{ "archive_path": "archive/1789769086.100515", "base_url": "code.claude.com/docs/en/changelog", "basename": "changelog", "bookmarked_date": "2026-09-18 22:04", "canonical": { "archive_org_path": "https://web.archive.org/web/code.claude.com/docs/en/changelog", "dom_path": "output.html", "favicon_path": "favicon.ico", "git_path": "git/", "google_favicon_path": "https://www.google.com/s2/favicons?domain=code.claude.com", "headers_path": "headers.json", "htmltotext_path": "htmltotext.txt", "index_path": "index.html", "media_path": "media/", "mercury_path": "mercury/content.html", "pdf_path": "output.pdf", "readability_path": "readability/content.html", "screenshot_path": "screenshot.png", "singlefile_path": "singlefile.html", "warc_path": "warc/", "wget_path": "code.claude.com/docs/en/changelog.html" }, "domain": "code.claude.com", "extension": "", "hash": "1ESWR8NVWY9D09RQPG16", "history": { "archive_org": [], "dom": [], "favicon": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--max-time", "120", "--output", "favicon.ico", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://www.google.com/s2/favicons?domain=code.claude.com" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-18T22:07:03.082696+00:00", "index_texts": null, "output": "favicon.ico", "pwd": "/data/archive/1789769086.100515", "schema": "ArchiveResult", "start_ts": "2026-09-18T22:07:02.610336+00:00", "status": "succeeded" }, { "cmd": [ "curl", "--silent", "--location", "--compressed", "--max-time", "120", "--output", "favicon.ico", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://www.google.com/s2/favicons?domain=code.claude.com" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-19T01:43:43.367961+00:00", "index_texts": null, "output": "favicon.ico", "pwd": "/data/archive/1789769086.100515", "schema": "ArchiveResult", "start_ts": "2026-09-19T01:43:42.733754+00:00", "status": "succeeded" } ], "git": [], "headers": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--head", "--max-time", "120", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://code.claude.com/docs/en/changelog" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-18T22:07:03.450391+00:00", "index_texts": null, "output": "headers.json", "pwd": "/data/archive/1789769086.100515", "schema": "ArchiveResult", "start_ts": "2026-09-18T22:07:03.088695+00:00", "status": "succeeded" }, { "cmd": [ "curl", "--silent", "--location", "--compressed", "--head", "--max-time", "120", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://code.claude.com/docs/en/changelog" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-19T01:43:43.796151+00:00", "index_texts": null, "output": "headers.json", "pwd": "/data/archive/1789769086.100515", "schema": "ArchiveResult", "start_ts": "2026-09-19T01:43:43.377389+00:00", "status": "succeeded" } ], "htmltotext": [], "media": [], "mercury": [ { "cmd": [ "/app/node_modules/@postlight/parser/cli.js", "https://code.claude.com/docs/en/changelog" ], "cmd_version": "1.0.0", "end_ts": "2026-09-18T22:07:46.470315+00:00", "index_texts": null, "output": "mercury", "pwd": "/data/archive/1789769086.100515", "schema": "ArchiveResult", "start_ts": "2026-09-18T22:07:34.954008+00:00", "status": "succeeded" }, { "cmd": [ "/app/node_modules/@postlight/parser/cli.js", "https://code.claude.com/docs/en/changelog" ], "cmd_version": "1.0.0", "end_ts": "2026-09-19T01:44:26.002599+00:00", "index_texts": null, "output": "mercury", "pwd": "/data/archive/1789769086.100515", "schema": "ArchiveResult", "start_ts": "2026-09-19T01:44:15.490887+00:00", "status": "succeeded" } ], "pdf": [], "readability": [ { "cmd": [ "/app/node_modules/readability-extractor/readability-extractor", "/tmp/tmp1lxgsajt", "https://code.claude.com/docs/en/changelog" ], "cmd_version": "0.0.11", "end_ts": "2026-09-18T22:07:34.944503+00:00", "index_texts": [ "Getting startedPlatforms and integrations2.1.2772.1.2762.1.2752.1.2742.1.2732.1.2722.1.2712.1.2702.1.2692.1.2682.1.2672.1.2662.1.2652.1.2632.1.2612.1.2602.1.2592.1.2582.1.2572.1.2522.1.2512.1.2502.1.2482.1.2472.1.2462.1.2452.1.2432.1.2412.1.2402.1.2392.1.2382.1.2372.1.2362.1.2352.1.2342.1.2332.1.2322.1.2312.1.2292.1.2282.1.2272.1.2262.1.2252.1.2242.1.2232.1.2222.1.2212.1.2202.1.2192.1.2182.1.2172.1.2162.1.2152.1.2142.1.2122.1.2112.1.2102.1.2092.1.2082.1.2072.1.2062.1.2052.1.2042.1.2032.1.2022.1.2012.1.2002.1.1992.1.1982.1.1972.1.1962.1.1952.1.1932.1.1912.1.1902.1.1872.1.1862.1.1852.1.1832.1.1812.1.1792.1.1782.1.1762.1.1752.1.1742.1.1732.1.1722.1.1702.1.1692.1.1682.1.1672.1.1662.1.1652.1.1632.1.1622.1.1612.1.1602.1.1592.1.1582.1.1572.1.1562.1.1542.1.1532.1.1522.1.1502.1.1492.1.1482.1.1472.1.1452.1.1442.1.1432.1.1422.1.1412.1.1402.1.1392.1.1382.1.1372.1.1362.1.1332.1.1322.1.1312.1.1292.1.1282.1.1262.1.1232.1.1222.1.1212.1.1202.1.1192.1.1182.1.1172.1.1162.1.1142.1.1132.1.1122.1.1112.1.1102.1.1092.1.1082.1.1072.1.1052.1.1012.1.982.1.972.1.962.1.942.1.922.1.912.1.902.1.892.1.872.1.862.1.852.1.842.1.832.1.812.1.802.1.792.1.782.1.772.1.762.1.752.1.742.1.732.1.722.1.712.1.702.1.692.1.682.1.662.1.632.1.622.1.612.1.592.1.582.1.562.1.552.1.532.1.522.1.512.1.502.1.492.1.472.1.462.1.452.1.442.1.432.1.422.1.412.1.392.1.382.1.372.1.362.1.342.1.332.1.322.1.312.1.302.1.292.1.272.1.252.1.232.1.222.1.212.1.202.1.192.1.182.1.172.1.162.1.152.1.142.1.122.1.112.1.102.1.92.1.72.1.62.1.52.1.42.1.32.1.22.1.02.0.762.0.752.0.742.0.732.0.722.0.712.0.702.0.692.0.682.0.672.0.652.0.642.0.622.0.612.0.602.0.592.0.582.0.572.0.562.0.552.0.542.0.522.0.512.0.502.0.492.0.472.0.462.0.452.0.432.0.422.0.412.0.372.0.362.0.352.0.342.0.332.0.322.0.312.0.302.0.282.0.272.0.252.0.242.0.222.0.212.0.202.0.192.0.172.0.152.0.142.0.132.0.122.0.112.0.102.0.92.0.82.0.52.0.12.0.01.0.1261.0.1241.0.1231.0.1201.0.1191.0.1171.0.1151.0.1131.0.1121.0.1111.0.1101.0.1091.0.1061.0.971.0.941.0.931.0.901.0.881.0.861.0.851.0.841.0.831.0.821.0.811.0.801.0.771.0.731.0.721.0.711.0.701.0.691.0.681.0.651.0.641.0.631.0.621.0.611.0.601.0.591.0.581.0.571.0.561.0.551.0.541.0.531.0.521.0.511.0.481.0.451.0.441.0.431.0.421.0.411.0.401.0.391.0.381.0.371.0.361.0.351.0.341.0.331.0.321.0.311.0.301.0.291.0.281.0.271.0.251.0.241.0.231.0.221.0.211.0.181.0.171.0.111.0.101.0.81.0.71.0.61.0.41.0.11.0.00.2.1250.2.1170.2.1080.2.1070.2.1060.2.1050.2.1020.2.1000.2.980.2.960.2.930.2.820.2.750.2.740.2.720.2.700.2.690.2.670.2.660.2.630.2.610.2.590.2.540.2.530.2.500.2.490.2.470.2.440.2.410.2.370.2.360.2.340.2.320.2.310.2.300.2.260.2.21Getting startedClaude Code changelogRelease notes for Claude Code, including new features, improvements, and bug fixes by version.This page is generated from the CHANGELOG.md on GitHub.\nRun claude --version to check your installed version.