add custom gateway selection to TUI sign-in [@alexsong-oai](https://github.com/alexsong-oai)\n- [#60010](https://github.com/openai/codex/pull/60010) Add a shared Windows Bazel setup action [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#60011](https://github.com/openai/codex/pull/60011) Adapt multi-agent mode guidance for hosted tools [@rhan-oai](https://github.com/rhan-oai)\n- [#60014](https://github.com/openai/codex/pull/60014) Honor managed runtime authentication modes and expand coverage [@cooper-oai](https://github.com/cooper-oai)\n- [#60017](https://github.com/openai/codex/pull/60017) Disable bundled skills in the prompt-prefix comparison test [@felixxia-oai](https://github.com/felixxia-oai)\n- [#60018](https://github.com/openai/codex/pull/60018) Propagate turn triggers to usage analytics events [@chrisdong-oai](https://github.com/chrisdong-oai)\n- [#60019](https://github.com/openai/codex/pull/60019) Add theme-aware semantic colors to extension UI trees [@joesavona](https://github.com/joesavona)\n- [#60020](https://github.com/openai/codex/pull/60020) Strip terminal commands from UI tree strings [@joesavona](https://github.com/joesavona)\n- [#60021](https://github.com/openai/codex/pull/60021) Normalize inline UI tree text across styled leaves [@joesavona](https://github.com/joesavona)\n- [#60022](https://github.com/openai/codex/pull/60022) Separate UI tree preparation from line rendering [@joesavona](https://github.com/joesavona)\n- [#60026](https://github.com/openai/codex/pull/60026) Add multiline layout and row wrapping to UI trees [@joesavona](https://github.com/joesavona)\n- [#60027](https://github.com/openai/codex/pull/60027) Keep configured status lines on the native renderer [@joesavona](https://github.com/joesavona)\n- [#60028](https://github.com/openai/codex/pull/60028) Reuse prepared UI trees for footer rendering [@joesavona](https://github.com/joesavona)\n- [#60029](https://github.com/openai/codex/pull/60029) Render UI trees directly into terminal buffers [@joesavona](https://github.com/joesavona)\n- [#60031](https://github.com/openai/codex/pull/60031) Unify TUI status line layout for extension content [@joesavona](https://github.com/joesavona)\n- [#60032](https://github.com/openai/codex/pull/60032) Allow extension status content to use multiple footer rows [@joesavona](https://github.com/joesavona)\n- [#60034](https://github.com/openai/codex/pull/60034) Omit sensitive process inputs from spawn trace logs [@etraut-openai](https://github.com/etraut-openai)\n- [#60035](https://github.com/openai/codex/pull/60035) Fix authentication for git-attribution settings requests [@cooper-oai](https://github.com/cooper-oai)\n- [#60037](https://github.com/openai/codex/pull/60037) Preserve MCP sharing discovery when deferring cached stdio startup [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60038](https://github.com/openai/codex/pull/60038) Surface MCP server startup errors when preparing cached tool calls [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60039](https://github.com/openai/codex/pull/60039) Add opt-in lazy startup for cached stdio MCP servers [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60040](https://github.com/openai/codex/pull/60040) Restrict legacy incremental tools to Amazon Bedrock providers [@akira-oai](https://github.com/akira-oai)\n- [#60041](https://github.com/openai/codex/pull/60041) Require explicit project trust in non-interactive runs [@etraut-openai](https://github.com/etraut-openai)\n- [#60042](https://github.com/openai/codex/pull/60042) Centralize Windows sandbox filesystem override resolution [@viyatb-oai](https://github.com/viyatb-oai)\n- [#60046](https://github.com/openai/codex/pull/60046) Derive MCP network policy and preserve saved denials [@viyatb-oai](https://github.com/viyatb-oai)\n- [#60048](https://github.com/openai/codex/pull/60048) Allow plugin starter prompts up to 512 characters [@karenwong-openai](https://github.com/karenwong-openai)\n- [#60049](https://github.com/openai/codex/pull/60049) Preserve model catalogs when starting child agents [@evanz-oai](https://github.com/evanz-oai)\n- [#60055](https://github.com/openai/codex/pull/60055) Reauthenticate runtime requests before reusing WebSocket access [@cooper-oai](https://github.com/cooper-oai)\n- [#60061](https://github.com/openai/codex/pull/60061) Support Cloud image messages and editable draft handoffs [@bc-openai](https://github.com/bc-openai)\n- [#60064](https://github.com/openai/codex/pull/60064) Use thread runtime auth for Guardian Decisions and memory quota checks [@cooper-oai](https://github.com/cooper-oai)\n- [#60067](https://github.com/openai/codex/pull/60067) Ensure side-chat policy reaches the first model request [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60073](https://github.com/openai/codex/pull/60073) Use runtime authentication for remote-control requests [@cooper-oai](https://github.com/cooper-oai)\n- [#60075](https://github.com/openai/codex/pull/60075) Support asynchronous host service-tier reads and acceptance [@rhan-oai](https://github.com/rhan-oai)\n- [#60080](https://github.com/openai/codex/pull/60080) Allow asynchronous model request budget checks [@rhan-oai](https://github.com/rhan-oai)\n- [#60082](https://github.com/openai/codex/pull/60082) Fetch account security reminders through app-server [@cooper-oai](https://github.com/cooper-oai)\n- [#60096](https://github.com/openai/codex/pull/60096) Stop retrying terminal runtime authentication denials [@cooper-oai](https://github.com/cooper-oai)\n- [#60101](https://github.com/openai/codex/pull/60101) Use runtime authentication for Analytics requests [@cooper-oai](https://github.com/cooper-oai)\n- [#60118](https://github.com/openai/codex/pull/60118) Buffer payload JSON reads during rollout trace replay [@tz-openai](https://github.com/tz-openai)\n- [#60122](https://github.com/openai/codex/pull/60122) Preserve code mode dispatch errors in rollout traces [@tz-openai](https://github.com/tz-openai)\n- [#60152](https://github.com/openai/codex/pull/60152) Persist local workspace side chats and add tab close controls [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60157](https://github.com/openai/codex/pull/60157) Add support for incremental code mode tool definitions [@akira-oai](https://github.com/akira-oai)\n- [#60163](https://github.com/openai/codex/pull/60163) Wire incremental exec tool updates into Code Mode sessions [@akira-oai](https://github.com/akira-oai)\n- [#60172](https://github.com/openai/codex/pull/60172) Preserve display instructions in answer moderation [@psachdeva-oai](https://github.com/psachdeva-oai)\n- [#60173](https://github.com/openai/codex/pull/60173) Test exec catalog preservation across session and compaction changes [@akira-oai](https://github.com/akira-oai)\n- [#60177](https://github.com/openai/codex/pull/60177) Add metrics for incremental exec catalog updates [@akira-oai](https://github.com/akira-oai)\n- [#60180](https://github.com/openai/codex/pull/60180) Attach typed punctuation to completed skill mentions [@bc-openai](https://github.com/bc-openai)\n- [#60181](https://github.com/openai/codex/pull/60181) Add keyboard and mouse navigation for sidebar tabs [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60186](https://github.com/openai/codex/pull/60186) Allow creating custom providers during sign-in [@alexsong-oai](https://github.com/alexsong-oai)\n- [#60188](https://github.com/openai/codex/pull/60188) Allow model overrides for incremental Code Mode tool messages [@akira-oai](https://github.com/akira-oai)\n- [#60190](https://github.com/openai/codex/pull/60190) Expose agent tree IDs and environment removal [@richardopenai](https://github.com/richardopenai)\n- [#60206](https://github.com/openai/codex/pull/60206) Increase Bazel core unit test shards from 8 to 16 [@seanh-oai](https://github.com/seanh-oai)\n- [#60270](https://github.com/openai/codex/pull/60270) Warn when a CLI model is absent from the ChatGPT model list [@etraut-openai](https://github.com/etraut-openai)\n- [#60271](https://github.com/openai/codex/pull/60271) Keep the agents overview open when there is no chat to return to [@etraut-openai](https://github.com/etraut-openai)\n- [#60280](https://github.com/openai/codex/pull/60280) Fix MCP startup tracking when attaching to shared app servers [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60286](https://github.com/openai/codex/pull/60286) Search task history from the command center [@etraut-openai](https://github.com/etraut-openai)\n- [#60309](https://github.com/openai/codex/pull/60309) Defer executor readiness when prewarming a selected project [@bc-openai](https://github.com/bc-openai)\n- [#60319](https://github.com/openai/codex/pull/60319) Improve side-chat discovery and navigation in the TUI [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60320](https://github.com/openai/codex/pull/60320) Include older itemless Cloud failures in full transcripts [@bc-openai](https://github.com/bc-openai)\n- [#60327](https://github.com/openai/codex/pull/60327) Suppress stale cloud stream disconnects after reopening a thread [@bc-openai](https://github.com/bc-openai)\n- [#60335](https://github.com/openai/codex/pull/60335) Avoid duplicate assistant messages when resuming Cloud threads [@bc-openai](https://github.com/bc-openai)\n- [#60342](https://github.com/openai/codex/pull/60342) Treat goal objectives as untrusted context [@dylan-hurd-oai](https://github.com/dylan-hurd-oai)\n- [#60389](https://github.com/openai/codex/pull/60389) Include Codex Apps configuration in cloud prewarm requests [@bc-openai](https://github.com/bc-openai)\n- [#60390](https://github.com/openai/codex/pull/60390) Allow deferring plain-text MCP output in thread item listings [@btraut-openai](https://github.com/btraut-openai)\n- [#60391](https://github.com/openai/codex/pull/60391) Simplify Guardian's post-denial approval guidance [@olliem-oai](https://github.com/olliem-oai)\n- [#60414](https://github.com/openai/codex/pull/60414) Enable model catalog discovery with gateway OAuth [@smurching-oai](https://github.com/smurching-oai)\n- [#60416](https://github.com/openai/codex/pull/60416) Keep fitting URLs intact in fullscreen Markdown transcripts [@etraut-openai](https://github.com/etraut-openai)\n- [#60418](https://github.com/openai/codex/pull/60418) Let terminal status become done while background terminals run [@etraut-openai](https://github.com/etraut-openai)\n- [#60423](https://github.com/openai/codex/pull/60423) Expose deferred MCP servers in resource listings [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60432](https://github.com/openai/codex/pull/60432) Upgrade bundled SQLite to 3.54 for WAL fixes [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#60433](https://github.com/openai/codex/pull/60433) Bind API workload identity source reuse to session claims [@cooper-oai](https://github.com/cooper-oai)\n- [#60434](https://github.com/openai/codex/pull/60434) Reset app-server turn summaries when a turn starts [@cooper-oai](https://github.com/cooper-oai)\n- [#60435](https://github.com/openai/codex/pull/60435) Dismiss the model switch prompt when account usage recovers [@etraut-openai](https://github.com/etraut-openai)\n- [#60436](https://github.com/openai/codex/pull/60436) Revalidate request auth ownership after budget checks [@won-openai](https://github.com/won-openai)\n- [#60443](https://github.com/openai/codex/pull/60443) Support cancellation and deadlines for realtime attachments [@parabolala](https://github.com/parabolala)\n- [#60454](https://github.com/openai/codex/pull/60454) Update OpenSSL Rust dependencies [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#60455](https://github.com/openai/codex/pull/60455) Prefer native clipboard images on macOS and honor image orientation [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#60461](https://github.com/openai/codex/pull/60461) Honor permission grant scopes in executor capability discovery [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#60462](https://github.com/openai/codex/pull/60462) Route TUI question answers through Cloud message admission [@bc-openai](https://github.com/bc-openai)\n- [#60466](https://github.com/openai/codex/pull/60466) Keep thread storage available when diagnostic logs fail [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#60472](https://github.com/openai/codex/pull/60472) Serialize default thread permissions as a permission profile [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#60473](https://github.com/openai/codex/pull/60473) Respect error retry policy during local compaction [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60475](https://github.com/openai/codex/pull/60475) Stop retrying explicit invalid-workspace errors [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60476](https://github.com/openai/codex/pull/60476) Treat access program denials as terminal errors [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60479](https://github.com/openai/codex/pull/60479) Treat content policy violations as terminal errors [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60480](https://github.com/openai/codex/pull/60480) Treat workspace access denials as terminal errors [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60481](https://github.com/openai/codex/pull/60481) Treat invalid access program stream errors as terminal [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60482](https://github.com/openai/codex/pull/60482) Handle context length errors consistently across response transports [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60486](https://github.com/openai/codex/pull/60486) Add device authorization to app-server gateway OAuth [@smurching-oai](https://github.com/smurching-oai)\n- [#60487](https://github.com/openai/codex/pull/60487) Enable untrusted context tool outputs by default [@akira-oai](https://github.com/akira-oai)\n- [#60488](https://github.com/openai/codex/pull/60488) Remove the Xcode 26.4 MCP elicitation compatibility workaround [@jameswt-oai](https://github.com/jameswt-oai)\n- [#60489](https://github.com/openai/codex/pull/60489) Honor feature overrides when starting the app-server daemon [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#60492](https://github.com/openai/codex/pull/60492) Add local-only Claude Code skill migration [@aramaraju-oai](https://github.com/aramaraju-oai)\n- [#60496](https://github.com/openai/codex/pull/60496) Preserve typed gRPC code-mode session admission errors [@seanh-oai](https://github.com/seanh-oai)\n- [#60498](https://github.com/openai/codex/pull/60498) Unify Guardian context preparation across reviewers [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#60500](https://github.com/openai/codex/pull/60500) Derive sandbox telemetry labels directly from permission profiles [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#60505](https://github.com/openai/codex/pull/60505) Trace MCP refresh causes and plugin metadata changes [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60506](https://github.com/openai/codex/pull/60506) Preserve MCP trace context when tracing is disabled [@joesavona](https://github.com/joesavona)\n- [#60507](https://github.com/openai/codex/pull/60507) Use hermetic LLVM for Windows MSVC Bazel builds [@zbarsky-openai](https://github.com/zbarsky-openai)\n- [#60511](https://github.com/openai/codex/pull/60511) Treat inference hook denials as terminal errors [@ccui-oai](https://github.com/ccui-oai)\n- [#60513](https://github.com/openai/codex/pull/60513) Record an MCP refresh reason for session permission grants [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60517](https://github.com/openai/codex/pull/60517) Route sandbox CLI launches through the shared sandbox manager [@viyatb-oai](https://github.com/viyatb-oai)\n- [#60519](https://github.com/openai/codex/pull/60519) Handle