\nAdded AGENTS.md support: in a project with no CLAUDE.md, Claude Code reads AGENTS.md instead; change it under \u201cProject instructions\u201d in /config (not yet on Bedrock, Vertex or Foundry)\nAdded CLAUDE_GATEWAY_PROXY_IS_EGRESS_BOUNDARY=1 for Claude apps gateways whose only egress is a forward proxy: every outbound request hands the proxy the hostname instead of resolving it locally\nAdded an optional headers: map on Claude apps gateway upstreams, to send static headers to a proxy you run in front of a provider\nAdded a line saying a background task\u2019s update is waiting when it finishes while a panel such as /tasks is open\nFixed claude -p and Agent SDK sessions that could hang with no result after an internal error; they now report the error and exit with code 1\nFixed conversations failing every request with \u201ctext content blocks must be non-empty\u201d when an earlier assistant turn held an empty text block beside other content, including after --resume\nFixed being unexpectedly logged out when an older Claude Code build (for example an IDE extension\u2019s bundled CLI) runs on the same machine as the current one\nFixed interactive start-up hanging or showing an error for ANTHROPIC_API_KEY users when ~/.claude.json holds a malformed customApiKeyResponses value\nFixed update checks erroring every 30 minutes, and claude update hanging when a minimum or maximum version is set, if a proxy returns an invalid version; a malformed minimumVersion is now ignored\nFixed claude update on winget- or apk-managed installs reporting \u201cup to date\u201d when the version lookup failed\nFixed claude plugin install sometimes failing and breaking the installed copy when reinstalling a plugin version that a session or another program was using; an unchanged copy is now left alone\nFixed Grep and Glob reporting no matches when the search could not start because the system was out of processes, memory or file handles; they now return an error saying so\nFixed the Write tool silently ending the turn as a declined permission when the target path is an existing directory; it now reports a clear error\nFixed the Edit tool treating an escaped backslash followed by uXXXX text as a \\uXXXX escape, which could make an edit of a non-ASCII character rewrite an escaped backslash sequence instead\nFixed the Edit tool reporting \u201cInvalid regular expression: regular expression too large\u201d instead of \u201cString not found in file\u201d when a very large edit containing non-ASCII text did not match the file\nFixed a turn ending early with \u201cPath contains null bytes\u201d when a tool call\u2019s file path contained \\u0000 written as an escape sequence; escaped control characters now stay as literal text\nFixed background sessions (claude --bg) exiting when a plugin\u2019s LSP server exited or closed its stdin\nFixed a crash (\u201cType error\u201d) when opening /mcp or /plugin manage with a malformed claudeAiMcpEverConnected value in ~/.claude.json\nFixed a crash at launch when ~/.claude.json holds a malformed theme value\nFixed a crash (\u201cunrecoverable interface error\u201d) when the prompt held text containing terminal color codes, for example a prompt recalled from history or text loaded from the external editor\nFixed a crash when resuming a session whose saved history holds an assistant message stored as a plain string\nFixed sessions on slow or heavily loaded machines sometimes exiting with \u201cClaude Code exited after an unrecoverable interface error\u201d when the first spinner appeared\nFixed a rare case where the screen could stop updating for the rest of the session after an internal rendering error\nFixed a rare case on Windows where a turn could stop with an error such as \u201cOut of memory\u201d right after Claude replied, so that reply\u2019s tool calls never ran\nFixed sessions continued after /clear (restart, --continue, --resume) missing part of their first message when a SessionStart hook printed output, causing a full prompt-cache miss\nFixed messages from other agents (such as a subagent\u2019s SendMessage) that arrived mid-turn showing up below the \u201cRan N shell commands\u201d row instead of where they arrived\nFixed the \u201ccopied\u201d notice not appearing after drag-selecting text in the fullscreen /resume picker and other panels that cover the prompt area\nFixed $TMPDIR expanding empty in Bash commands that run outside the sandbox while sandboxing is enabled\nFixed WebFetch and WebSearch in Cowork cloud sessions not telling Claude why a request was refused, such as a used-up fetch budget or an admin policy\nFixed the Claude apps gateway\u2019s telemetry relay ignoring a collector hostname or domain listed in NO_PROXY when a proxy is set\nFixed one malformed strictKnownMarketplaces or blockedMarketplaces entry silently disabling the whole enterprise marketplace policy\nFixed failed auto-updates leaving large staged downloads behind in ~/.cache/claude/staging\nFixed /plugin not stripping terminal control characters from messages on the Installed tab, such as the error of a failed plugin update\nFixed /plugin \u2192 Installed and /skills crashing when a skill or legacy command is named like a built-in Object property such as constructor or toString\nFixed /plugin closing with no message when every install in a multi-select failed\nFixed uninstalled plugins reappearing as \u201cfailed to load\u201d rows in /plugin Installed, and Remove not clearing such a row\nFixed plugins from the official marketplace being recorded without their commit in installed_plugins.json, and installed_plugins.json keeping the old commit after updating a pinned-commit plugin\nFixed plugin reload previews keeping every previewed copy of a plugin archive unpacked until exit, and overwriting the cached --plugin-url archive a reload falls back to when its download fails\nFixed Remote Control session bookkeeping failing when ~/.claude.json holds a malformed placeholder record\nFixed the error after a revoked claude.ai login blaming an expired Anthropic profile; it now leads with /login\nFixed typed or pasted text occasionally coming out scrambled in the claude agents dispatch input during key repeat or very fast input\nFixed a crash (\u201cunrecoverable interface error\u201d) when resuming a session whose saved transcript contains a stop hook summary without a well-formed hook list\nFixed Enter on a selected agent panel row doing nothing when keybindings.json rebinds Enter in the Chat context, for example to chat:queueSubmit\nFixed PDF page reads on Windows failing when the working folder\u2019s path is long (about 120 characters or more)\nFixed a headless resume (claude -p --resume, the SDK, a VS Code extension window reload) starting the session\u2019s cost and usage totals at zero; headless sessions now save their totals at exit\nFixed project skills from the main repository not loading in --worktree sessions when .claude/skills is untracked\nFixed a sandbox.excludedCommands glob exempting an entire compound Bash command from the sandbox when only one part matched; every part must now match\nFixed resumed subagents and teammates re-rendering the MCP tool definitions they had loaded, which broke prompt caching for that agent\nFixed rate-limited artifact publishes telling Claude to stop retrying; Claude is now told nothing was published and when to send the same publish again\nFixed attachments recorded earlier in a conversation being re-rendered after a resume or relaunch, which dropped extended thinking and missed the prompt cache\nFixed Console sign-in showing only \u201cRequest failed with status code 400\u201d when the server refuses to create an API key; it now shows the server\u2019s message\nFixed messages typed while Claude is still working sometimes being ignored by the model\nImproved session start-up for SDK and headless (-p) use: the first turn no longer waits on the per-directory CLAUDE.md lookup\nImproved the Claude apps gateway\u2019s loopback error messages to name CLAUDE_GATEWAY_ALLOW_LOOPBACK\nImproved /plugin Installed: an MCP server listed apart from its plugin now shows which plugin it belongs to\nImproved claude plugin install on an already-installed plugin: it now says when the marketplace offers a newer version and names the claude plugin update command\nImproved the startup notice overflow line under the logo: it now reads \u201cN more notices hidden\u201d instead of \u201c+N more \u00b7 /status\u201d\nImproved prompt handling: invisible Unicode formatting and tag characters in a prompt are removed and the cleaned prompt is shown for review before it is sent\nImproved /ultrareview when there\u2019s nothing to review: messages say which case you\u2019re in, offer a command that reviews your latest commit, and a new repository\u2019s first commit is reviewed in full\nImproved artifact link handling so Claude reads claude.ai artifact links with the Artifact tool instead of WebFetch when that tool is available\nImproved the dangerous-rm permission prompt to name the flagged rm command and suggest a ${VAR:?