unfinished tool calls in restored history [@etraut-openai](https://github.com/etraut-openai)\n- [#60522](https://github.com/openai/codex/pull/60522) Fragment large Responses WebSocket messages into bounded frames [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60523](https://github.com/openai/codex/pull/60523) Add policy-checked TCP upstream tunneling for SOCKS [@viyatb-oai](https://github.com/viyatb-oai)\n- [#60524](https://github.com/openai/codex/pull/60524) Encapsulate thread metadata construction [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#60525](https://github.com/openai/codex/pull/60525) Preserve first-party catalog display order in app-server model lists [@dpatel-oai](https://github.com/dpatel-oai)\n- [#60526](https://github.com/openai/codex/pull/60526) Disable Daybreak when the selected model lacks access [@jamy-OAI](https://github.com/jamy-OAI)\n- [#60528](https://github.com/openai/codex/pull/60528) Reuse installed Git objects when upgrading marketplaces [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#60542](https://github.com/openai/codex/pull/60542) Add local diagnostic ZIP saves to the TUI [@smurching-oai](https://github.com/smurching-oai)\n- [#60554](https://github.com/openai/codex/pull/60554) Add optional tool call summaries to the full-screen transcript [@etraut-openai](https://github.com/etraut-openai)\n- [#60555](https://github.com/openai/codex/pull/60555) Extend output-free transcript summaries to more tools [@etraut-openai](https://github.com/etraut-openai)\n- [#60556](https://github.com/openai/codex/pull/60556) Group consecutive tool calls in the full-screen conversation [@etraut-openai](https://github.com/etraut-openai)\n- [#60557](https://github.com/openai/codex/pull/60557) Collapse completed turn activity beneath transcript prompts [@etraut-openai](https://github.com/etraut-openai)\n- [#60558](https://github.com/openai/codex/pull/60558) Discard opening prompts when side chat creation fails [@etraut-openai](https://github.com/etraut-openai)\n- [#60559](https://github.com/openai/codex/pull/60559) Preserve inherited previews for paginated thread forks [@etraut-openai](https://github.com/etraut-openai)\n- [#60560](https://github.com/openai/codex/pull/60560) Avoid periodic wakeups in idle Windows file watchers [@chess-oai](https://github.com/chess-oai)\n- [#60568](https://github.com/openai/codex/pull/60568) Add an OpenSSH ProxyCommand mode to `codex tcp-tunnel` [@apanasenko-oai](https://github.com/apanasenko-oai)\n- [#60574](https://github.com/openai/codex/pull/60574) Allow searching persisted attached threads [@btraut-openai](https://github.com/btraut-openai)\n- [#60578](https://github.com/openai/codex/pull/60578) Add published OAuth metadata discovery to the RMCP client [@vishnu-oai](https://github.com/vishnu-oai)\n- [#60599](https://github.com/openai/codex/pull/60599) Keep the TUI responsive while opening saved side chats [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60613](https://github.com/openai/codex/pull/60613) Clarify safety recourse messaging and add a reporting option [@xiaohk](https://github.com/xiaohk)\n- [#60617](https://github.com/openai/codex/pull/60617) Support incremental tool declarations on GCP [@akira-oai](https://github.com/akira-oai)\n- [#60620](https://github.com/openai/codex/pull/60620) Allow sampling in-progress turns with `thread/sample` [@keyz](https://github.com/keyz)\n- [#60622](https://github.com/openai/codex/pull/60622) Use app-server MCP configuration and startup events in the TUI [@etraut-openai](https://github.com/etraut-openai)\n- [#60624](https://github.com/openai/codex/pull/60624) Preserve script quotes in Windows sandbox cmd invocations [@malsamiri-oai](https://github.com/malsamiri-oai)\n- [#60635](https://github.com/openai/codex/pull/60635) Support archiving and restoring cloud threads [@bc-openai](https://github.com/bc-openai)\n- [#60637](https://github.com/openai/codex/pull/60637) Show pending key chords in a shared floating shortcut guide [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60638](https://github.com/openai/codex/pull/60638) Honor read-only MCP tool hints in automatic approval [@cazhang-oai](https://github.com/cazhang-oai)\n- [#60639](https://github.com/openai/codex/pull/60639) Refresh Apps tools outside the MCP runtime gate [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#60644](https://github.com/openai/codex/pull/60644) Keep interrupts responsive during host service-tier updates [@rhan-oai](https://github.com/rhan-oai)\n- [#60645](https://github.com/openai/codex/pull/60645) Update the safety consent test for the current dialog [@ambrosino-oai](https://github.com/ambrosino-oai)\n- [#60646](https://github.com/openai/codex/pull/60646) Use trusted Guardian templates for managed policies [@won-openai](https://github.com/won-openai)\n- [#60647](https://github.com/openai/codex/pull/60647) Restore PowerShell's working directory in the MXC sandbox [@zm-oai](https://github.com/zm-oai)\n- [#60651](https://github.com/openai/codex/pull/60651) Enable model and reasoning effort selection for Cloud conversations [@bc-openai](https://github.com/bc-openai)\n- [#60655](https://github.com/openai/codex/pull/60655) Support environment updates in app-server settings APIs [@sayan-oai](https://github.com/sayan-oai)\n- [#60662](https://github.com/openai/codex/pull/60662) Support gateway device-code sign-in in headless TUI sessions [@smurching-oai](https://github.com/smurching-oai)\n- [#60663](https://github.com/openai/codex/pull/60663) Defer exec-server sandbox preparation to the execution host [@viyatb-oai](https://github.com/viyatb-oai)\n- [#60665](https://github.com/openai/codex/pull/60665) Add a content helper for forwarding MCP blocks in Code Mode [@cconger](https://github.com/cconger)\n- [#60669](https://github.com/openai/codex/pull/60669) Skip submodule worktree scans when checking repository changes [@dhavalkapil-oai](https://github.com/dhavalkapil-oai)\n- [#60670](https://github.com/openai/codex/pull/60670) Initialize Guardian evidence cursors from thread history state [@felixxia-oai](https://github.com/felixxia-oai)\n- [#60671](https://github.com/openai/codex/pull/60671) Add sign-only mode and tighten macOS signing verification [@tarunoai](https://github.com/tarunoai)\n- [#60673](https://github.com/openai/codex/pull/60673) Preserve agent message deliveries in rollout traces [@cassirer-openai](https://github.com/cassirer-openai)\n- [#60675](https://github.com/openai/codex/pull/60675) Avoid redundant reads and screen clears when switching side chats [@fcoury-oai](https://github.com/fcoury-oai)\n- [#60680](https://github.com/openai/codex/pull/60680) Keep TUI status content separate from composer hints [@joesavona](https://github.com/joesavona)\n- [#60681](https://github.com/openai/codex/pull/60681) Prioritize draft space when allocating composer footer rows [@joesavona](https://github.com/joesavona)\n- [#60682](https://github.com/openai/codex/pull/60682) Wrap built-in TUI status lines within the footer row budget [@joesavona](https://github.com/joesavona)\n- [#60683](https://github.com/openai/codex/pull/60683) Fit composer footer hints to available terminal space [@joesavona](https://github.com/joesavona)\n- [#60695](https://github.com/openai/codex/pull/60695) Add a Signer Service backend for macOS notarization [@tarunoai](https://github.com/tarunoai)\n- [#60697](https://github.com/openai/codex/pull/60697) Limit MCP product attribution to configured REPL-backed calls [@trpcic-openai](https://github.com/trpcic-openai)\n- [#60700](https://github.com/openai/codex/pull/60700) Finalize aborted turns and ignore unrelated late aborts [@etraut-openai](https://github.com/etraut-openai)\n- [#60702](https://github.com/openai/codex/pull/60702) Centralize permission config selection and legacy translation [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#60708](https://github.com/openai/codex/pull/60708) Canonicalize Windows sandbox homes and migrate saved junction paths [@zm-oai](https://github.com/zm-oai)\n- [#60721](https://github.com/openai/codex/pull/60721) Gate plugin recommendation cache warming on feature flags [@adaley-openai](https://github.com/adaley-openai)\n- [#60730](https://github.com/openai/codex/pull/60730) Apply session app overrides during MCP config refresh [@jameswt-oai](https://github.com/jameswt-oai)\n- [#60732](https://github.com/openai/codex/pull/60732) Allow model-facing ID sequences to resume from allocated IDs [@euroelessar](https://github.com/euroelessar)\n- [#60735](https://github.com/openai/codex/pull/60735) Preserve file-backed audio in tool outputs [@hung-oai](https://github.com/hung-oai)\n- [#60746](https://github.com/openai/codex/pull/60746) Add a Git credential helper for existing Codex logins [@ccy-oai](https://github.com/ccy-oai)\n- [#60748](https://github.com/openai/codex/pull/60748) Label seed arguments in the model-facing ID recovery test [@bc-openai](https://github.com/bc-openai)\n- [#60749](https://github.com/openai/codex/pull/60749) Capture saved Daybreak preferences for native helpers [@jamy-OAI](https://github.com/jamy-OAI)\n- [#60753](https://github.com/openai/codex/pull/60753) Add snapshot coverage for history and notes tool contracts [@euroelessar](https://github.com/euroelessar)\n- [#60757](https://github.com/openai/codex/pull/60757) Use typed context window IDs throughout thread persistence [@euroelessar](https://github.com/euroelessar)\n- [#60759](https://github.com/openai/codex/pull/60759) Fix Unicode caseless matching in thread search [@etraut-openai](https://github.com/etraut-openai)\n- [#60763](https://github.com/openai/codex/pull/60763) Hand off approved plans through attachments in fresh sessions [@dylan-hurd-oai](https://github.com/dylan-hurd-oai)","publishedAt":"2026-11-13T00:00:00.000Z","url":"https://developers.openai.com/codex/changelog/#github-release-443990865","media":[],"prerelease":false,"source":{"slug":"openai-codex-changelog","name":"Codex Changelog","type":"feed"},"org":{"slug":"openai","name":"OpenAI"},"product":{"slug":"codex","name":"Codex"},"groupSlug":"codex","groupName":"Codex","coverageCount":0,"composition":{"bugs":8,"features":6,"enhancements":6}},{"id":"rel_HMxN8qIoB--mMFIotPM4_","version":"0.168.0","type":"feature","title":"Codex CLI Release: 0.168.0","summary":"Codex can browse model providers under Sessions and pick a default for new chats, and an opt-in features.tui_workspace setting opens a resizable side conversation beside the main chat. Chats keep their saved model provider when the default changes, and idle Windows MCP servers no longer block filesystem I/O.","titleGenerated":"Codex CLI 0.168.0 adds model provider picker and side conversation panes","titleShort":"Model provider picker; side conversation panes ship","importance":4,"content":"## New Features\n\n- Browse model providers under Sessions → Model providers and choose the default for new chats. ([#59541](https://github.com/openai/codex/pull/59541), [#59722](https://github.com/openai/codex/pull/59722))\n- Enable `features.tui_workspace` to view a side conversation beside the main chat. Drag the divider to resize the panes or press `leader z` to zoom the focused pane. ([#58992](https://github.com/openai/codex/pull/58992), [#59527](https://github.com/openai/codex/pull/59527), [#59595](https://github.com/openai/codex/pull/59595))\n- Attach recordings when using an audio-capable model. Codex skips confirmed rejected recordings on later turns and uses clip duration to keep more short recordings in context. ([#59234](https://github.com/openai/codex/pull/59234), [#59268](https://github.com/openai/codex/pull/59268))\n- Enable `features.image_generation_animation` to generate or edit animated GIFs with optional frame count and timing controls. ([#59073](https://github.com/openai/codex/pull/59073))\n- Fullscreen Codex displays image previews and pets in tmux on Ghostty and Kitty with `allow-passthrough all`. ([#59579](https://github.com/openai/codex/pull/59579), [#59580](https://github.com/openai/codex/pull/59580))\n- Set `tui.notification_method = \"osc777\"` for terminal notifications. Approval notifications show longer command previews and distinguish terminal input from network access requests. ([#59256](https://github.com/openai/codex/pull/59256))\n\n## Bug Fixes\n\n- Local TUI history and resume pickers filter newly created threads by the signed-in ChatGPT account. ([#58918](https://github.com/openai/codex/pull/58918))\n- Open and resumed chats use their saved model provider after you change the default for new chats. ([#59503](https://github.com/openai/codex/pull/59503))\n- The TUI renders replies interrupted by tool calls as a single Markdown message and restores reasoning summaries when replaying a conversation. ([#59698](https://github.com/openai/codex/pull/59698), [#59700](https://github.com/openai/codex/pull/59700))\n- Steering messages reach running turns when sent during the final transcript save or after switching back to a long-running chat. ([#58877](https://github.com/openai/codex/pull/58877), [#58998](https://github.com/openai/codex/pull/58998))\n- Fuzzy file search gives filename matches more weight so matching parent directories are less likely to crowd them out. ([#59709](https://github.com/openai/codex/pull/59709))\n- Idle local MCP servers on Windows leave filesystem I/O available to the rest of Codex. ([#59726](https://github.com/openai/codex/pull/59726))\n\n## Documentation\n\n- The app-server reference covers listing model providers, creating custom gateways, and updating gateway, Bedrock, and GCP settings. ([#58845](https://github.com/openai/codex/pull/58845), [#58876](https://github.com/openai/codex/pull/58876))\n\n## Chores\n\n- Codex warns when `features.multi_agent` is explicitly configured and points to `agents.enabled` for disabling subagent tools. ([#59479](https://github.com/openai/codex/pull/59479))\n- Contributors can run Bazel CI jobs locally without BuildBuddy credentials. ([#59055](https://github.com/openai/codex/pull/59055))\n\n## Changelog\n\nFull Changelog: [rust-v0.167.0...rust-v0.168.0](https://github.com/openai/codex/compare/rust-v0.167.0...rust-v0.168.0)\n\n- [#58766](https://github.com/openai/codex/pull/58766) Distinguish update handoffs in TUI startup telemetry [@etraut-openai](https://github.com/etraut-openai)\n- [#58767](https://github.com/openai/codex/pull/58767) Validate `/fork` before consuming the composer draft [@etraut-openai](https://github.com/etraut-openai)\n- [#58829](https://github.com/openai/codex/pull/58829) Accept agent follow-ups through native turn input delivery [@won-openai](https://github.com/won-openai)\n- [#58845](https://github.com/openai/codex/pull/58845) Add model provider discovery to the app server [@alexsong-oai](https://github.com/alexsong-oai)\n- [#58876](https://github.com/openai/codex/pull/58876) Add model provider configuration writes to the app-server [@alexsong-oai](https://github.com/alexsong-oai)\n- [#58877](https://github.com/openai/codex/pull/58877) Handle steers accepted during the final transcript flush [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58886](https://github.com/openai/codex/pull/58886) Preserve timestamp ties when paginating threads [@jif-oai](https://github.com/jif-oai)\n- [#58890](https://github.com/openai/codex/pull/58890) Add a feature-gated account check for CLI thread resume [@jif-oai](https://github.com/jif-oai)\n- [#58894](https://github.com/openai/codex/pull/58894) Support model catalog overrides for plugin guidance [@goodman-oai](https://github.com/goodman-oai)\n- [#58899](https://github.com/openai/codex/pull/58899) Add native patch support to direct Guardian review [@swastik-oai](https://github.com/swastik-oai)\n- [#58903](https://github.com/openai/codex/pull/58903) Support permission requests in direct Guardian review [@swastik-oai](https://github.com/swastik-oai)\n- [#58904](https://github.com/openai/codex/pull/58904) Add a standalone HTTP MCP server for the agent message board [@jif-oai](https://github.com/jif-oai)\n- [#58908](https://github.com/openai/codex/pull/58908) Package the message board MCP connection as a plugin [@jif-oai](https://github.com/jif-oai)\n- [#58909](https://github.com/openai/codex/pull/58909) Add an in-memory launcher for the message-board MCP server [@jif-oai](https://github.com/jif-oai)\n- [#58918](https://github.com/openai/codex/pull/58918) Scope local history discovery to the current ChatGPT account [@jif-oai](https://github.com/jif-oai)\n- [#58927](https://github.com/openai/codex/pull/58927) Fix live stream reflow at the allocated conversation width [@fcoury-oai](https://github.com/fcoury-oai)\n- [#58931](https://github.com/openai/codex/pull/58931) Support direct Guardian review of patches invoked through exec\\_command [@swastik-oai](https://github.com/swastik-oai)\n- [#58933](https://github.com/openai/codex/pull/58933) Release message-board state before HTTP shutdown returns [@jif-oai](https://github.com/jif-oai)\n- [#58935](https://github.com/openai/codex/pull/58935) Defer Guardian context trimming in elastic preparation [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58936](https://github.com/openai/codex/pull/58936) Extract model-provider unit tests into separate files [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58937](https://github.com/openai/codex/pull/58937) Verify queued network approval wakeups before polling again [@seanh-oai](https://github.com/seanh-oai)\n- [#58938](https://github.com/openai/codex/pull/58938) Skip attribution for direct MCP tool calls [@victor-openai](https://github.com/victor-openai)\n- [#58939](https://github.com/openai/codex/pull/58939) Refine additional-context byte histogram buckets [@akira-oai](https://github.com/akira-oai)\n- [#58940](https://github.com/openai/codex/pull/58940) Retire the daemon\\_auto\\_start feature flag [@etraut-openai](https://github.com/etraut-openai)\n- [#58945](https://github.com/openai/codex/pull/58945) Fix model selection after provider setup [@etraut-openai](https://github.com/etraut-openai)\n- [#58946](https://github.com/openai/codex/pull/58946) Preserve Guardian sender context for cloud deliveries with orbit IDs [@felixxia-oai](https://github.com/felixxia-oai)\n- [#58950](https://github.com/openai/codex/pull/58950) Recover WebSocket transport after transient HTTP fallback [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58953](https://github.com/openai/codex/pull/58953) Support caller-owned sideband channels for realtime attachments [@jiahao-oai](https://github.com/jiahao-oai)\n- [#58954](https://github.com/openai/codex/pull/58954) Add a thread-store builder with host cleanup support [@felixxia-oai](https://github.com/felixxia-oai)\n- [#58955](https://github.com/openai/codex/pull/58955) Delete Guardian evidence boards when permanently deleting threads [@felixxia-oai](https://github.com/felixxia-oai)\n- [#58956](https://github.com/openai/codex/pull/58956) Allow volatile World State sections to refresh after compaction [@rhan-oai](https://github.com/rhan-oai)\n- [#58957](https://github.com/openai/codex/pull/58957) Cache marketplace name mismatches during automatic upgrades [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58959](https://github.com/openai/codex/pull/58959) Recover inherited history from alternate rollout copies [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58960](https://github.com/openai/codex/pull/58960) Prefer indexed rollout paths when resolving ancestors [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58962](https://github.com/openai/codex/pull/58962) Preserve Guardian budget policy through review recovery [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58963](https://github.com/openai/codex/pull/58963) Add host-owned turn admission to the agent runtime [@rhan-oai](https://github.com/rhan-oai)\n- [#58964](https://github.com/openai/codex/pull/58964) Stabilize and enable client tool output retention by default [@jif-oai](https://github.com/jif-oai)\n- [#58966](https://github.com/openai/codex/pull/58966) Move protocol permissions tests into a separate file [@viyatb-oai](https://github.com/viyatb-oai)\n- [#58969](https://github.com/openai/codex/pull/58969) Avoid unnecessary rendering during TUI history loading and insertion [@etraut-openai](https://github.com/etraut-openai)\n- [#58970](https://github.com/openai/codex/pull/58970) Keep automatic thread title saves off the TUI event loop [@etraut-openai](https://github.com/etraut-openai)\n- [#58971](https://github.com/openai/codex/pull/58971) Fix Escape handling in memory reset confirmation [@imac-oai](https://github.com/imac-oai)\n- [#58973](https://github.com/openai/codex/pull/58973) Add opt-in MCP discovery for in-memory agent message boards [@jif-oai](https://github.com/jif-oai)\n- [#58974](https://github.com/openai/codex/pull/58974) Bind final-message output before running sandboxed tools [@viyatb-oai](https://github.com/viyatb-oai)\n- [#58975](https://github.com/openai/codex/pull/58975) Add usage accounting for host-owned agent runtimes [@rhan-oai](https://github.com/rhan-oai)\n- [#58978](https://github.com/openai/codex/pull/58978) Enable the segment heap on Windows CI runners [@anp-oai](https://github.com/anp-oai)\n- [#58980](https://github.com/openai/codex/pull/58980) Record turn critical path metrics and trace fields [@bkotsopoulos](https://github.com/bkotsopoulos)\n- [#58983](https://github.com/openai/codex/pull/58983) Extend and bound exec-server session recovery [@richardopenai](https://github.com/richardopenai)\n- [#58984](https://github.com/openai/codex/pull/58984) Allow hosts to restrict a thread's root-only capabilities [@rhan-oai](https://github.com/rhan-oai)\n- [#58986](https://github.com/openai/codex/pull/58986) Move agent message-board backend construction into the extension [@jif-oai](https://github.com/jif-oai)\n- [#58988](https://github.com/openai/codex/pull/58988) Respect delivery boundaries for injected agent messages [@won-openai](https://github.com/won-openai)\n- [#58989](https://github.com/openai/codex/pull/58989) Keep HTTP fallback active for the rest of the session [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58990](https://github.com/openai/codex/pull/58990) Preserve modified Ctrl+C keys in active TUI views [@imac-oai](https://github.com/imac-oai)\n- [#58992](https://github.com/openai/codex/pull/58992) Add opt-in side conversation panes to the TUI [@fcoury-oai](https://github.com/fcoury-oai)\n- [#58996](https://github.com/openai/codex/pull/58996) Deliver environment hook decisions while output reads are queued [@andrewgu-oai](https://github.com/andrewgu-oai)\n- [#58998](https://github.com/openai/codex/pull/58998) Preserve steering when returning to a long-running turn [@tamird](https://github.com/tamird)\n- [#59000](https://github.com/openai/codex/pull/59000) Allow hosted child threads to inherit a root session ID [@rhan-oai](https://github.com/rhan-oai)\n- [#59001](https://github.com/openai/codex/pull/59001) Test connection events against active environment selections [@anp-oai](https://github.com/anp-oai)\n- [#59003](https://github.com/openai/codex/pull/59003) Preserve Guardian review framing under elastic context limits [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59007](https://github.com/openai/codex/pull/59007) Add host cycling and saved defaults for new TUI conversations [@bc-openai](https://github.com/bc-openai)\n- [#59009](https://github.com/openai/codex/pull/59009) Centralize filesystem permission resolution for sandbox enforcement [@viyatb-oai](https://github.com/viyatb-oai)\n- [#59011](https://github.com/openai/codex/pull/59011) Add optional total token limits to elastic section configuration [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59015](https://github.com/openai/codex/pull/59015) Make SQLite schema migrations safe for concurrent startup [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59016](https://github.com/openai/codex/pull/59016) Make observed thread metadata writes atomic and retryable [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59018](https://github.com/openai/codex/pull/59018) Add WebSocket connection metrics with reauthentication labels [@bromano-oai](https://github.com/bromano-oai)\n- [#59020](https://github.com/openai/codex/pull/59020) Preserve sandbox context in prepared MCP calls [@sayan-oai](https://github.com/sayan-oai)\n- [#59021](https://github.com/openai/codex/pull/59021) Allow concurrent readers of the installation ID [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59025](https://github.com/openai/codex/pull/59025) Remove strict third-party tool deferral in Code Mode Only [@rishab-openai](https://github.com/rishab-openai)\n- [#59028](https://github.com/openai/codex/pull/59028) Use prepared MCP owner permissions in Guardian reviews [@sayan-oai](https://github.com/sayan-oai)\n- [#59033](https://github.com/openai/codex/pull/59033) Add web search provider metadata and configuration [@yitaoh-oai](https://github.com/yitaoh-oai)\n- [#59034](https://github.com/openai/codex/pull/59034) Add model Search discovery to MCP bindings [@yitaoh-oai](https://github.com/yitaoh-oai)\n- [#59036](https://github.com/openai/codex/pull/59036) Resolve model-selected search providers with policy-aware fallback [@yitaoh-oai](https://github.com/yitaoh-oai)\n- [#59041](https://github.com/openai/codex/pull/59041) Expose assistant content token counts in completion notifications [@aojiao-oai](https://github.com/aojiao-oai)\n- [#59043](https://github.com/openai/codex/pull/59043) Tie executor attachments to environment selection lifetimes [@anp-oai](https://github.com/anp-oai)\n- [#59047](https://github.com/openai/codex/pull/59047) Decode live transcript updates in the cloud client [@bc-openai](https://github.com/bc-openai)\n- [#59055](https://github.com/openai/codex/pull/59055) Run Bazel CI locally without BuildBuddy credentials [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59057](https://github.com/openai/codex/pull/59057) Tailor MCP doctor remediation to detected problems [@etraut-openai](https://github.com/etraut-openai)\n- [#59058](https://github.com/openai/codex/pull/59058) Classify daemon auto-start failures in TUI telemetry [@etraut-openai](https://github.com/etraut-openai)\n- [#59066](https://github.com/openai/codex/pull/59066) Add typed item lifecycle events to the cloud client [@bc-openai](https://github.com/bc-openai)\n- [#59068](https://github.com/openai/codex/pull/59068) Enable Bedrock onboarding through the local daemon [@etraut-openai](https://github.com/etraut-openai)\n- [#59071](https://github.com/openai/codex/pull/59071) Enforce strict third-party tool deferral in Code Mode Only [@rishab-openai](https://github.com/rishab-openai)\n- [#59073](https://github.com/openai/codex/pull/59073) Add optional animated GIF output to image generation [@Gan-Tu](https://github.com/Gan-Tu)\n- [#59075](https://github.com/openai/codex/pull/59075) Add paginated history reads to the cloud client [@bc-openai](https://github.com/bc-openai)\n- [#59076](https://github.com/openai/codex/pull/59076) Clarify argument meanings in marketplace and guardian tests [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59082](https://github.com/openai/codex/pull/59082) Make native startup libraries available to Windows voice-host tests [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59107](https://github.com/openai/codex/pull/59107) Add opt-in reconsideration of Guardian denials [@jif-oai](https://github.com/jif-oai)\n- [#59111](https://github.com/openai/codex/pull/59111) Normalize saved chat settings after model catalog refreshes [@dpatel-oai](https://github.com/dpatel-oai)\n- [#59115](https://github.com/openai/codex/pull/59115) Preserve worker permissions across sibling-triggered reloads [@jif-oai](https://github.com/jif-oai)\n- [#59118](https://github.com/openai/codex/pull/59118) Preserve Guardian denials when reconsideration times out [@jif-oai](https://github.com/jif-oai)\n- [#59120](https://github.com/openai/codex/pull/59120) Allow dangerous-command review when sandbox approvals are disabled [@jif-oai](https://github.com/jif-oai)\n- [#59124](https://github.com/openai/codex/pull/59124) Remove a redundant permission profile clone in exec policy tests [@jif-oai](https://github.com/jif-oai)\n- [#59125](https://github.com/openai/codex/pull/59125) Add context to Guardian test setup failures [@jif-oai](https://github.com/jif-oai)\n- [#59130](https://github.com/openai/codex/pull/59130) Add a shortcut to hide the focused side conversation pane [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59131](https://github.com/openai/codex/pull/59131) Clarify test failures and annotate a backend registration invariant [@oh-openai](https://github.com/oh-openai)\n- [#59136](https://github.com/openai/codex/pull/59136) Parse Guardian review responses once [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59139](https://github.com/openai/codex/pull/59139) Route message-board notifications through agent mailboxes [@jif-oai](https://github.com/jif-oai)\n- [#59154](https://github.com/openai/codex/pull/59154) Expose thread APIs for message board access [@jif-oai](https://github.com/jif-oai)\n- [#59159](https://github.com/openai/codex/pull/59159) Move Guardian unit tests into their owning crates [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59161](https://github.com/openai/codex/pull/59161) Clarify token-count and screen-ownership arguments in tests [@jif-oai](https://github.com/jif-oai)\n- [#59162](https://github.com/openai/codex/pull/59162) Move agent message board host integration to app-server [@jif-oai](https://github.com/jif-oai)\n- [#59167](https://github.com/openai/codex/pull/59167) Move agent message board cleanup into the app server [@jif-oai](https://github.com/jif-oai)\n- [#59177](https://github.com/openai/codex/pull/59177) Wire shared runtime authentication through threads and API clients [@cooper-oai](https://github.com/cooper-oai)\n- [#59192](https://github.com/openai/codex/pull/59192) Resolve saved Daybreak preferences against advertised model programs [@jamy-OAI](https://github.com/jamy-OAI)\n- [#59201](https://github.com/openai/codex/pull/59201) Add cloud client message submission and turn interruption [@bc-openai](https://github.com/bc-openai)\n- [#59208](https://github.com/openai/codex/pull/59208) Add shortcuts to focus the main composer and side conversation [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59217](https://github.com/openai/codex/pull/59217) Use the public managed-features constructor in the guardian config test [@dylan-hurd-oai](https://github.com/dylan-hurd-oai)\n- [#59219](https://github.com/openai/codex/pull/59219) Use Bazel's zstd library for zstd-sys [@cayoub-oai](https://github.com/cayoub-oai)\n- [#59225](https://github.com/openai/codex/pull/59225) Revalidate Guardian denials when parent authorization changes [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59227](https://github.com/openai/codex/pull/59227) Preserve installer stderr when releasing the install lock [@etraut-openai](https://github.com/etraut-openai)\n- [#59228](https://github.com/openai/codex/pull/59228) Handle `budget_reached` as a terminal session budget error [@yuning-oai](https://github.com/yuning-oai)\n- [#59231](https://github.com/openai/codex/pull/59231) Fix test fixtures for Bazel sandboxes and Windows voice runtimes [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59232](https://github.com/openai/codex/pull/59232) Suppress macOS warnings in the V8 libc++ Bazel build [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59233](https://github.com/openai/codex/pull/59233) Use resolved filesystem permissions in Windows sandbox policies [@viyatb-oai](https://github.com/viyatb-oai)\n- [#59234](https://github.com/openai/codex/pull/59234) Enable uploaded audio with durable rejection recovery [@hung-oai](https://github.com/hung-oai)\n- [#59237](https://github.com/openai/codex/pull/59237) Budget retained file audio during remote compaction [@hung-oai](https://github.com/hung-oai)\n- [#59238](https://github.com/openai/codex/pull/59238) Register the Guardian additional-context feature flag [@akira-oai](https://github.com/akira-oai)\n- [#59239](https://github.com/openai/codex/pull/59239) Honor Guardian's additional-context setting for tool outputs [@akira-oai](https://github.com/akira-oai)\n- [#59240](https://github.com/openai/codex/pull/59240) Mark deny-path parent directories created by the Windows MXC launcher [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#59242](https://github.com/openai/codex/pull/59242) Set seven-day garbage collection ages for Bazel caches [@anp-oai](https://github.com/anp-oai)\n- [#59245](https://github.com/openai/codex/pull/59245) Expand per-turn critical path capture and split inference timing [@taohong339](https://github.com/taohong339)\n- [#59250](https://github.com/openai/codex/pull/59250) Expand Windows deny-read globs without invoking ripgrep [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#59251](https://github.com/openai/codex/pull/59251) Track automatic review overhead in critical-path durations [@taohong339](https://github.com/taohong339)\n- [#59253](https://github.com/openai/codex/pull/59253) Clarify TUI reconnect status for HTTP 429 stream errors [@etraut-openai](https://github.com/etraut-openai)\n- [#59254](https://github.com/openai/codex/pull/59254) Add stdin bearer authentication for exec-server registration [@apanasenko-oai](https://github.com/apanasenko-oai)\n- [#59255](https://github.com/openai/codex/pull/59255) Bind reusable backend clients to runtime authentication [@cooper-oai](https://github.com/cooper-oai)\n- [#59256](https://github.com/openai/codex/pull/59256) Add OSC 777 notifications and improve approval notification text [@etraut-openai](https://github.com/etraut-openai)\n- [#59257](https://github.com/openai/codex/pull/59257) Preserve prompt contrast when terminal colors are unavailable [@etraut-openai](https://github.com/etraut-openai)\n- [#59258](https://github.com/openai/codex/pull/59258) Refactor sandbox request preparation by backend [@viyatb-oai](https://github.com/viyatb-oai)\n- [#59260](https://github.com/openai/codex/pull/59260) Report auto-review overhead in turn critical-path analytics [@taohong339](https://github.com/taohong339)\n- [#59263](https://github.com/openai/codex/pull/59263) Normalize loading bullets in the safety recourse snapshot test [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59268](https://github.com/openai/codex/pull/59268) Use measured durations to budget uploaded audio [@hung-oai](https://github.com/hung-oai)\n- [#59272](https://github.com/openai/codex/pull/59272) Preserve requested approval reviewers across policy overrides [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59275](https://github.com/openai/codex/pull/59275) Support API workload identity authentication [@cooper-oai](https://github.com/cooper-oai)\n- [#59278](https://github.com/openai/codex/pull/59278) Enforce current filesystem permissions when reading repository skills [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59279](https://github.com/openai/codex/pull/59279) Preserve filesystem authority when loading skills [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59280](https://github.com/openai/codex/pull/59280) Respect filesystem permissions when discovering repository skills [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59282](https://github.com/openai/codex/pull/59282) Enable unified image budgeting by default [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59287](https://github.com/openai/codex/pull/59287) Add directional conversation focus shortcuts in workspace mode [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59291](https://github.com/openai/codex/pull/59291) Respect turn filesystem permissions during executor skill discovery [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59295](https://github.com/openai/codex/pull/59295) Add outcome metrics for skill reads and installation listings [@TAFOYA-OAI](https://github.com/TAFOYA-OAI)\n- [#59299](https://github.com/openai/codex/pull/59299) Switch rusty\\_v8 to the openai-oss-forks repository [@seanh-oai](https://github.com/seanh-oai)\n- [#59312](https://github.com/openai/codex/pull/59312) Route response item IDs through a shared generator [@euroelessar](https://github.com/euroelessar)\n- [#59319](https://github.com/openai/codex/pull/59319) Restore background terminals from the displayed thread's live state [@etraut-openai](https://github.com/etraut-openai)\n- [#59328](https://github.com/openai/codex/pull/59328) Introduce a dedicated context window ID type [@euroelessar](https://github.com/euroelessar)\n- [#59330](https://github.com/openai/codex/pull/59330) Stop TUI voice calls when the OS session locks [@bc-openai](https://github.com/bc-openai)\n- [#59343](https://github.com/openai/codex/pull/59343) Add cloud thread activation with resolved settings [@bc-openai](https://github.com/bc-openai)\n- [#59351](https://github.com/openai/codex/pull/59351) Add managed thread prewarming to the cloud client [@bc-openai](https://github.com/bc-openai)\n- [#59354](https://github.com/openai/codex/pull/59354) Share context window ID allocation through the agent runtime [@euroelessar](https://github.com/euroelessar)\n- [#59369](https://github.com/openai/codex/pull/59369) Stabilize cross-platform tests and directory-change fork permissions [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59401](https://github.com/openai/codex/pull/59401) Honor section limits in elastic Guardian context budgeting [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59403](https://github.com/openai/codex/pull/59403) Fix missing item ID generators in retained instruction tests [@jif-oai](https://github.com/jif-oai)\n- [#59404](https://github.com/openai/codex/pull/59404) Preserve pending agent messages across compaction [@jif-oai](https://github.com/jif-oai)\n- [#59444](https://github.com/openai/codex/pull/59444) Add opt-in elastic budgeting for synchronous Guardian reviews [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59445](https://github.com/openai/codex/pull/59445) Warn when project instructions exceed the configured byte budget [@etraut-openai](https://github.com/etraut-openai)\n- [#59447](https://github.com/openai/codex/pull/59447) Restore the daemon auto-start feature flag [@etraut-openai](https://github.com/etraut-openai)\n- [#59448](https://github.com/openai/codex/pull/59448) Prefer Launchpad for Linux zlib downloads in `rules_rs` [@jif-oai](https://github.com/jif-oai)\n- [#59455](https://github.com/openai/codex/pull/59455) Carry per-input extension data to user message completion callbacks [@wemmanuel-oai](https://github.com/wemmanuel-oai)\n- [#59458](https://github.com/openai/codex/pull/59458) Drain queued side-conversation events before switching focus [@etraut-openai](https://github.com/etraut-openai)\n- [#59468](https://github.com/openai/codex/pull/59468) Allow prepared forks to use source revalidation [@owenlin0](https://github.com/owenlin0)\n- [#59472](https://github.com/openai/codex/pull/59472) Preserve transcript interaction when switching panes with arrow keys [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59473](https://github.com/openai/codex/pull/59473) Extract agent message board hosting into a dedicated crate [@seanh-oai](https://github.com/seanh-oai)\n- [#59474](https://github.com/openai/codex/pull/59474) Refactor MCP tool argument parsing and approval preparation [@swastik-oai](https://github.com/swastik-oai)\n- [#59476](https://github.com/openai/codex/pull/59476) Honor project agent opt-ins over lower-priority legacy opt-outs [@jif-oai](https://github.com/jif-oai)\n- [#59478](https://github.com/openai/codex/pull/59478) Record saved multi-agent opt-outs in session telemetry [@jif-oai](https://github.com/jif-oai)\n- [#59479](https://github.com/openai/codex/pull/59479) Deprecate the `multi_agent` feature flag [@jif-oai](https://github.com/jif-oai)\n- [#59480](https://github.com/openai/codex/pull/59480) Add per-turn prompt cache TTL options [@rm-openai](https://github.com/rm-openai)\n- [#59481](https://github.com/openai/codex/pull/59481) Allow per-turn overrides for direct-only tool namespaces [@hlevy-openai](https://github.com/hlevy-openai)\n- [#59489](https://github.com/openai/codex/pull/59489) Use MCP owner permissions for Guardian reviews and score reuse [@sayan-oai](https://github.com/sayan-oai)\n- [#59491](https://github.com/openai/codex/pull/59491) Preserve external auth through transient provider failures [@cooper-oai](https://github.com/cooper-oai)\n- [#59493](https://github.com/openai/codex/pull/59493) Add a native cloud backend to the app-server client [@bc-openai](https://github.com/bc-openai)\n- [#59495](https://github.com/openai/codex/pull/59495) Honor the configured credential store in gateway test fixtures [@jif-oai](https://github.com/jif-oai)\n- [#59496](https://github.com/openai/codex/pull/59496) Preserve project agent opt-ins across enabled legacy feature flags [@jif-oai](https://github.com/jif-oai)\n- [#59497](https://github.com/openai/codex/pull/59497) Add debug timing for thread item pagination [@yiqian-oai](https://github.com/yiqian-oai)\n- [#59501](https://github.com/openai/codex/pull/59501) Fix the workspace policy test's auth loading call [@bc-openai](https://github.com/bc-openai)\n- [#59503](https://github.com/openai/codex/pull/59503) Keep threads on their saved model provider when defaults change [@abhinav-oai](https://github.com/abhinav-oai)\n- [#59505](https://github.com/openai/codex/pull/59505) Upgrade argument comment lint to Dylint 6.3.0 [@seanh-oai](https://github.com/seanh-oai)\n- [#59507](https://github.com/openai/codex/pull/59507) Bind memory consolidation permissions to the memory root [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59509](https://github.com/openai/codex/pull/59509) Forward caller metadata on MCP resource reads [@joesavona](https://github.com/joesavona)\n- [#59510](https://github.com/openai/codex/pull/59510) Use server model defaults for fresh TUI sessions on every connection [@etraut-openai](https://github.com/etraut-openai)\n- [#59512](https://github.com/openai/codex/pull/59512) Bump the Python SDK build backend to `uv_build` 0.13 [@ww-oai](https://github.com/ww-oai)\n- [#59514](https://github.com/openai/codex/pull/59514) Add paginated cloud thread discovery to the cloud client [@bc-openai](https://github.com/bc-openai)\n- [#59517](https://github.com/openai/codex/pull/59517) Simplify pending interactive request tracking in the TUI [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#59518](https://github.com/openai/codex/pull/59518) Validate UI trees at decode time and simplify display elements [@joesavona](https://github.com/joesavona)\n- [#59520](https://github.com/openai/codex/pull/59520) Populate the local secrets cache after successful saves [@celia-oai](https://github.com/celia-oai)\n- [#59524](https://github.com/openai/codex/pull/59524) Forward cloud message item lifecycle notifications [@bc-openai](https://github.com/bc-openai)\n- [#59527](https://github.com/openai/codex/pull/59527) Make side conversation panes resizable and interactive [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59530](https://github.com/openai/codex/pull/59530) Resolve workspace permissions at the sandbox launch boundary [@viyatb-oai](https://github.com/viyatb-oai)\n- [#59531](https://github.com/openai/codex/pull/59531) Render host-owned V2 subagent rosters in environment context [@rhan-oai](https://github.com/rhan-oai)\n- [#59535](https://github.com/openai/codex/pull/59535) Add opt-in Unicode canonical matching to exec policy rules [@dylan-hurd-oai](https://github.com/dylan-hurd-oai)\n- [#59540](https://github.com/openai/codex/pull/59540) Add a token-efficient model-facing ID generator [@anp-oai](https://github.com/anp-oai)\n- [#59541](https://github.com/openai/codex/pull/59541) Add model provider browsing to the TUI configuration panel [@alexsong-oai](https://github.com/alexsong-oai)\n- [#59544](https://github.com/openai/codex/pull/59544) Use paused time throughout the Noise relay recovery test [@dylan-hurd-oai](https://github.com/dylan-hurd-oai)\n- [#59564](https://github.com/openai/codex/pull/59564) Extend runtime auth to account, plugin, and file requests [@cooper-oai](https://github.com/cooper-oai)\n- [#59569](https://github.com/openai/codex/pull/59569) Add ChatGPT token stdin authentication to exec-server [@aagrawal-oai](https://github.com/aagrawal-oai)\n- [#59572](https://github.com/openai/codex/pull/59572) Use runtime authentication across backend request paths [@cooper-oai](https://github.com/cooper-oai)\n- [#59579](https://github.com/openai/codex/pull/59579) Enable pane-local image previews in tmux [@etraut-openai](https://github.com/etraut-openai)\n- [#59580](https://github.com/openai/codex/pull/59580) Enable pets in fullscreen tmux panes with Kitty passthrough [@etraut-openai](https://github.com/etraut-openai)\n- [#59590](https://github.com/openai/codex/pull/59590) Increase shared code-mode host capacity safely [@pash-openai](https://github.com/pash-openai)\n- [#59595](https://github.com/openai/codex/pull/59595) Add keyboard controls for TUI pane resizing and zoom [@fcoury-oai](https://github.com/fcoury-oai)\n- [#59598](https://github.com/openai/codex/pull/59598) Add a root-only option for background compaction [@oliverye-oai](https://github.com/oliverye-oai)\n- [#59601](https://github.com/openai/codex/pull/59601) Add experimental command output deferral to thread item listings [@btraut-openai](https://github.com/btraut-openai)\n- [#59632](https://github.com/openai/codex/pull/59632) Remove the legacy Seatbelt network policy wrapper [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59633](https://github.com/openai/codex/pull/59633) Expose retained command sessions in incremental world state [@jif-oai](https://github.com/jif-oai)\n- [#59638](https://github.com/openai/codex/pull/59638) Ignore the unmaintained pin-utils advisory in dependency audits [@jif-oai](https://github.com/jif-oai)\n- [#59642](https://github.com/openai/codex/pull/59642) Preserve turn configuration in Guardian scoring and approvals [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59659](https://github.com/openai/codex/pull/59659) Resample Guardian denials with the original request [@jif-oai](https://github.com/jif-oai)\n- [#59669](https://github.com/openai/codex/pull/59669) Add original-history evidence snapshots for Guardian V3 [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59670](https://github.com/openai/codex/pull/59670) Capture Guardian evidence incrementally with explicit acknowledgements [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59671](https://github.com/openai/codex/pull/59671) Add evidence housekeeping for Guardian v3 [@felixxia-oai](https://github.com/felixxia-oai)\n- [#59677](https://github.com/openai/codex/pull/59677) Remove the legacy sandbox policy summary helper [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#59688](https://github.com/openai/codex/pull/59688) Extract channel notification formatting into `PostPreview` [@jif-oai](https://github.com/jif-oai)\n- [#59691](https://github.com/openai/codex/pull/59691) Split MCP direct-mode exposure coverage into separate test cases [@jif-oai](https://github.com/jif-oai)\n- [#59693](https://github.com/openai/codex/pull/59693) Support Cloud conversation requests in the app-server client [@bc-openai](https://github.com/bc-openai)\n- [#59698](https://github.com/openai/codex/pull/59698) Fix reasoning summary restoration in the TUI [@etraut-openai](https://github.com/etraut-openai)\n- [#59700](https://github.com/openai/codex/pull/59700) Fix TUI assistant messages interrupted by tool execution [@etraut-openai](https://github.com/etraut-openai)\n- [#59709](https://github.com/openai/codex/pull/59709) Prioritize filename matches in fuzzy file search [@omers-oai](https://github.com/omers-oai)\n- [#59711](https://github.com/openai/codex/pull/59711) Expand Linux sandbox deny globs in process [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#59713](https://github.com/openai/codex/pull/59713) Clarify message board notification delivery documentation [@jif-oai](https://github.com/jif-oai)\n- [#59714](https://github.com/openai/codex/pull/59714) Set `output_deferred` to false for cloud thread items [@jif-oai](https://github.com/jif-oai)\n- [#59722](https://github.com/openai/codex/pull/59722) Allow choosing the default provider for new chats in the TUI [@alexsong-oai](https://github.com/alexsong-oai)\n- [#59725](https://github.com/openai/codex/pull/59725) Add host reporting for terminal agent turn outcomes [@rhan-oai](https://github.com/rhan-oai)\n- [#59726](https://github.com/openai/codex/pull/59726) Prevent idle Windows MCP pipes from blocking unrelated I/O [@majapw-oai](https://github.com/majapw-oai)\n- [#59728](https://github.com/openai/codex/pull/59728) Use Codex releases as the primary libcap source for musl builds [@anp-oai](https://github.com/anp-oai)\n- [#59732](https://github.com/openai/codex/pull/59732) Omit native multi-agent guidance when native tools are disabled [@won-openai](https://github.com/won-openai)\n- [#59735](https://github.com/openai/codex/pull/59735) Support direct Guardian review of recorded MCP calls [@swastik-oai](https://github.com/swastik-oai)\n- [#59736](https://github.com/openai/codex/pull/59736) Stabilize the unreadable-glob traversal error test @maxwell-oai\\[bot\\]\n- [#59737](https://github.com/openai/codex/pull/59737) Restrict Guardian native multi-agent history to the OpenAI provider [@jif-oai](https://github.com/jif-oai)","publishedAt":"2026-11-13T00:00:00.000Z","url":"https://developers.openai.com/codex/changelog/#github-release-443793906","media":[],"prerelease":false,"source":{"slug":"openai-codex-changelog","name":"Codex Changelog","type":"feed"},"org":{"slug":"openai","name":"OpenAI"},"product":{"slug":"codex","name":"Codex"},"groupSlug":"codex","groupName":"Codex","coverageCount":0,"composition":{"bugs":8,"features":10,"enhancements":6}},{"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","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","url":"https://code.claude.com/docs/en/changelog#2-1-323","media":[],"prerelease":false,"source":{"slug":"claude-code-releases","name":"Claude Code","type":"scrape"},"org":{"slug":"anthropic","name":"Anthropic"},"product":{"slug":"claude-code","name":"Claude Code"},"groupSlug":"claude-code","groupName":"Claude Code","coverageCount":0,"composition":{"bugs":150,"features":16,"enhancements":40}},{"id":"rel_pS5FkMCnrm81jhAm6MKMq","version":null,"type":"feature","title":"November 12, 2026","summary":"The Devin Review page is redesigned for all organizations with merge controls and actions moved to the top bar, and whether Devin fixes Devin Review findings on pull requests is now a separate setting from Responding to bots. Security gains separate Source filtering per connection, per-finding chats, and undoable status changes, plus closed-beta finding imports from Veracode and Black Duck.","titleGenerated":"Devin Review gets redesigned page and its own auto-fix setting","titleShort":"Devin Review redesigned; findings auto-fix gets own setting","importance":3,"content":"### Sessions & Workspace\n\n**Same-kind requests merge into one card**\n\nWhen Devin asks for several things of the same kind at once (credentials, approvals, network access), they appear together in one compact card instead of a stack of cards.\n\n**\"Reminder set\" row**\n\nWhen Devin sets itself a reminder, the session shows a \"Reminder set\" row with the time or schedule and a link to the reminder.\n\n**Repo file links show as pills**\n\nBare links to a file on github.com or gitlab.com appear in chat as a short pill naming the file, lines and commit instead of the full URL.\n\n**Drag an open session by its header**\n\nDrag an open session by its pane header onto a sidebar folder, Recent or another session, with the same result as dragging its sidebar row.\n\n**Clearer upload and folder-attach errors**\n\nAttaching a folder no longer fails when one file cannot be read (the rest is attached and you are told what was left out), and uploads blocked by your network's security policy say so with the reason.\n\n### Devin Review\n\n**Redesigned Devin Review page**\n\nThe redesigned Devin Review page is now available to all organizations: merge controls, the actions menu and the diff-size pill live in the top bar, the sidebar collapses out of the way, and findings are presented in the redesigned cards and panels.\n\n**\"Automatically fix Devin Review findings\" setting**\n\nWhether Devin fixes Devin Review findings on pull requests from its sessions is now its own setting, separate from Responding to bots. The account default is in Settings > Review, and each person can change it for their own sessions in Settings > Preferences, under Devin sessions.\n\n### Security\n\n**Scan syncs with several connections**\n\nWhen starting a Security scan you can pick several connections so its findings sync to more than one external destination.\n\n**Source filter lists each connection**\n\nThe Source filter on the Findings page lists each connection separately, so findings can be filtered to a single scanner connection.\n\n**Named Custom connections**\n\nA Custom security connection now asks for a connection name (for example \"Semgrep\"), which is the name it is shown under.\n\n**Delete a finding tag org-wide**\n\nHover a tag in the Findings filter and click the X to delete it from every finding at once.\n\n**Each finding has its own chat**\n\nThe chat button on a finding opens that finding's own chat; the Findings page launcher keeps its general chat.\n\n**Finding status changes can be undone**\n\nDismissing a finding or marking it reviewed shows exactly what changed, with an Undo button (or Cmd/Ctrl+Z).\n\n**Veracode and Black Duck findings sources (closed beta)**\n\nSecurity connections can import findings from Veracode (Static, Dynamic, Manual and SCA findings) and Black Duck (vulnerable components of a project version). Dismissing or reopening a finding writes back as a Veracode mitigation or Black Duck remediation status. Available as a closed beta.\n\n### Enterprise & Settings\n\n**Seven more optional GitHub permissions**\n\nEnterprise admins can stop Devin from using the GitHub code quality, deployments, pages, discussions, repository projects, repository hooks and packages permissions under Optional GitHub permissions.\n\n**Approve personal secrets shared with service users**\n\nAnyone who can write to a session can approve a request to use a personal secret they shared with a service user.\n\n**Enterprise-disabled agent modes are shown**\n\nOrganization settings show agent modes disabled by your enterprise and explain who can enable them.\n\n**Login page explains IP allowlist denials**\n\nUsers signing in from outside their organization's IP allowlist see why on the login page, including their current IP, instead of a blank login page.\n\n**Multiple installations per MCP in Customize**\n\nMCPs signed in with several accounts list each installation separately with its own settings; add another installation from the marketplace card; every installation offers Disconnect and Uninstall with a short explanation. Supabase can switch between read-only and full access from its settings.\n\n**Pylon EU region**\n\nAdmins can connect a Pylon workspace hosted in Pylon's EU region, and Pylon issue links open the EU app for those workspaces.\n\n**Bitbucket Data Center clone fixes**\n\nClones through Devin's git proxy work with Bitbucket Data Center `/scm/` clone URLs and on instances with Basic authentication disabled (the default in version 10).\n\n**MCP tool permissions**\n\nSet each MCP tool to Allow, Needs approval or Block from Customize. Enterprise admins can also limit individual tools to specific organizations.\n\n### Integrations\n\n**Mention-only Slack thread mode**\n\nNew Slack thread mode \"Reply in the same thread, only when mentioned\": Devin stays in the thread but only responds to messages that tag it. Available as the org default and as a personal override.\n\n**Archiving a session archives its Slack code channel**\n\nArchiving a Devin session also archives its dedicated Slack code channel.\n\n**Teams: Devin edits messages in place**\n\nIn Microsoft Teams, when Devin edits a message it is updated in place, and a replaced message collapses behind \"Show earlier\".\n\n**Slack working status for @mentions**\n\nIn Slack threads with several people, Devin shows its working status right away when you @mention it, and after it starts acting on an untagged reply.\n\n### Automations & Oncall\n\n**Readable schedules on automation cards**\n\nAutomation cards show a readable schedule (\"Weekly on Thursday at 6:00 PM...\") for schedules that previously showed raw RRULE text.\n\n**Oncall report notifications and PR panel**\n\nOncall report notifications in Slack open with the report period and list action items as bullets; the report's pull requests panel shows each PR's state, repository and date (GitLab orgs see \"Merge requests\").","publishedAt":"2026-11-12T19:00:05.000Z","url":"https://docs.devin.ai/release-notes/overview#november-12-2026","media":[],"prerelease":false,"source":{"slug":"release-notes","name":"Release Notes","type":"feed"},"org":{"slug":"cognition","name":"Cognition"},"product":{"slug":"devin","name":"Devin"},"groupSlug":"devin","groupName":"Devin","coverageCount":0,"composition":{"bugs":2,"features":14,"enhancements":16}},{"id":"rel_lYX1dQXO2iq684c6VNJtH","version":null,"type":"feature","title":"November 11, 2026","summary":"Design Mode now opens a compact prompt bar with per-property editing when you select an element, and supports spotlight boxes and pencil annotations in five colors. Devin can set itself reminders to wake up in the future, including in sessions started by automations or service users.","titleGenerated":"Devin adds Design Mode with per-property editing and spotlight annotations","titleShort":"Design Mode adds per-property editing; Devin sets itself reminders","importance":4,"content":"**Customizable Sidebar Navigation Shortcuts**\n\nThe keys for Option-hold sidebar session navigation can be changed in Settings › Keyboard shortcuts.\n\n**Filter Sessions by Service User**\n\nThe sessions list's Created by filter includes service users, and sessions created by a service user show its name in Properties.\n\n**Improved Visibility of Stacked PRs**\n\nImproved visibility of stacked PRs within the UI, with an option for stack order preference (shared with Devin Review).\n\n**Design Mode**\n\nSelecting an element in Design Mode opens a compact prompt bar with per-property editing. Draw spotlight boxes or pencil annotations in five colors and send the annotated page image with a note; pick colors from the preview with the eyedropper; and search and use your page's CSS variables in property controls.\n\n**Grouped Repository Setup Suggestions**\n\nYou can approve adjacent repository setup suggestions together, or review each diff and approve it separately.\n\n**Mark the Open Session Unread**\n\nMark unread is available in the sidebar menu for the session you currently have open.\n\n**Bulk Add Secrets Accepts Shell-style Lines**\n\nBulk add secrets accepts `export KEY=value` lines and keeps a `#` that is part of a value (put a space before `#` to add a note).\n\n**Devin Can Set Itself Reminders**\n\nDevin can set itself reminders to wake up in the future, including in sessions started by automations or service users.\n\n**Link Findings and Threats**\n\nCreate a threat directly from a finding's Threat menu, remove a finding from the threat it is linked to, and find saved threats by title or number when filtering findings or starting a scan.\n\n**Start Scan Page with Threat Focus**\n\nStart scan opens as its own page (links can preselect a repository or threat) and includes a Focus on threats section with a searchable table.\n\n**Severity Letters**\n\nFindings use C/H/M/L letter badges instead of severity icons, and Repositories show medium and low counts separately.\n\n**Dismissal Reasons**\n\nRecord a reason or note when dismissing a finding from its row, from the finding's PR view, or across a selection; the reason is shown in the finding detail sheet.\n\n**Auto-scan Schedules**\n\nNew auto-scans show their schedule before launch and keep their local time when daylight saving time changes.\n\n**Failed Scans Explain Themselves**\n\nWhen a scan fails before finding anything, the scan page, the session's Findings tab and the per-scan history say why and offer Recover scan or Start new scan.\n\n**Findings and Repositories, Smaller Changes**\n\nOpen a repository's scan history from its row; scan summaries show medium and low findings; dragged column widths are remembered; drag a CSV report into a Security connection; remediation sessions link to their assigned finding.\n\n**Ask Before Loading External Content**\n\nSession security profiles gain an Ask before loading external content option. In sessions it governs, images from other sites and HTML/PDF attachment previews load only after you click them.\n\n**Guided Perforce Setup**\n\nPerforce setup guides you through server details and depot selection, reusing known details from a saved connection or importing the server address from your local Perforce configuration.\n\n**Edit Enterprise Plugin MCPs in Place**\n\nEnterprise admins can edit an active plugin MCP's Access, OAuth client and secrets in place.\n\n**Improved In-app UI for Grouped Oncall Alerts**\n\nOncall alerts with many occurrences can load older ones, and the events view shows whether each event was investigated, forwarded, grouped or skipped, with a link to the session.","publishedAt":"2026-11-12T03:22:13.000Z","url":"https://docs.devin.ai/release-notes/overview#november-11-2026","media":[],"prerelease":false,"source":{"slug":"release-notes","name":"Release Notes","type":"feed"},"org":{"slug":"cognition","name":"Cognition"},"product":{"slug":"devin","name":"Devin"},"groupSlug":"devin","groupName":"Devin","coverageCount":0,"composition":{"bugs":4,"features":16,"enhancements":4}},{"id":"rel_3V4yuy2hoChfnl7xH7O8W","version":"0.167.2","type":"feature","title":"Codex CLI Release: 0.167.2","summary":"Codex now sends tool updates incrementally by default on Google Cloud Vertex AI; setting incremental_tools_legacy_gcp = false under [features] in config.toml opts out.","titleGenerated":"Codex CLI 0.167.2 sends tool updates incrementally on Vertex AI by default","titleShort":"Vertex AI tool updates now incremental by default","importance":2,"content":"## New Features\n\n- Codex now sends tool updates incrementally by default on Google Cloud Vertex AI. To opt out, set `incremental_tools_legacy_gcp = false` under `[features]` in `config.toml`. ([#60617](https://github.com/openai/codex/pull/60617), [#60848](https://github.com/openai/codex/pull/60848))\n\n## Changelog\n\nFull Changelog: [rust-v0.167.1...rust-v0.167.2](https://github.com/openai/codex/compare/rust-v0.167.1...rust-v0.167.2)","publishedAt":"2026-11-11T00:00:00.000Z","url":"https://developers.openai.com/codex/changelog/#github-release-442267349","media":[],"prerelease":false,"source":{"slug":"openai-codex-changelog","name":"Codex Changelog","type":"feed"},"org":{"slug":"openai","name":"OpenAI"},"product":{"slug":"codex","name":"Codex"},"groupSlug":"codex","groupName":"Codex","coverageCount":0,"composition":{"bugs":0,"features":1,"enhancements":0}},{"id":"rel_E5YAJrRofwxZeF1etLIQ7","version":null,"type":"feature","title":"November 9, 2026","summary":"A new GET /v3/enterprise/consumption/daily/sessions endpoint returns per-day session consumption for an enterprise. Devin now refreshes expiring GitHub user tokens automatically for GitHub Enterprise Server and token-issuing GitHub apps, and every member including Teams users can open My Analytics to see their own usage and limits.","titleGenerated":"Devin adds daily session consumption API and auto-refreshing GitHub user tokens","titleShort":"GitHub user tokens refresh automatically; daily session usage API ships","importance":3,"content":"**\"Redacted Secret\" Pills Link to the Secret**\n\nWhen a transcript shows a redacted secret, the pill links to the matching entry in Settings › Secrets.