} guard, so headless runs can recover\nImproved the Artifact tool\u2019s permission prompts: shorter sentences, pages and artifacts named by title or file name, and links listed after the text\nChanged Fable to always appear in /model on the Anthropic API; it is greyed out only when your organization\u2019s settings disable it\nChanged the Bash sandbox instructions on Bedrock, Vertex and Foundry to the first-party wording, which frames the sandbox as the boundary of what the task was given\nChanged /ultrareview in non-interactive sessions to refuse when the repository has no base branch or shared history\nChanged subagent results to reach the main agent under a header marking them as subagent output, with the result indented, so text in a subagent\u2019s result cannot pass as the session\u2019s own instructions\nChanged workflow scripts\u2019 computed agent() prompts on Bedrock, Vertex and Foundry to reach the subagent framed as script-authored text, so the safety classifier does not read them as the user\nRemoved the background Haiku auto-title request from claude -p runs launched outside an SDK or IDE\nRemoved the deprecated TaskOutput tool; Claude reads a background task\u2019s output file with Read instead, and the taskOutputMaxChars setting and TASK_MAX_OUTPUT_LENGTH no longer have any effect\n[VSCode] Added a Sign out row to the panel menu, with /logout in the typed command menu\n[VSCode] Added background shells and other running tasks to the agent map, each with a Stop, and a typed /tasks that opens it\n[VSCode] Added a Copy response button on responses and a typed /copy\n[VSCode] Added a one-time notice when inactive sessions are archived automatically, and an \u201cUnarchive all\u201d action on the Archived sessions group\n[VSCode] Added the session\u2019s cost and token usage to the Account & usage dialog and the session manager where plan limits do not apply (Vertex, Bedrock, Foundry, API key)\n[VSCode] Fixed the \u201cGeneral config\u201d menu row showing /config usage text instead of opening settings, and made typed /mcp, /hooks, /memory, /rewind and similar commands open their dialogs\n[VSCode] Fixed the effort slider\u2019s level not persisting into later sessions on a model that already had a level saved with /effort\n[VSCode] Fixed Auto missing from the mode picker for conversations opened in an already-used panel when the saved model setting is a differently-cased alias such as \u201cSonnet\u201d\n[VSCode] Fixed /fast not saving fast mode as the default, so it was lost when the extension relaunched Claude Code\n[Claude Code on the web] Added Personal and Organization sections to the environment picker on Team and Enterprise plans, and admins can now share a personal environment with the organization\n[Claude Code on the web] Changed organization environments to open as a read-only summary from the Code tab on Team and Enterprise plans, with editing under Admin settings \u2192 Cloud environments\n[Claude Code on the web] Fixed a cloud environment saved with Custom network access and no domains silently reverting to Trusted; the dialog now asks for at least one domain\n[Claude Code on the web] Changed the admin Claude Code setting labeled \u201cWeb\u201d to \u201cCloud sessions\u201d and removed the redundant read-only Mobile row beneath it\n[Claude Tag] Fixed routines created in a Slack channel on an Enterprise Grid org-wide install failing to read other public channels in their workspace when they ran\n[Claude Tag] Fixed the \u201cLearn more\u201d links on credential presets in Claude Tag access bundles to open each vendor\u2019s credential-setup page instead of a generic API reference\n[Claude Tag] Changed the Pylon credential preset in Claude Tag access bundles so admins can point it at Pylon\u2019s EU host\n[Claude Tag] Fixed Google Cloud credential forms in Claude Tag access bundles: a refused key file now says why, the website and scopes stay locked, and a rejected rotation keeps the pasted key\n[Claude Tag] Fixed the network events log in Claude Tag admin settings showing no response status for requests through connections that use AWS signing, client certificates or a custom CA\n\n\nFixed every request failing with 400 \u2026 Input tag 'advisor_20260301' when ANTHROPIC_BASE_URL points at a proxy or gateway (2.1.275 regression)\n\n\nAdded the signed-in account to Claude apps gateway sign-in: when the gateway names it, you confirm it before the credential is saved, and /status shows it\nAdded a send-now key (ctrl+enter, or ctrl+x ctrl+s) that interrupts the current turn and sends all queued messages at once; sent and queued messages show in gray until the model receives them\nAdded a startup warning when a configured otelHeadersHelper fails, so sessions that silently export no telemetry are noticed\nAdded syncing of the skills and plugins enabled on your claude.ai account to terminal sessions signed in with it; opt out with syncClaudeAiSkills: false or syncClaudeAiPlugins: false\nAdded /plugin install --marketplace , which offers to add the marketplace before installing the plugin\nFixed a restored memory file\u2019s age note changing between requests after a compaction or resume, which caused prompt cache misses\nFixed --forward-subagent-text stream-json and SDK output dropping the messages of subagents spawned by a context: fork skill, and of forked skills invoked by a subagent or another forked skill\nFixed @-mention file suggestions being buried below MCP resources when using a custom fileSuggestion command or typing @./@./\nFixed fullscreen mode placing background-task completion notices beneath a long turn\u2019s collapsed tool row instead of where they arrived; each notice now closes the open row\nFixed claude plugin marketplace update deleting a GitHub marketplace\u2019s local copy when the fetch failed and the marketplace was named after its repository\nFixed plugin and marketplace messages, logs and claude plugin marketplace list showing a password or token stored in a git, ssh or marketplace URL\nFixed a resumed cloud session leaving an unanswered question open in the transcript after a queued message superseded it\nFixed vim mode placing the cursor one character right after a dot-repeated \u201d!\u201d or a fast-typed \u201ci!\u201d switched a non-empty prompt into shell mode\nFixed fullscreen mode freezing or blanking for several seconds when scrolling up past a large file diff\nFixed a stray -style closing tag occasionally appearing in responses\nFixed plugin messages, logs and the VS Code plugin dialog showing the wrong server for some git addresses\nFixed a terminal API Error: 400 on every turn for users behind a network gateway that rewrites API error responses when a beta request header is rejected\nFixed sandboxed Bash commands on Linux reporting exit code 0 for failed commands when the shell is zsh\nFixed the Read tool hanging instead of reporting an error when part of a large file could not be decoded under memory pressure\nFixed --resume, the resume picker preview, resumed background agents and the transcript view failing on a session whose saved history contains a malformed task-reminder or @-file attachment entry\nFixed a crash when resuming a conversation whose transcript contains a malformed message entry, and a fullscreen crash when such a conversation received new messages while scrolled up\nFixed sessions failing to resume or start when their saved transcript contains a malformed message content block\nFixed Grep, Glob and @-file suggestions hanging or running out of memory on searches over the 20MB output cap, and system ripgrep reporting \u201cno matches\u201d instead of an error after a flood of warnings\nFixed /rewind in a forked or background session restoring a zero-filled or truncated file when the session\u2019s file-history backups could not be fully copied\nFixed fullscreen sessions sometimes exiting with \u201cClaude Code exited after an unrecoverable interface error\u201d when typing fast or holding a key with the slash-command dropdown open\nFixed background sessions crashing and restarting their worker when a command fed through stdin ran on a machine that had run out of file descriptors\nFixed a crash at launch when ~/.claude.json holds a malformed mcpNeedsAuthNoticed value\nFixed --resume and --continue dropping a conversation\u2019s earlier thinking when a built-in tool it started with has since been switched off by a server-side flag\nFixed text selected with the mouse in the fullscreen claude --resume session picker never reaching the clipboard\nFixed plugin reload previews replacing a running session\u2019s extracted plugin files when the plugin was loaded from a --plugin-dir or --plugin-url archive\nFixed self-hosted runners with --drain-wait-sec losing the final result of a turn that finished during a SIGTERM drain; the runner now waits briefly for the turn to be reported\nFixed SubagentStop hooks with a specific matcher firing for every stopping subagent whose agent type was empty\nFixed sandboxed Bash commands being unable to write to project directories named hooks/ or config/\nFixed Artifact updates failing with \u201cFile not found\u201d after a session resumes on another machine or its scratchpad is cleared: the page\u2019s last published version is restored\nFixed /update-config writing Write(path) permission rules, which file permission checks don\u2019t match, instead of Edit(path) rules\nFixed four dead documentation URLs (Pricing, Computer Use, Skills, CLI) in the bundled claude-api skill\u2019s live-sources table\nImproved prompt caching for a --system-prompt that contains a __SYSTEM_PROMPT_DYNAMIC_BOUNDARY__ line: the text above it is now cached globally, as the SDK\u2019s array form already is\nImproved the /desktop error when Claude Desktop does not open: it now says why and what to do next\nImproved the Artifact tool\u2019s publish and read results: they now say who can open the page and what the owner\u2019s Share menu offers\nImproved artifact publish results: they name the tab icon sent, warn when the page contains a NUL byte, and retry a flaky fetch of the newer page to merge after a stale publish\nImproved pasted and attached images: they are now saved where Claude can open them as files without a permission prompt, including in Desktop and VS Code\nImproved the Artifact tool\u2019s guidance so Claude updates a shared artifact in place when you were given edit access to it, instead of publishing a separate copy\nImproved plan-usage reads: editor windows and non-interactive sessions on one machine now share a read made in the last minute instead of each calling the usage endpoint\nImproved the ListPlugins tool description so Claude knows it lists plugins enabled on your claude.ai account, not plugins installed locally with /plugin\nImproved responsiveness when the terminal is slow or paused: output no longer falls further behind while the terminal catches up\nImproved Write and Edit results for files in the synced account-skills folder: they now say the change is not saved to your account and how to save it\nUpdated /logout for Claude apps gateway sign-ins to also end the session on gateways that advertise token revocation\nChanged hosted sessions to keep an unanswered permission prompt up after a container restart, instead of asking again\nChanged the Artifact tool to ask for a one-word tab icon on a first publish instead of an emoji favicon\nChanged Claude in Chrome in auto mode to skip the extension\u2019s per-site check for classifier-approved calls, as bypass mode does, fixing browser_batch \u201cPermission denied\u201d after a redirect\nChanged plugins installed from an npm source to be fetched with npm pack --ignore-scripts and integrity-verified, so a package\u2019s install scripts no longer run\nChanged scheduled and Run now routine runs to save data to, and republish the page of, an artifact you can edit without asking; public artifacts, first publishes and deletes still ask\nRemoved the startup notice that told you a one-off scheduled routine had run since your last session\n[VSCode] Added viewing, editing and deleting a saved memory inside the Memory dialog\n[VSCode] Added sending an attached image without typing any text\n[VSCode] Added a Retry link to the MCP servers dialog when the server list fails to load\n[VSCode] Added accept and reject buttons under each change in the proposed-change diff tab, so an edit can be reviewed change by change\n[VSCode] Fixed the transcript creeping toward the bottom in small steps while a permission card waits and content keeps arriving\n[VSCode] Fixed rewound and forked conversations not keeping the permission mode you had picked for the original conversation\n[VSCode] Fixed an empty CLAUDE_CONFIG_DIR entry in the environmentVariables setting making Claude Code keep its files in the workspace\n[VSCode] Fixed plugin install links opening the Manage plugins dialog for plugin names and marketplace addresses that can\u2019t be used in a link\n[VSCode] Fixed Remote Control staying shown as connected after a turn-off that Claude Code reported as failed; it now shows as off\n[VSCode] Fixed the scroll to the bottom on send stopping short of the reply when the reply starts arriving during the scroll\n[VSCode] Fixed the agent map showing agents a crash left unfinished as stopped instead of failed once the session is reopened\n[VSCode] Fixed the \u201cContinuing the step\u201d notice not appearing, and the continue limit resetting, after a reload that follows a crash with background tasks still running\n[VSCode] Fixed the session list showing when a session was last reopened, such as after a window reload, instead of when its last message was sent\n[VSCode] Fixed \u201cFork conversation from here\u201d failing on the message right after one sent while Claude was working\n[VSCode] Fixed the prompt cache clock showing too few minutes after reopening a session with a message sent while Claude was working\n[VSCode] Fixed a background agent that finished while Claude was running a tool losing its completion