\n\n**Forwarded Messages Show the Original Sender**\n\nMessages forwarded into a session show who originally sent them, with a \"Forwarded by / via\" label.\n\n**Stacked PRs Show \"In Merge Queue\"**\n\nPRs in a stack that are waiting in the merge queue are labeled as such in Devin Review.\n\n**One Source Filter for Findings**\n\nA single **Source** filter (Connections / Scans) replaces the old Scan type filter on Security findings pages.\n\n**Scan Findings Tab: Real Columns and Easier Navigation**\n\nThe Findings tab on a scan uses proper columns with horizontal scrolling, breadcrumbs lead back to the finding you came from, and you can open scan runs directly after a split scan.\n\n**Unlink a Finding from Its Remediation Session**\n\nFindings can be unlinked from a remediation session via the API and MCP tools, and batch remediation now paces sessions to avoid overloading repositories.\n\n**My Analytics for Every Member**\n\nAll members, including Teams users, can open **My Analytics** to see their own usage and limits and request more usage.\n\n**GitHub User Tokens Refresh Automatically**\n\nFor GitHub Enterprise Server and GitHub apps that issue expiring user tokens, Devin refreshes the token automatically; relink once and you won't be asked again.\n\n**Enterprise Environment: Active Build Queue**\n\nEnterprise admins can see the queue of environment builds currently running or waiting.\n\n**Daily Session Consumption API**\n\nNew endpoint `GET /v3/enterprise/consumption/daily/sessions` returns per-day session consumption for an enterprise.\n\n**Incident Reports**\n\nOncall incident reports open in a dedicated view with a timeline, a UTC/local time switch and copy buttons; summaries are shorter.\n\n**Custom Plugin Logos**\n\nPlugins that declare a `logo` in their manifest show it in the plugin list and overview.\n\n**Deny Button on Slack Permission Cards**\n\nPermission cards in Slack now have a Deny button alongside Approve.\n\n**PR Cards Update When a PR Is Reopened**\n\nSlack and Teams PR cards turn green again when a closed PR is reopened.\n\n**Devin's Slack Posts Keep Their Session Link**\n\nPosts containing tables or headings keep the session link, and messages that fail to format are retried as plain text.\n\n**Teams: Suggestions After Linking**\n\nAfter linking your Teams account, Devin suggests recent work to pick up.","publishedAt":"2026-11-10T21:15:58.000Z","url":"https://docs.devin.ai/release-notes/overview#november-9-2026","media":[],"prerelease":false,"source":{"slug":"release-notes","name":"Release Notes","type":"feed"},"org":{"slug":"cognition","name":"Cognition"},"product":{"slug":"devin","name":"Devin"},"groupSlug":"devin","groupName":"Devin","coverageCount":0,"composition":{"bugs":0,"features":17,"enhancements":2}},{"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","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","url":"https://code.claude.com/docs/en/changelog#2-1-322","media":[],"prerelease":false,"source":{"slug":"claude-code-releases","name":"Claude Code","type":"scrape"},"org":{"slug":"anthropic","name":"Anthropic"},"product":{"slug":"claude-code","name":"Claude Code"},"groupSlug":"claude-code","groupName":"Claude Code","coverageCount":0,"composition":{"bugs":0,"features":1,"enhancements":0}},{"id":"rel_Y_Eki_1F2UPJhZcVcy2XC","version":null,"type":"feature","title":"ChatGPT for iOS","summary":"Paste images and files into the full-screen composer, and open file attachments from sent prompts. Switch between drives in the folder picker on Windows hosts. Dictate annotation notes. Hide the task list on iPhone Duo without leaving the current task.","titleGenerated":null,"titleShort":null,"importance":null,"content":"# ChatGPT for iOS\n\n### New features\n\n- Paste images and files into the full-screen composer, and open file attachments from sent prompts.\n- Switch between drives in the folder picker on Windows hosts.\n- Dictate annotation notes.\n- Hide the task list on iPhone Duo without leaving the current task.\n\n### Improvements and bug fixes\n\n- Reopening a running Codex Cloud task keeps loaded messages visible.\n- Sent prompts stay above their replies, even when the host confirms receipt late.\n- Expanded command cards show the full command and more of its output.\n- Improved keyboard and toolbar behavior on iPad and iPhone Duo.\n- Fixed opening changed files and loading response images on Windows hosts. The folder picker now opens on Linux hosts that can't locate a home folder.\n- Manual pairing works with VoiceOver and on devices that can't scan QR codes.","publishedAt":"2026-11-10T00:00:00.000Z","url":"https://developers.openai.com/codex/changelog/#codex-2026-11-10-mobile","media":[],"prerelease":false,"source":{"slug":"openai-codex-changelog","name":"Codex Changelog","type":"feed"},"org":{"slug":"openai","name":"OpenAI"},"product":{"slug":"codex","name":"Codex"},"groupSlug":"codex","groupName":"Codex","coverageCount":0,"composition":null},{"id":"rel_lsmRg9u9pQ933MVCs2bDq","version":"0.167.1","type":"feature","title":"Codex CLI Release: 0.167.1","summary":"Full Changelog: rust-v0.167.0...rust-v0.167.1","titleGenerated":null,"titleShort":null,"importance":null,"content":"## Changelog\n\nFull Changelog: [rust-v0.167.0...rust-v0.167.1](https://github.com/openai/codex/compare/rust-v0.167.0...rust-v0.167.1)","publishedAt":"2026-11-10T00:00:00.000Z","url":"https://developers.openai.com/codex/changelog/#github-release-440934118","media":[],"prerelease":false,"source":{"slug":"openai-codex-changelog","name":"Codex Changelog","type":"feed"},"org":{"slug":"openai","name":"OpenAI"},"product":{"slug":"codex","name":"Codex"},"groupSlug":"codex","groupName":"Codex","coverageCount":0,"composition":null},{"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","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","url":"https://code.claude.com/docs/en/changelog#2-1-321","media":[],"prerelease":false,"source":{"slug":"claude-code-releases","name":"Claude Code","type":"scrape"},"org":{"slug":"anthropic","name":"Anthropic"},"product":{"slug":"claude-code","name":"Claude Code"},"groupSlug":"claude-code","groupName":"Claude Code","coverageCount":0,"composition":{"bugs":70,"features":2,"enhancements":30}},{"id":"rel_nooq0KygKKnTx1eR2wjE0","version":null,"type":"feature","title":"November 6, 2026","summary":"Admins can set default session visibility across the enterprise or per organization and limit sharing to the session's organization. Settings now has a Log Out of All Devices option covering the web app, Desktop, CLI, and mobile apps. Gerrit connections can sync group membership and project access into repo access, requiring Gerrit 3.13+ for new connections.","titleGenerated":"Devin adds session sharing controls, sign-out everywhere, and Gerrit permission sync","titleShort":"Session sharing gets org caps; log out of all devices","importance":3,"content":"**\"Reveal Child Sessions\" Switch in the Sidebar**\n\nThe sidebar's Properties menu has a \"Reveal child sessions\" switch; turn it off to keep child sessions hidden under their parent until you expand it.\n\n**Side Chats: Queue or Interrupt, and the List Persists**\n\nYou can now send messages while a side chat is working: send to interrupt it or queue to run after it finishes, as in the main chat. Older side chats stay in a session's side chat list after a reload or reconnect.\n\n**Setup Steps on Secret Request Cards**\n\nWhen Devin needs an API key from you, its request card now shows step-by-step instructions for getting it.\n\n**Redesigned Share Artifact Dialog**\n\nThe Share artifact dialog has a new layout: set the link name and publish, then see the link name and URL with an inline copy button, sharing options and a \"General access\" row showing who in your organization can view the artifact.\n\n**Ask Screenshots Carry into the Agent Session**\n\nScreenshots attached in Ask now carry over automatically when you create an Agent session from the thread.\n\n**Customizations Tabs in Settings Search**\n\nSettings search now finds each Customizations tab (Plugins, Skills, MCPs, Hooks, Rules, Memory) and opens that tab.\n\n**Gerrit Changes in \"PRs Needing You\"**\n\nGerrit changes you own or review now appear in Devin Review's \"PRs needing you\", matched to your Gerrit account by verified email.\n\n**Repository Shown on Findings and in Scan Details**\n\nThe finding detail sheet names the finding's repository with a link to it, and the Scan details popover lists every repository in a multi-repo scan.\n\n**Sort the Security Repositories Table**\n\nSort the Security Repositories table by repository name, number of open findings, or last scan time from the column headers; the sort is kept in the page URL. Long repository names show in full on hover.\n\n**Session Sharing: Account Defaults and an Organization Cap**\n\nAdmins can set the default visibility for new sessions to \"Account, read-only\" or \"Account, read and write\", for the enterprise or per organization, and can limit session sharing to the session's organization.\n\n**Pylon Connection: Allowed Organizations**\n\nEnterprise admins can limit which organizations can use the Pylon connection.\n\n**Gerrit Permission Sync**\n\nGerrit connections can now sync Gerrit group membership and project access into Devin repository access, like GitHub and GitLab member sync. New Gerrit connections use a long-lived service account password that Devin exchanges for short-lived tokens (requires Gerrit 3.13+).\n\n**Jira Admin Connection Finishes in the App**\n\nConnecting Jira as an admin now finishes inside the Devin app for the person who started it, and any problem during the connection is shown on the Jira settings page in your language.\n\n**Machine Catalog and Machines Tab**\n\nThe enterprise Machine catalog settings show OS icons per platform and tier names with slot counts, and the Machines tab shows one row per platform with a dropdown to pick the machine.\n\n**Enterprise Environments Page Loads for Large Enterprises**\n\nThe Environments page for large enterprises no longer times out; the organization list sorts by name only.\n\n**Log Out of All Devices**\n\nLog out of the web app, Devin Desktop, the CLI and the mobile apps on every device at once from Settings → Connections.\n\n**Enterprise Multi-organization Incidents**\n\nEnterprise admins get an Oncall tab in Enterprise Settings to enable incidents across multiple organizations and route PagerDuty incidents to the right organization; incidents turn on there automatically.\n\n**PagerDuty Incidents Start from the PagerDuty Slack Channel**\n\nPagerDuty incidents now start from the Slack channel PagerDuty creates, with no time limit on how late the channel appears. The PagerDuty incident source now requires a Slack channel name prefix matching the channel names set in PagerDuty. New PagerDuty responder triggers start with an empty filter.\n\n**Service User Slack Apps: Watch Channels, DMs, Icons, Enterprise Setup**\n\nAutomations and oncall responders that run as a service user with its own Slack app can watch channels that app is in (including private channels you're in), reply in DMs when the responder allows them, and post answers when installed org-wide on Enterprise Grid. You can upload an icon for the app, and enterprise admins can set up Slack apps from the Enterprise service users page.\n\n**Manage Oncall Role Can Edit Responders**\n\nMembers whose role grants Manage Oncall (without Manage Automations) can now fully load and save the Oncall responder editor and incident settings.\n\n**Devin MCP: Run Automations and Oncall as a Service User**\n\nDevin can now create automations and Oncall responders that run as a specific service user. Terraform devin\\_automation and devin\\_oncall\\_responder resources with run\\_as = \"organization\" no longer show drift on every plan.\n\n**Plugins Show an Installing State**\n\nPlugins show \"Installing\" on the Marketplace card, in Customize and in the / menu until indexing finishes, and refresh their resources automatically when done.\n\n**Plugin MCPs Honor oauthResource**\n\nPlugin MCP declarations now honor oauthResource, so a web plugin can omit the OAuth resource parameter for authorization servers such as Microsoft Entra that reject it.\n\n**GitLab Fine-grained Personal Access Tokens**\n\nViewing merge requests and Devin Review now work with GitLab fine-grained personal access tokens, and repo permissions set by GitLab group list their repositories with these tokens. Line counts and reactions are not shown until GitLab adds GraphQL support.\n\n**Slack-synced Sessions Start with the Session Title**\n\nSessions synced to Slack from the composer now start with a top-level message showing the session title and link (updated on rename), with your prompt and all replies in its thread. Sessions linked to Slack later use the same format.\n\n**Teams Messages Retried on Connection Failures**\n\nDevin's messages in Microsoft Teams are less likely to be lost during a brief connection failure.\n\n**V3 API: Memory\\_enabled on Session Create**\n\nThe v3 create session API accepts memory\\_enabled: false to keep memory off for that session and its child sessions. The grants API no longer lets anyone change or remove a secret owner's own access to that secret.","publishedAt":"2026-11-08T17:56:37.000Z","url":"https://docs.devin.ai/release-notes/overview#november-6-2026","media":[],"prerelease":false,"source":{"slug":"release-notes","name":"Release Notes","type":"feed"},"org":{"slug":"cognition","name":"Cognition"},"product":{"slug":"devin","name":"Devin"},"groupSlug":"devin","groupName":"Devin","coverageCount":0,"composition":{"bugs":1,"features":20,"enhancements":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","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","url":"https://code.claude.com/docs/en/changelog#2-1-319","media":[],"prerelease":false,"source":{"slug":"claude-code-releases","name":"Claude Code","type":"scrape"},"org":{"slug":"anthropic","name":"Anthropic"},"product":{"slug":"claude-code","name":"Claude Code"},"groupSlug":"claude-code","groupName":"Claude Code","coverageCount":0,"composition":{"bugs":60,"features":11,"enhancements":20}},{"id":"rel_KRTpUnDHC_oFwQX5qyxA_","version":null,"type":"feature","title":"Voice Chat with Agents","summary":"Agents can now be reached in a realtime voice call where you describe a change out loud, ask what an agent is working on, and hear progress updates while it runs. Voice chat is available in the Agents Window, the IDE, on web, and Cursor for iOS, runs on Grok Voice with selectable voices, and keeps zero data retention under Privacy Mode.","titleGenerated":"Cursor adds realtime voice chat with coding agents across IDE, web, and iOS","titleShort":"Voice chat with agents lands in IDE, web, iOS","importance":4,"content":"You can now talk with your agents in a realtime voice call. Describe a change out loud, ask what an agent is working on, and hear updates as the work moves forward.