notice, and its result on the agent map, after a window reload\n[VSCode] Fixed a rare case where text selected in a git-ignored file could be sent to Claude after the extension was unresponsive for several seconds\n[VSCode] Fixed renaming a running session reverting to the generated name (regression in 2.1.269)\n[VSCode] Fixed some claude.ai/code sessions opening in VS Code as an empty conversation with no messages\n[VSCode] Fixed slash commands typed while Claude is responding being sent to the model as text instead of running once the response finishes\n[VSCode] Fixed unreadable code in the plan preview and the Hooks and Permission rules dialogs with the High Contrast Light theme\n[VSCode] Fixed /remote-control being ignored while Remote Control is still connecting: running it again now turns Remote Control off immediately\n[VSCode] Fixed the conversation pulling you back to the bottom while a reply streams after you scroll up, and added a claudeCode.scrollToBottomOnSend setting to turn off the jump on send\n[VSCode] Fixed the Manage plugins dialog showing a password or token that was typed into a marketplace URL\n[VSCode] Improved the agent map: the pill counts running agents and turns red after a failure, the main agent stays in view while the map scrolls, and agents sort by state then end time\n[VSCode] Changed New session in a Claude editor tab to open in the sidebar when Preferred Location is set to Sidebar, instead of always opening another tab\n[VSCode] Changed a message sent while Claude is working to wait at the bottom of the conversation until Claude starts on it\n[Claude Code on the web] Added a \u201cNew routine\u201d button to the page shown when a routine link no longer resolves, next to the link back to your routines list\n[Claude Code on the web] Fixed routine \u201cpaused\u201d and \u201con hold\u201d notifications being cut off mid-sentence; the paused-subscription notice now says to turn the routine back on yourself\n[Claude Code on the web] Fixed cloud environments with a very long allowed-domains list saving fine and then failing every session start; saving now fails up front and says how much to trim\n[Claude Code on the web] Fixed Claude\u2019s guidance when a cloud session on a personal account is denied GitHub access: it now links to claude.ai/connect-github instead of an admin settings page\n[Claude Code on the web] Improved what Claude tells you when asked to edit, delete or run a routine it didn\u2019t create: it now links to the routine\u2019s page so you can do it yourself\n[Claude Tag] Added attach conditions for access bundles in Claude Tag settings: an Owner can let a bundle also apply in channels with guests or Slack Connect channels, not just member-only\n[Claude Tag] Added Amazon CloudWatch, CloudWatch Logs, Amazon SNS, Google Cloud Monitoring and Cloud Logging presets to an access bundle\u2019s Credentials tab in Claude Tag admin settings\n[Claude Tag] Added Datadog presets for the US3, AP1, AP2 and US1-FED sites; new Datadog connections are now limited to Datadog\u2019s read and query API routes\n[Claude Tag] Fixed S3 uploads from recent AWS CLI and SDK versions failing with a 502 error when sent through an AWS connection\n[Claude Tag] Fixed Claude treating a channel as inactive, and skipping untagged messages there, while it was still posting in that channel from a routine or a thread\n[Claude Tag] Fixed a thread\u2019s \u201cClaude [task]\u201d display name reverting to plain \u201cClaude\u201d after the session behind that thread was refreshed or restarted\n[Claude Tag] Fixed the model you switched to in a Slack thread silently reverting to the channel\u2019s default after that thread\u2019s session was restarted or refreshed\n[Claude Tag] Fixed Claude sometimes replying twice when another app or bot @mentioned it in a top-level channel message\n[Claude Tag] Improved Claude\u2019s notices in Enterprise Grid channels shared across workspaces: they now say when no workspace is set up yet, or why only organization defaults apply\n[Code Review] Fixed reviews occasionally dropping part of their analysis when one of the reviewing agents returned its findings in an unexpected format\n[Code Review] Fixed pull requests with more than 100 Claude reviews getting a full re-review on every clean merge from the base branch instead of the lighter merge-focused review\n\n\nAdded a visible warning when memory usage is critical, with steps to free memory or restart safely\nAdded CLAUDE_CODE_MCP_STARTUP_WAIT_MS to bound how long the first non-interactive turn waits for connecting MCP servers (0 = don\u2019t wait)\nAdded effort attribute to the claude_code.llm_request OpenTelemetry trace span, matching the api_request event\nAdded claude_code.managed_settings_resolved OTel event: managed-settings sources and policy helper state; redacted settings and digests with OTEL_LOG_MANAGED_SETTINGS=1\nAdded store.connect_timeout_seconds to the Claude apps gateway config to lengthen the Postgres connect timeout (default 5 seconds), and improved the boot error when the database is unreachable to point to store.postgres_url and the configured timeout\nAdded enduser.sub, the IdP subject, to the telemetry Claude Desktop and Cowork send through a Claude apps gateway\nAdded a Claude apps gateway warning when a replica has more requests open than the 256 it sends upstream at once, and a startup log line showing that limit\nAdded click-to-expand for collapsed teammate and agent messages in fullscreen mode\nFixed sessions getting stuck endlessly retrying \u201cunexpected tool_use_id\u201d 400 errors: corrupted transcripts now self-heal where possible, and otherwise a clear error (with a /rewind hint) ends the loop\nFixed MCP servers configured as http that only speak legacy HTTP+SSE failing to connect when they answer the first request with 422 or another 4xx error\nFixed Streamable HTTP MCP tool calls timing out after about 5 minutes even when a longer per-server timeout was set\nFixed MCP prompts and resources not refreshing when a server sends list-changed notifications without declaring listChanged\nFixed MCP tool calls refused with 403 insufficient_scope being reported as an expired sign-in: the error now names the missing permissions and points to /mcp re-authentication\nFixed hook-driven sessions (such as an active /goal) ending with \u201cPrompt is too long\u201d instead of compacting when the context overflowed again after a reactive compaction\nFixed an active /goal being lost when resuming (--continue / --resume) a session that had compacted\nFixed claude agents losing --model, --effort, --permission-mode, --allow-dangerously-skip-permissions and --agent after an auto-update relaunch\nFixed a per-turn slowdown when a language server publishes project-wide diagnostics for thousands of files\nFixed subagents with model: \"opus\" on Bedrock, Vertex or Foundry leaving the session\u2019s model when its id has no recognizable model family (unless ANTHROPIC_DEFAULT_OPUS_MODEL is set)\nFixed self-hosted runner sessions failing every turn with a 401 after a few failed token refreshes, until the next scheduled refresh; the runner now keeps retrying, and fetches a new token after a 401\nFixed clickable links to local file paths doing nothing in VS Code and other terminals that require a file:// URI\nFixed the transcript renumbering ordered lists in your own messages (typing \u201c3. 2. 1.\u201d displayed \u201c3. 4. 5.