\n\nVoice chat works in the Agents Window, the IDE, on web, and in Cursor for iOS.\n\n## Start a call\n\nWhen the agent is idle and the chat input is empty, click the waveform button labeled \\*\\*Start voice chat\\*\\*. It sits where Send usually is, next to the microphone button for dictation.\n\nOn iOS, you can also say \"Hey Siri, call Cursor\" to start a call with a new agent, and mute or hang up from the Lock Screen with a Live Activity.\n\n## Talk while the agent works\n\nThe voice agent listens, replies, and hands coding work to the agent in your chat. While that agent runs, you can keep the conversation going. Ask what changed in a diff, check on a run, search earlier messages and calls with this agent, or switch between workers in a Project that has them.\n\nThe coding agent speaks for itself too. It shares short progress notes while it works and a brief answer when it's done, so you usually hear a short update instead of the full written reply.\n\n## Powered by Grok Voice\n\nVoice chat runs on Grok Voice. Pick from many voices in Voice settings, including Ara, Eve, Leo, Rex, and Sal. With Privacy Mode on, voice chat runs with zero data retention.\n\n## Voice chat and dictation\n\nVoice chat is separate from dictation, which earlier releases called Voice Mode. Dictation still turns your speech into text in the chat input, and Cmd+Shift+Space still toggles it in the IDE. Voice chat is a two-way conversation where the agent answers out loud.\n\nUpdate Cursor to try voice chat. Learn more in our [docs](https://cursor.com/docs/agent/voice-chat).","publishedAt":"2026-11-06T00:00:00.000Z","url":"https://cursor.com/changelog/voice-chat","media":[],"prerelease":false,"source":{"slug":"cursor-changelog","name":"Cursor Changelog","type":"feed"},"org":{"slug":"cursor","name":"Cursor"},"product":null,"groupSlug":"cursor-changelog","groupName":"Cursor Changelog","coverageCount":0,"composition":{"bugs":0,"features":1,"enhancements":0}},{"id":"rel_sAKqckb6H-y44wHzHs5E1","version":"0.167.0","type":"feature","title":"Codex CLI Release: 0.167.0","summary":"Codex CLI gains an agent command center via codex agents --no-daemon that marks unread results, and the terminal session picker can now switch between connected backends while keeping drafts. Saved chats stay visible after switching model providers, network approval prompts close when their requests end, and Ctrl+L clears the display while keeping history available through Ctrl+T.","titleGenerated":"Codex CLI 0.167.0 adds agent command center and cross-backend session switching","titleShort":"Agent command center lands; sessions switch across backends","importance":3,"content":"## New Features\n\n- Open the agent command center with `codex agents --no-daemon`. The command center and session picker now mark unread results ([#58238](https://github.com/openai/codex/pull/58238), [#58239](https://github.com/openai/codex/pull/58239), [#58263](https://github.com/openai/codex/pull/58263), [#58264](https://github.com/openai/codex/pull/58264)).\n- Switch between connected backends from the terminal session picker and keep your drafts when switching ([#58280](https://github.com/openai/codex/pull/58280), [#58457](https://github.com/openai/codex/pull/58457)).\n- Preview local image attachments while composing on supported terminals, and open viewed local images from clickable transcript paths ([#58261](https://github.com/openai/codex/pull/58261), [#58446](https://github.com/openai/codex/pull/58446)).\n- Find API key and cloud provider options under “Sign in another way.” Enable `gcp_setup_wizard` to set up Google Cloud Vertex AI with Application Default Credentials or a credential file ([#57857](https://github.com/openai/codex/pull/57857), [#58427](https://github.com/openai/codex/pull/58427)).\n- Opt into macOS or Windows system proxy settings from `/experimental` ([#58157](https://github.com/openai/codex/pull/58157)).\n- Expand patch details directly in the transcript while an approval prompt is open ([#58281](https://github.com/openai/codex/pull/58281)).\n\n## Bug Fixes\n\n- Saved chats stay visible after you switch model providers. If a chat can’t resume, you can read its transcript and prepare an editable prompt to continue in a new chat ([#58202](https://github.com/openai/codex/pull/58202)).\n- `Ctrl+L` clears the terminal display while keeping earlier messages available through `Ctrl+T` ([#58266](https://github.com/openai/codex/pull/58266)).\n- Deleting text that exposes a leading `!` keeps an ordinary draft in chat mode ([#58412](https://github.com/openai/codex/pull/58412)).\n- Network approval prompts close when their requests end. Concurrent requests to the same endpoint prompt one at a time ([#58350](https://github.com/openai/codex/pull/58350), [#58351](https://github.com/openai/codex/pull/58351)).\n- Long realtime V3 conversations save completed speech to chat history as the transcript grows ([#58221](https://github.com/openai/codex/pull/58221)).\n- Automatic context compaction now accounts for retained output from failed responses and responses with missing token usage data ([#58404](https://github.com/openai/codex/pull/58404)).\n\n## Documentation\n\n- The app-server README covers managed network constraints under `[permissions.network]` and `prompt-user` command rules that require fresh human approval ([#58045](https://github.com/openai/codex/pull/58045), [#58182](https://github.com/openai/codex/pull/58182)).\n- The `openai-docs` skill recommends the documented replacement when migrating away from a deprecated model ([#58134](https://github.com/openai/codex/pull/58134)).\n\n## Chores\n\n- DotSlash manifests now reference the Cargo packages shipped with the release ([#58717](https://github.com/openai/codex/pull/58717), [#58757](https://github.com/openai/codex/pull/58757)).\n- Linux musl builds use fewer allocator arenas to reduce idle memory after conversations unload ([#58399](https://github.com/openai/codex/pull/58399)).\n\n## Changelog\n\nFull Changelog: [rust-v0.166.0...rust-v0.167.0](https://github.com/openai/codex/compare/rust-v0.166.0...rust-v0.167.0)\n\n- [#57848](https://github.com/openai/codex/pull/57848) Add bounded Code Mode output collection and rendering [@seanh-oai](https://github.com/seanh-oai)\n- [#57850](https://github.com/openai/codex/pull/57850) Wire bounded output collection into the code mode runtime [@seanh-oai](https://github.com/seanh-oai)\n- [#57855](https://github.com/openai/codex/pull/57855) Load TUI plugin mentions from installed plugins [@zamoshchin-openai](https://github.com/zamoshchin-openai)\n- [#57857](https://github.com/openai/codex/pull/57857) Add Google Cloud Vertex AI setup to TUI onboarding [@alexsong-oai](https://github.com/alexsong-oai)\n- [#57859](https://github.com/openai/codex/pull/57859) Extend the provider OAuth sign-in timeout to ten minutes [@smurching-oai](https://github.com/smurching-oai)\n- [#57861](https://github.com/openai/codex/pull/57861) Skip resource template discovery for hosted Apps [@khong-oai](https://github.com/khong-oai)\n- [#57862](https://github.com/openai/codex/pull/57862) Keep resume picker pagination moving across backends [@bc-openai](https://github.com/bc-openai)\n- [#57864](https://github.com/openai/codex/pull/57864) Allow upstream interruption when canceling Responses Lite turns [@liame-oai](https://github.com/liame-oai)\n- [#57866](https://github.com/openai/codex/pull/57866) Retry transient environment startup failures on the next turn [@pansul47-oai](https://github.com/pansul47-oai)\n- [#57870](https://github.com/openai/codex/pull/57870) Enable server-side output truncation for Code Mode sessions [@seanh-oai](https://github.com/seanh-oai)\n- [#57876](https://github.com/openai/codex/pull/57876) Reuse the cached runtime keymap for default configuration [@bc-openai](https://github.com/bc-openai)\n- [#57883](https://github.com/openai/codex/pull/57883) Preserve status-line hyperlinks through footer layout and animations [@joesavona](https://github.com/joesavona)\n- [#57970](https://github.com/openai/codex/pull/57970) Add an experimental UI tree renderer for the status line [@joesavona](https://github.com/joesavona)\n- [#57974](https://github.com/openai/codex/pull/57974) Apply session plugin overrides during MCP config refresh [@TAFOYA-OAI](https://github.com/TAFOYA-OAI)\n- [#57978](https://github.com/openai/codex/pull/57978) Add per-chat leases for shared MCP connections [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#57985](https://github.com/openai/codex/pull/57985) Skip read-only volume-root expansion in the MXC sandbox [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#58013](https://github.com/openai/codex/pull/58013) Fix argument comment lint tests after rust-clippy branch rename [@seanh-oai](https://github.com/seanh-oai)\n- [#58025](https://github.com/openai/codex/pull/58025) Add trace-safe diagnostics for Noise relay request progress [@apanasenko-oai](https://github.com/apanasenko-oai)\n- [#58028](https://github.com/openai/codex/pull/58028) Add direct Guardian review of recorded exec calls [@swastik-oai](https://github.com/swastik-oai)\n- [#58034](https://github.com/openai/codex/pull/58034) Add an option to disable successful SSE delta logs [@jif-oai](https://github.com/jif-oai)\n- [#58040](https://github.com/openai/codex/pull/58040) Preserve native tool-call evidence through Guardian review [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58045](https://github.com/openai/codex/pull/58045) Add managed command checkpoints requiring fresh human approval [@jif-oai](https://github.com/jif-oai)\n- [#58057](https://github.com/openai/codex/pull/58057) Initialize `native_call` in agent host test fixtures [@jif-oai](https://github.com/jif-oai)\n- [#58071](https://github.com/openai/codex/pull/58071) Share the Guardian context delivery token limit [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58075](https://github.com/openai/codex/pull/58075) Use serving assessments for adaptive Guardian review [@jif-oai](https://github.com/jif-oai)\n- [#58083](https://github.com/openai/codex/pull/58083) Use a dedicated binary for Bazel schema generation [@jif-oai](https://github.com/jif-oai)\n- [#58091](https://github.com/openai/codex/pull/58091) Update default models for memory extraction and consolidation [@jif-oai](https://github.com/jif-oai)\n- [#58110](https://github.com/openai/codex/pull/58110) Ignore RUSTSEC-2026-0418 for the Windows-only libz-sys dependency [@swastik-oai](https://github.com/swastik-oai)\n- [#58119](https://github.com/openai/codex/pull/58119) Add experimental Guardian review for injected thread commands [@swastik-oai](https://github.com/swastik-oai)\n- [#58121](https://github.com/openai/codex/pull/58121) Centralize provider request preparation auth checks [@sgarg-oai](https://github.com/sgarg-oai)\n- [#58122](https://github.com/openai/codex/pull/58122) Add an incremental exec tools feature flag [@akira-oai](https://github.com/akira-oai)\n- [#58123](https://github.com/openai/codex/pull/58123) Add validated elastic section budget configuration for Guardian [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58125](https://github.com/openai/codex/pull/58125) Support negotiated server-side truncation in Code Mode clients [@seanh-oai](https://github.com/seanh-oai)\n- [#58127](https://github.com/openai/codex/pull/58127) Honor the tooltip preference for TUI startup help [@imac-oai](https://github.com/imac-oai)\n- [#58129](https://github.com/openai/codex/pull/58129) Persist subagent relationships before creating child storage [@radai-oai](https://github.com/radai-oai)\n- [#58130](https://github.com/openai/codex/pull/58130) Support pasted search text in the skills toggle view [@xma-oai](https://github.com/xma-oai)\n- [#58134](https://github.com/openai/codex/pull/58134) Prefer documented replacements for deprecation-driven model migrations [@kkahadze-oai](https://github.com/kkahadze-oai)\n- [#58140](https://github.com/openai/codex/pull/58140) Route standalone web search through the thread's workspace context [@sgarg-oai](https://github.com/sgarg-oai)\n- [#58144](https://github.com/openai/codex/pull/58144) Record additional context counts, sizes, and truncation metrics [@akira-oai](https://github.com/akira-oai)\n- [#58147](https://github.com/openai/codex/pull/58147) Return standalone web search failures to the model [@sgarg-oai](https://github.com/sgarg-oai)\n- [#58157](https://github.com/openai/codex/pull/58157) Expose system proxy support as an experimental setting [@abhinav-oai](https://github.com/abhinav-oai)\n- [#58159](https://github.com/openai/codex/pull/58159) Add a feature flag for code-mode host output truncation [@seanh-oai](https://github.com/seanh-oai)\n- [#58161](https://github.com/openai/codex/pull/58161) Include tool names and annotation hints in MCP approval metadata [@brisebois-oai](https://github.com/brisebois-oai)\n- [#58165](https://github.com/openai/codex/pull/58165) Keep remote plugin bundle URLs out of diagnostics [@mzeng-openai](https://github.com/mzeng-openai)\n- [#58166](https://github.com/openai/codex/pull/58166) Make rollout recording failures and completion explicit [@cassirer-openai](https://github.com/cassirer-openai)\n- [#58168](https://github.com/openai/codex/pull/58168) Create missing parent directories for MXC deny paths on Windows [@iceweasel-oai](https://github.com/iceweasel-oai)\n- [#58171](https://github.com/openai/codex/pull/58171) Stabilize cancellation and tracing tests [@aakash-oai](https://github.com/aakash-oai)\n- [#58172](https://github.com/openai/codex/pull/58172) Validate trace completion records and payload UTF-8 during replay [@cassirer-openai](https://github.com/cassirer-openai)\n- [#58182](https://github.com/openai/codex/pull/58182) Support managed network requirements under `permissions.network` [@viyatb-oai](https://github.com/viyatb-oai)\n- [#58189](https://github.com/openai/codex/pull/58189) Use turn-scoped permissions for standalone web search [@sayan-oai](https://github.com/sayan-oai)\n- [#58190](https://github.com/openai/codex/pull/58190) Allow providers to opt into Cyber access programs [@jamy-OAI](https://github.com/jamy-OAI)\n- [#58195](https://github.com/openai/codex/pull/58195) Preserve remote read compatibility with older executors [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#58196](https://github.com/openai/codex/pull/58196) Require Bazel release artifacts and publish DotSlash manifests [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#58198](https://github.com/openai/codex/pull/58198) Centralize TOML config deserialization [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#58202](https://github.com/openai/codex/pull/58202) Keep chats accessible across model provider changes [@abhinav-oai](https://github.com/abhinav-oai)\n- [#58219](https://github.com/openai/codex/pull/58219) Add file-backed storage for gateway OAuth credentials [@alexsong-oai](https://github.com/alexsong-oai)\n- [#58221](https://github.com/openai/codex/pull/58221) Preserve realtime V3 transcripts in conversation history [@guinness-oai](https://github.com/guinness-oai)\n- [#58237](https://github.com/openai/codex/pull/58237) Preserve unrelated side conversations when archiving or deleting threads [@etraut-openai](https://github.com/etraut-openai)\n- [#58238](https://github.com/openai/codex/pull/58238) Enable the agent command center for embedded sessions [@etraut-openai](https://github.com/etraut-openai)\n- [#58239](https://github.com/openai/codex/pull/58239) Enable the agent command center in embedded mode [@etraut-openai](https://github.com/etraut-openai)\n- [#58240](https://github.com/openai/codex/pull/58240) Enable TUI task tools for embedded app-server sessions [@etraut-openai](https://github.com/etraut-openai)\n- [#58241](https://github.com/openai/codex/pull/58241) Update `mio` from 1.2.0 to 1.2.4 [@euroelessar](https://github.com/euroelessar)\n- [#58251](https://github.com/openai/codex/pull/58251) Reload remote executor credentials safely across token rotation [@tgrue-openai](https://github.com/tgrue-openai)\n- [#58260](https://github.com/openai/codex/pull/58260) Reuse `LocalImageAttachment` throughout the TUI composer [@etraut-openai](https://github.com/etraut-openai)\n- [#58261](https://github.com/openai/codex/pull/58261) Add local image thumbnails to the TUI composer [@etraut-openai](https://github.com/etraut-openai)\n- [#58263](https://github.com/openai/codex/pull/58263) Show unread results in the TUI command center [@etraut-openai](https://github.com/etraut-openai)\n- [#58264](https://github.com/openai/codex/pull/58264) Show unread sessions and acknowledge viewed transcripts in the picker [@etraut-openai](https://github.com/etraut-openai)\n- [#58266](https://github.com/openai/codex/pull/58266) Preserve transcript history when clearing the terminal with Ctrl+L [@xma-oai](https://github.com/xma-oai)\n- [#58270](https://github.com/openai/codex/pull/58270) Recognize Rex terminals and enable Kitty graphics for pets [@etraut-openai](https://github.com/etraut-openai)\n- [#58280](https://github.com/openai/codex/pull/58280) Reuse connected backends in the in-app resume picker [@bc-openai](https://github.com/bc-openai)\n- [#58281](https://github.com/openai/codex/pull/58281) Allow inline patch review during approval prompts [@etraut-openai](https://github.com/etraut-openai)\n- [#58324](https://github.com/openai/codex/pull/58324) Treat auth reload failures as transient during token refresh [@jif-oai](https://github.com/jif-oai)\n- [#58341](https://github.com/openai/codex/pull/58341) Add an MCP adapter for message-board tools [@jif-oai](https://github.com/jif-oai)\n- [#58343](https://github.com/openai/codex/pull/58343) Preserve fallback token refreshes when the keyring is unreadable [@jif-oai](https://github.com/jif-oai)\n- [#58350](https://github.com/openai/codex/pull/58350) Resolve network approval prompts when their requests end [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58351](https://github.com/openai/codex/pull/58351) Serialize manual network approval prompts per endpoint [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58352](https://github.com/openai/codex/pull/58352) Allow Apps catalog discovery before the initial tool inventory completes [@bkotsopoulos](https://github.com/bkotsopoulos)\n- [#58355](https://github.com/openai/codex/pull/58355) Fix remote executor authentication timeout recovery [@jif-oai](https://github.com/jif-oai)\n- [#58357](https://github.com/openai/codex/pull/58357) Extract pre-tool-use hook handling into a helper [@swastik-oai](https://github.com/swastik-oai)\n- [#58361](https://github.com/openai/codex/pull/58361) Preserve model metadata origin through runtime resolution [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58368](https://github.com/openai/codex/pull/58368) Apply PreToolUse hooks before direct Guardian review policy checks [@swastik-oai](https://github.com/swastik-oai)\n- [#58378](https://github.com/openai/codex/pull/58378) Scope agent MCP metadata to a host-selected plugin registration [@rhan-oai](https://github.com/rhan-oai)\n- [#58380](https://github.com/openai/codex/pull/58380) Broadcast thread section changes to connected clients [@charliemarsh-oai](https://github.com/charliemarsh-oai)\n- [#58381](https://github.com/openai/codex/pull/58381) Fix unsettled refresh detection after workspace switching [@jif-oai](https://github.com/jif-oai)\n- [#58385](https://github.com/openai/codex/pull/58385) Test V8 inline typed array storage limits [@fcoury-oai](https://github.com/fcoury-oai)\n- [#58388](https://github.com/openai/codex/pull/58388) Add whole-entry elastic allocation for Guardian context [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58390](https://github.com/openai/codex/pull/58390) Expose parent context for host-managed child startup [@rhan-oai](https://github.com/rhan-oai)\n- [#58395](https://github.com/openai/codex/pull/58395) Add scoped host metadata hooks for MCP tool requests [@rhan-oai](https://github.com/rhan-oai)\n- [#58399](https://github.com/openai/codex/pull/58399) Reduce jemalloc arena count for Linux musl builds [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#58403](https://github.com/openai/codex/pull/58403) Run package updates outside the project on non-Windows platforms [@viyatb-oai](https://github.com/viyatb-oai)\n- [#58404](https://github.com/openai/codex/pull/58404) Fix context token estimates for responses without usable usage [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#58405](https://github.com/openai/codex/pull/58405) Add an option to suppress incremental model change announcements [@chetan-oai](https://github.com/chetan-oai)\n- [#58406](https://github.com/openai/codex/pull/58406) Add host-inherited thread instructions and execution policies [@rhan-oai](https://github.com/rhan-oai)\n- [#58410](https://github.com/openai/codex/pull/58410) Enable deferred tool world state by default [@adaley-openai](https://github.com/adaley-openai)\n- [#58411](https://github.com/openai/codex/pull/58411) Expose non-consuming input activity waits to hosts [@rhan-oai](https://github.com/rhan-oai)\n- [#58412](https://github.com/openai/codex/pull/58412) Keep exposed leading bangs literal in the chat composer [@xma-oai](https://github.com/xma-oai)\n- [#58418](https://github.com/openai/codex/pull/58418) Simplify recorded exec command review [@swastik-oai](https://github.com/swastik-oai)\n- [#58427](https://github.com/openai/codex/pull/58427) Group alternative sign-in methods in a submenu [@alexsong-oai](https://github.com/alexsong-oai)\n- [#58428](https://github.com/openai/codex/pull/58428) Support app settings in plugin configuration [@mzeng-openai](https://github.com/mzeng-openai)\n- [#58430](https://github.com/openai/codex/pull/58430) Add audio file references with explicit runtime rejection [@hung-oai](https://github.com/hung-oai)\n- [#58434](https://github.com/openai/codex/pull/58434) Treat audio validation failures as non-retryable input errors [@hung-oai](https://github.com/hung-oai)\n- [#58436](https://github.com/openai/codex/pull/58436) Add interactive controls and refresh inline visualization styles [@philipp-oai](https://github.com/philipp-oai)\n- [#58441](https://github.com/openai/codex/pull/58441) Preserve Apps usage instructions in model catalog messages [@goodman-oai](https://github.com/goodman-oai)\n- [#58444](https://github.com/openai/codex/pull/58444) Cap Guardian context entries before elastic allocation [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58446](https://github.com/openai/codex/pull/58446) Make viewed image paths clickable in the TUI [@etraut-openai](https://github.com/etraut-openai)\n- [#58454](https://github.com/openai/codex/pull/58454) Add elastic budgeting for Guardian context preparation [@ioanbudea-OAI](https://github.com/ioanbudea-OAI)\n- [#58456](https://github.com/openai/codex/pull/58456) Preserve incomplete response output in failed inference traces [@cassirer-openai](https://github.com/cassirer-openai)\n- [#58457](https://github.com/openai/codex/pull/58457) Enable cross-backend session switching in the TUI [@bc-openai](https://github.com/bc-openai)\n- [#58459](https://github.com/openai/codex/pull/58459) Tighten Bazel CI snapshot handling and OTLP test cleanup [@imac-oai](https://github.com/imac-oai)\n- [#58713](https://github.com/openai/codex/pull/58713) Extract fresh session adoption from TUI session startup [@bc-openai](https://github.com/bc-openai)\n- [#58714](https://github.com/openai/codex/pull/58714) Sort agent center tasks by recency within status groups [@etraut-openai](https://github.com/etraut-openai)\n- [#58716](https://github.com/openai/codex/pull/58716) Support OSC 52 clipboard forwarding through GNU screen [@etraut-openai](https://github.com/etraut-openai)\n- [#58717](https://github.com/openai/codex/pull/58717) Make Bazel checks advisory and pause Bazel release artifacts [@dkovalenko-oai](https://github.com/dkovalenko-oai)\n- [#58718](https://github.com/openai/codex/pull/58718) Support rendering TUI conversations within an inset area [@fcoury-oai](https://github.com/fcoury-oai)\n- [#58719](https://github.com/openai/codex/pull/58719) Wire native voice dependencies into full Rust CI [@bc-openai](https://github.com/bc-openai)\n- [#58720](https://github.com/openai/codex/pull/58720) Prioritize pending approvals in the subagent picker [@tamird](https://github.com/tamird)\n- [#58722](https://github.com/openai/codex/pull/58722) Add naming and navigation helpers for TUI backend sessions [@bc-openai](https://github.com/bc-openai)\n- [#58727](https://github.com/openai/codex/pull/58727) Use local Bazel for voice release builds [@fcoury-oai](https://github.com/fcoury-oai)\n- [#58751](https://github.com/openai/codex/pull/58751) Release superseded assistant text from live paginated history [@btraut-openai](https://github.com/btraut-openai)\n- [#58757](https://github.com/openai/codex/pull/58757) Remove Bazel outputs from DotSlash release configs [@fcoury-oai](https://github.com/fcoury-oai)","publishedAt":"2026-11-06T00:00:00.000Z","url":"https://developers.openai.com/codex/changelog/#github-release-435979892","media":[],"prerelease":false,"source":{"slug":"openai-codex-changelog","name":"Codex Changelog","type":"feed"},"org":{"slug":"openai","name":"OpenAI"},"product":{"slug":"codex","name":"Codex"},"groupSlug":"codex","groupName":"Codex","coverageCount":0,"composition":{"bugs":6,"features":7,"enhancements":1}},{"id":"rel_GAUPJ-p-drVgzgzRwXRT_","version":"v1.0.77","type":"feature","title":"Grok Build 1.0.77","summary":"Setting kitty_keyboard = \"query\" now lets Shift+Enter insert newlines in unidentified terminals, and MCP server connections no longer restart unnecessarily on session reloads or updates. The footer's context usage no longer drops to zero or shows stale values after a rewind, and quote copy now preserves ampersands in links.","titleGenerated":"Grok Build 1.0.77 adds kitty keyboard query mode and fixes MCP restarts","titleShort":"MCP connections no longer restart; footer context usage fixed","importance":2,"content":"- `[ui] kitty_keyboard = \"query\"` lets Shift+Enter insert newlines in unidentified terminals\n- Quote copy now preserves `&` in links, and double-clicks no longer fold thoughts when missing text\n- MCP server connections no longer restart unnecessarily on session reloads or updates\n- `Ctrl+L` now works on more screens, and copy hints match what actually copies to the clipboard\n- Context usage shown in the footer no longer drops to zero or shows stale values after a rewind\n- Rows in the `/resume` list now expand and collapse the same way they do inside a session","publishedAt":"2026-11-05T00:00:00.000Z","url":"https://x.ai/build/changelog#v1-0-77","media":[],"prerelease":false,"source":{"slug":"grok-build-releases","name":"Grok Build","type":"scrape"},"org":{"slug":"xai","name":"xAI"},"product":{"slug":"grok-build","name":"Grok Build"},"groupSlug":"grok-build","groupName":"Grok Build","coverageCount":0,"composition":{"bugs":4,"features":2,"enhancements":0}},{"id":"rel_Y24yyNC_kryzrH41z4zeS","version":null,"type":"feature","title":"Build Live Apps in Cursor","summary":"Cloud agents can scaffold a web app, provision Postgres, set up auth behind Cursor SSO, and deploy it to a live link you pick a domain for. Publishing runs deterministic and agentic security checks for exposed secrets or credentials, and nothing goes live until they pass.","titleGenerated":"Cursor adds live app building and hosting with shareable links","titleShort":"Cloud agents now build, deploy, and share live apps","importance":4,"content":"Agents have lowered the cost of writing code, but most apps and tools that users build never make it off of localhost.\n\nNow, you can build live apps directly in Cursor and share them with a link. Describe what you want to create and Cursor will handle the rest: build the app, set up a database, configure auth, and deploy it.\n\n[Video](https://ptht05hbb1ssoooe.public.blob.vercel-storage.com/Cursor_PublishApps_15.mp4)\n\nWith this release, anyone in your org can now build useful tools for their work and share them with teammates. Update Cursor to try it out.\n\n## How it works\n\nTo get started, kick off a [cloud agent](https://cursor.com/docs/cloud-agent) and tell it explicitly that you want to build an app.\n\nCursor will scaffold the web app, spin up a Postgres database if needed, set up authentication behind Cursor SSO, and provision hosting. If the app needs to connect to your systems, like a data warehouse, Cursor will ask for the secrets it needs.\n\nWhen you go to publish your app, you pick a unique domain and Cursor gets you back a live link.\n\nAt Cursor, users across sales, operations, legal, finance, and engineering are building internal apps. Some of the most useful ones are:\n\n- Internal ops tools for event registration, account assignments, and desk seating\n- Multimedia presentations for customers that combine slides, videos, and interactive demos\n- Training, enablement, and onboarding experiences for new hires\n- Live dashboards and tools for sales teams to understand their pipeline and opportunities\n- Portals to manage engineering services\n\nCursor can deploy apps from new or existing repos across Cursor Origin and GitHub. Support for other git providers like GitLab is coming soon.\n\n## Controls and guardrails\n\nWhen you publish an app, Cursor will run deterministic and agentic security checks to catch common issues, like secrets written in app files or exposed credentials. If a check fails, the agent tells you what it needs to fix. Nothing goes live until the checks pass and you explicitly publish the app.\n\nTeam admins can see every app and secret on the team. They can also add their own rules to the security review, approve publishes the review blocked, and unpublish any app.\n\n## Share with your team\n\nOnce the checks pass, Cursor deploys the app to the URL you chose. You can then share your app with specific people, groups, or your entire team, all behind Cursor sign-in. On individual plans, you can also share apps publicly.\n\nYou can also provision a Cursor API key if you want a script or Grok Bot to use your app.\n\n## Get started\n\nThis capability is now available on all paid plans with cloud agents enabled. Learn more by reading our [docs](https://cursor.com/docs/cloud-agent/apps).","publishedAt":"2026-11-05T00:00:00.000Z","url":"https://cursor.com/changelog/build-live-apps","media":[{"type":"video","url":"https://ptht05hbb1ssoooe.public.blob.vercel-storage.com/Cursor_PublishApps_15.mp4"}],"prerelease":false,"source":{"slug":"cursor-changelog","name":"Cursor Changelog","type":"feed"},"org":{"slug":"cursor","name":"Cursor"},"product":null,"groupSlug":"cursor-changelog","groupName":"Cursor Changelog","coverageCount":0,"composition":{"bugs":0,"features":1,"enhancements":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","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","url":"https://code.claude.com/docs/en/changelog#2-1-318","media":[],"prerelease":false,"source":{"slug":"claude-code-releases","name":"Claude Code","type":"scrape"},"org":{"slug":"anthropic","name":"Anthropic"},"product":{"slug":"claude-code","name":"Claude Code"},"groupSlug":"claude-code","groupName":"Claude Code","coverageCount":0,"composition":{"bugs":40,"features":7,"enhancements":15}}],"pagination":{"nextCursor":"2026-11-04T00:01:03.000Z|2026-11-04T02:54:30.956Z|rel_2RHL6lTHakEm6erqNrDGL","limit":20}}