\u201d); numbers and \u201cN)\u201d markers now show as typed\nFixed AskUserQuestion preview notes being attached to a previously chosen option instead of the highlighted one\nFixed AskUserQuestion preview mode dropping the highlighted option when submitting a note with Enter\nFixed a resumed background agent keeping half of an interrupted tool batch when one of its calls was approved with a message\nFixed a local claude -p --resume started with CLAUDE_CODE_RESUME_INTERRUPTED_TURN not reporting background tasks the previous process left unfinished\nFixed the first turn of a cloud session sometimes starting without the tools of an SDK-hosted MCP server that was still connecting\nFixed background agent notifications claiming the agent had no live background work when it was still waiting on its own background task and would resume\nFixed error hints in Claude Desktop sessions to suggest slash commands like /usage-credits instead of CLI flags that cannot be used there\nFixed /schedule saving a routine\u2019s prompt without its message role when Claude writes the routine in the shape that listing routines returns\nFixed /status not showing the apiKeyHelper failure that its own error banner told you to check\nFixed /fast on in non-interactive sessions reporting on and then turning off under an organization\u2019s managed fast mode policy; it now says the organization has disabled it\nFixed the Artifact tool asking you to approve an update to an artifact that it then refused because the session had not read the latest version\nFixed Cowork and claude.ai cloud sessions with network access on treating reads of a teammate\u2019s artifact as if network access were off\nFixed a plugin or marketplace directory with no git repository of its own taking its version from an enclosing git repository, such as a git-managed ~/.claude\nFixed --strict-mcp-config with an empty --mcp-config holding the first non-interactive turn for up to MCP_TIMEOUT on incidental MCP servers\nFixed Stop prompt hooks re-sending their whole prompt on every block in a conversation; repeat blocks now name the condition with a 500-character label\nFixed extra empty editor windows opening at startup on Linux under Wayland when running inside the Cursor or VS Code terminal\nFixed an unhandled promise rejection in the Claude apps gateway when Postgres drops a connection during a spend check\nFixed Claude apps gateway cutting every open stream on SIGTERM: it now lets in-flight requests finish for up to 25 seconds before exiting (CLAUDE_GATEWAY_DRAIN_TIMEOUT_MS)\nFixed installed_plugins.json being rewritten on nearly every start-up when plugin policy comes from remote managed settings, which made Claude Desktop reload every open session\u2019s plugins\nFixed headless and SDK sessions making a separate model call for every background task that finished; completions already queued are now answered by one call\nFixed the Bash tool re-sourcing the shell profile (a multi-second stall on the next command) after every plugin reload; it now does so only when the plugins\u2019 bin/ directories changed\nFixed plugins with a top-level $schema in hooks/hooks.json showing an \u201cunknown key\u201d notice\nFixed MCP connection errors and the MCP login tool\u2019s description showing secrets resolved from ${VAR} placeholders in MCP configs\nFixed Bash permission checks for commands that loop over or assign certain special shell variables; these commands now ask for permission\nFixed worktree-isolated sessions accepting Bash commands with certain nested shell expansions; these are now refused\nFixed the Edit permission prompt preview sometimes showing a different location than the approved edit in files with multi-byte characters\nFixed background commands being stopped after 30 idle minutes on machines under mild memory pressure; they\u2019re now stopped only when memory is critically low, and the debug log says why\nFixed a message a subagent sends to the main session disappearing from the Claude Desktop transcript after a relaunch\nFixed a plugin loaded from a .zip being served from a stale extraction after several overlapping reloads\nFixed a sub-agent\u2019s progress summary being replaced by a runaway multi-paragraph reply\nImproved startup in --input-format stream-json sessions: the first turn no longer waits up to 2s for still-connecting MCP servers whose tools tool search defers; they arrive on a later turn\nImproved Monitor tool notifications: a script\u2019s final output and its exit now arrive as one notification instead of two, saving a model turn\nImproved Artifact tool errors: when you are not signed in to claude.ai the terminal now says so on the first attempt, and Claude is told to stop retrying a rejected call sooner\nImproved artifact publishing: a publish built on an older version is stopped before it is sent, with the newer page to merge\nImproved safety checks before removing an agent worktree that contains submodule checkouts\nImproved OTEL_LOG_RAW_API_BODIES=file: output: a new index.jsonl and request_body_id / message.id event attributes link each response to its request file and transcript message\nImproved Claude apps gateway boot: it now tries the first Postgres connection up to three times before exiting, so a database that is reachable a few seconds late no longer fails the boot\nImproved the Claude apps gateway\u2019s spend-limit check under load: it now takes one database round trip instead of four, so fewer checks time out on a busy gateway\nImproved Claude apps gateway sign-in rate limit errors: /login now explains the refusal, and the gateway log says which limit was hit and which setting to change\nChanged Bedrock, Vertex, Foundry and telemetry-disabled installs to use the v2 MCP client and MCP 2026-07-28 negotiation with direct HTTP servers by default, as other installs already do (opt out: MCP_SDK_GENERATION=v1 or MCP_PROTOCOL_NEGOTIATION=legacy)\nChanged /code-review to use leaner inline review prompts for every model that has no tuned settings of its own, instead of spawning many review subagents\nChanged \"type\": \"sdk\" MCP entries in .mcp.json, settings, plugins and agent files to be skipped with a warning: only an SDK host application can register in-process servers\nChanged artifact watching in local sessions: a new version published elsewhere no longer starts a turn; Claude learns of it from a later Artifact tool result\nChanged plugin and marketplace clones to leave Git LFS files as pointers instead of downloading them; git lfs pull in the checkout fetches them\nChanged self-hosted runners to skip a read-only repository the git host refuses at the access check instead of failing the session start\nChanged the /status GitHub line to read \u201cCloud sessions\u201d, and /web-setup, /ultrareview, and teleport messages to say \u201ccloud session\u201d instead of \u201cClaude Code on the web\u201d\n[VSCode] Added continuation of the step a window reload interrupted, labeled in the chat, with a Claude Code: Continue After Reload setting to turn it off\n[VSCode] Added Memory and Instructions entries to the Customize menu: Memory shows the auto-memory toggles, the saved memories and the memory folders, and Instructions edits the CLAUDE.md files\n[VSCode] Added a claudeCode.lockEditorGroups setting to stop Claude from locking the editor groups it opens in\n[VSCode] Fixed a /btw side question asked in a new conversation\u2019s first seconds occasionally showing another session\u2019s side-question history\n[VSCode] Fixed a brief freeze when the extension first looks up your global gitignore file\n[VSCode] Fixed a message sent while Claude was running a tool disappearing from the conversation after a window reload\n[VSCode] Fixed the Manage Plugins enable toggle and MCP servers dialog rows being unreachable from the keyboard\n[VSCode] Fixed sign-ins and sign-outs made in a terminal not showing until a reload after CLAUDE_CONFIG_DIR changed in the Environment Variables setting\n[VSCode] Fixed Edit diffs in the chat being cut off at the bottom at some panel widths and for long wrapped lines; diff boxes now fit the rows shown\n[VSCode] Fixed overlapping settings writes from the extension leaving ~/.claude/settings.json unparseable or dropping a setting\n[VSCode] Fixed Open in New Tab (Ctrl/Cmd+Shift+Esc) sometimes leaving the new tab\u2019s message box unfocused, so typing went nowhere until you clicked it\n[VSCode] Fixed reopening a closed Claude tab splitting the editor layout when its locked group still holds another Claude tab and a file\n[VSCode] Fixed New session opening another locked editor group whenever a file tab shared the group with your Claude tab\n[VSCode] Fixed session names shifting sideways in the session picker while typing a search query\n[VSCode] Fixed the plan review card cutting off its Send feedback button and reason field when a plan has several comments; the comment list now scrolls\n[VSCode] Fixed inline code and code blocks in chat replies being unreadable under the High Contrast themes\n[VSCode] Improved screen reader navigation of the conversation: each message is announced as \u201cYou\u201d or \u201cClaude\u201d, with the tool name for tool steps\n[VSCode] Changed the default global gitignore file to $XDG_CONFIG_HOME/git/ignore when XDG_CONFIG_HOME is an absolute path\n[Claude Code on the web] Added a \u201cCompare against\u201d branch picker to a cloud session\u2019s diff view, so you can diff its changes against any branch instead of only the base branch\n[Claude Code on the web] Fixed git operations in cloud sessions failing with \u201cservice unavailable\u201d when GitHub\u2019s token renewal briefly errors\n[Claude Code on the web] Fixed editing a routine occasionally making it fire twice or re-enabling a routine that had just been paused\n[Claude Code on the web] Fixed commits in cloud sessions occasionally failing with a signing error for a few minutes after the session\u2019s credentials refreshed\n[Claude Code on the web] Fixed the toast after saving a routine whose GitHub trigger couldn\u2019t be linked to show the reason, such as a per-repository trigger limit, instead of only \u201cedit to retry\u201d\n[Claude Code on the web] Fixed sessions sometimes flipping back to unread right after you mark them read\n[Claude Code on the web] Changed routines to skip a run and retry for up to 72 hours when the owner\u2019s GitHub connection is missing, instead of switching the routine off at the first failed check\n[Claude Code on the web] Changed a routine\u2019s on-hold notice: when your subscription is paused it now tells you to turn the routine back on yourself instead of promising an automatic resume\n[Claude Tag] Added a Guests setting to the Add channel and Add workspace forms in Claude Tag admin settings, so owners can pick Inherit, Allow, Channel only or Restrict up front\n[Claude Tag] Fixed Claude not answering when another Slack app or bot @mentions it; the tag now gets a reply and wakes Claude in a channel it had stopped following after days of inactivity\n[Claude Tag] Fixed Claude missing another app\u2019s message that tagged @Claude right after a new Slack channel was created; it\u2019s now delivered once Claude has joined\n[Claude Tag] Fixed Claude folding a follow-up sent minutes after its last Slack message into it as a silent edit; late updates such as blockers now post as a new reply that notifies\n[Claude Tag] Fixed Claude\u2019s Slack search failing with an error whenever it searched within a single channel; it now returns that channel\u2019s matching messages\n[Claude Tag] Fixed a safety-filter stop silently resetting a Slack thread\u2019s context when nobody was waiting; Claude now always says so and no longer cancels background work still running\n[Claude Tag] Fixed email addresses in Claude\u2019s Slack replies rendering with a visible mailto: prefix; they now show as the plain, clickable address\n[Claude Tag] Fixed Claude refusing to watch an Enterprise Grid channel shared with the whole organization when asked from another workspace in the grid\n[Claude Tag] Fixed the Environment picker in Claude Tag admin settings showing a raw environment ID instead of the environment\u2019s name for archived or app-created environments\n[Claude Tag] Improved Claude\u2019s live progress checklist in Slack: capped at 2,000 characters, reposted at most every 15 minutes in busy threads, with older \u201cLatest task list\u201d links updated\n[Claude Tag] Removed the repeated guest-attribution note Claude appended to a Slack canvas each time it edited one in a channel using the \u201cChannel only\u201d guest setting\n[Code Review] Fixed re-reviews occasionally leaving a fixed finding\u2019s thread open when the new review also filed a lower-severity note under it\n[Code Review] Fixed rare reviews ending with \u201cCode review encountered an error\u201d when GitHub or an internal service failed transiently at launch; they now wait and retry\n[Code Review] Improved how Code Review words each posted finding: short plain sentences that say who is affected, where the code goes wrong, and the fix up front\n[Code Review] Improved the check-run card and PR comment when a review is skipped because of an organization limit: each cause now links the admin page that fixes it\n\n\nAdded x-claude-code-request-class, x-claude-code-agent-type, x-claude-code-prev-tool-durations, x-claude-code-compaction and x-claude-code-context-compacted request headers for LLM gateways; opt in with CLAUDE_CODE_GATEWAY_HINT_HEADERS=1\nAdded a notification when an MCP server disconnects mid-session and automatic reconnection gives up, pointing at /mcp\nAdded forking a session started with claude --remote-control or /remote-control from the Claude app; the fork runs as a background session on your computer\nFixed Bash commands the permission checker cannot fully analyze skipping the prompt under permissions.blockReadsOutsideWorkingDirectories, and a subshell hiding a dangerous rm in bypass mode\nFixed skills synced from claude.ai staying available after your organization turns Skills off; they now move to the recoverable trash\nFixed allowManagedMcpServersOnly, deniedMcpServers and disableClaudeAiConnectors set via MDM or managed-settings.json being ignored when server-managed settings are also present\nFixed 401/403 errors on Bedrock, Vertex and Foundry, and Claude apps gateway 403s, telling you to run /login; the message now names the credential to refresh or points to your gateway administrator\nFixed /login, /upgrade, and /extra-usage discarding earlier thinking from the conversation, which forced a full prompt-cache rewrite on the next request\nFixed auto mode stopping for approval when the Artifact tool uploads a file you attached to the chat in a cloud or Remote Control session\nFixed a long-running session recreating a stub .git/info/exclude after the repository\u2019s .git directory was removed or moved away\nFixed the main prompt dropping a ! typed at the start while already in shell mode, so negated commands like ! grep \u2026 can be typed\nFixed Read on macOS refusing a dragged-in screenshot, or any file the system reports under a second path, with \u201csymlink resolution changed after permission was checked\u201d\nFixed permissions.blockReadsOutsideWorkingDirectories: a memory directory chosen by a repository\u2019s settings is no longer loaded into the prompt, recalled, indexed, or used by memory extraction\nFixed sub-agents and background agents being reported as failed, with their result never delivered, when the final streamed reply omitted token usage or carried no model id\nFixed the context meter and auto-compact counting advisor-tool turns at roughly twice their real context size, which made auto-compact fire at about half the real window\nFixed /tui refusing to restart because of an agent-team teammate that had already finished its work and was no longer shown in the agents panel\nFixed saved scheduled tasks running in the wrong session after .claude/scheduled_tasks.json was copied into another folder, such as a new worktree\nFixed SDK and --output-format stream-json output dropping a subagent\u2019s remaining messages and final report after it is moved to the background mid-run (e.g. by CLAUDE_AUTO_BACKGROUND_TASKS)\nFixed /install-github-app reporting a SAML single sign-on block as \u201cadmin permissions required\u201d\nFixed Remote Control clients attached to a Claude Desktop, VS Code or JetBrains session being refused when they ask for the session\u2019s context window usage\nFixed the spinner showing a doubled ellipsis (\u201d\u2026\u2026\u201d) on compaction status lines such as \u201cRunning PreCompact hooks\u2026\u201d\nFixed a false-positive spinner tip suggesting the frontend-design plugin after reading or publishing Artifacts\nReverted a 2.1.268 change that checked Read and Edit deny rules on Bash lines the permission checker can\u2019t analyze (eval, env -C); commands like time -p make build prompt again instead of being denied\nImproved responsiveness in long sessions: hook progress and sub-agent activity no longer re-process the whole conversation on every update\nImproved the Artifact tool\u2019s error when a publish includes a file type artifacts don\u2019t serve: Claude is told which types are served and what to do instead, and the terminal shows one plain line\nImproved the Artifact tool\u2019s page read to state the capabilities and database rules the artifact service holds for the page, for anyone who can publish to it\nImproved artifact database writes: an update can now remove a single field instead of rewriting the whole document\nImproved artifact publishing: a publish whose connection drops after reaching claude.ai is now re-sent safely instead of failing or creating a duplicate version\nImproved the cloud-session GitHub error for an IP allow list, a suspended app installation or SAML single sign-on to show the cause instead of a generic install hint\nImproved /autofix-pr: when gh pr view fails it now shows gh\u2019s own error (sign-in, SAML, rate limit) instead of a generic exit-code line\nImproved /autofix-pr to say why GitHub webhook delivery couldn\u2019t be set up for the PR (for example, no linked GitHub account) instead of a generic warning\nImproved /web-setup errors: a refused GitHub token now lists the likely reasons and the fix, and a connection failure names a configured proxy or TLS certificate problem\nImproved the in-session SSL certificate and proxy connection errors to name the error code and what to fix, such as NODE_EXTRA_CA_CERTS for an untrusted corporate CA\nImproved the error when a cloud session can\u2019t be created because your Claude login expired or was revoked: it now tells you to run /login\nImproved the error shown when an MCP server\u2019s sign-in expires mid-session to say how to re-authenticate (/mcp)\nChanged auto mode on Bedrock, Vertex and Foundry to use the local classifier by default for now; set CLAUDE_CODE_AUTO_MODE_SERVER=1 to use the platform\u2019s server-side classifier\nChanged OTEL_LOG_TOOL_DETAILS=1 to also include real agent, skill, plugin and MCP server names on cost and token metrics\nChanged sign-in with a Claude account to also request access to your claude.ai plugins\nChanged /bug and /feedback reports to include only model-behavior params (model, system prompt, tools) from the last API request, omitting request metadata and CLAUDE_CODE_EXTRA_BODY fields\n[VSCode] Fixed \u201cReport a problem\u201d still appearing, and /bug / /feedback opening a report form, for organizations that have product feedback disabled\n[VSCode] Fixed a red \u201cClaude Code process exited with code 4294967295\u201d banner appearing after completed turns on Windows\nWindows: Improved the network-path permission check for UNC paths when a mapped network drive was added with --add-dir\n[Claude Code on the web] Fixed routines losing access to an organization connector, and still calling the old one, after an admin removed and re-added that connector\n[Claude Code on the web] Fixed creating a self-hosted environment from organization settings occasionally failing with a server error and leaving a half-created environment behind\n[Claude Code on the web] Changed the admin \u201cShare cloud sessions\u201d setting to live under Data and privacy instead of the Claude Code page, where Data and privacy admins can also manage it\n[Claude Code on the web] Added a \u201cDiscard unsaved changes?\u201d confirmation before the New routine page or the Edit routine dialog throws away a routine name, prompt or edit you typed\n[Claude Code on the web] Removed the full-page desktop-app download screen that new users without a cloud environment saw on Mac and Windows; they now go straight to setup\n[Claude Code on the web] Improved the routine detail page: menu and rename in the breadcrumb, the on/off switch and Run now at the top, and run history beside the routine\u2019s settings\n[Claude Tag] Fixed Claude going silent minutes after reinstalling the app when an Enterprise Grid was disconnected but one of its workspaces stayed connected\n[Claude Tag] Fixed scheduled tasks set up in an organization-shared private Slack channel silently never posting; they now keep running in the thread they were created in\n[Claude Tag] Fixed replying in an older Slack thread while Claude is mid-task sometimes restarting it from scratch and losing work it had not pushed yet\n[Claude Tag] Fixed Claude occasionally dropping a message with an incorrect \u201ccouldn\u2019t find a Claude Code environment\u201d notice right after your account token refreshed\n[Claude Tag] Fixed AWS connections refusing region-less endpoints such as Budgets, Savings Plans, WAF Classic and Import/Export; Global Accelerator requests now sign correctly\n[Claude Tag] Improved AWS connection failures: when a request can\u2019t be signed, such as a hostname with no region, Claude is told why and how to fix it instead of a bare error\n[Claude Tag] Fixed OAuth client-credentials and JWT-bearer connections failing with providers that return a lowercase token type; requests now send the standard Bearer scheme\n[Claude Tag] Fixed adding a channel manager being refused on Enterprise Grid shared channels, on channels where Claude hasn\u2019t been used yet, and on legacy private channels\n[Claude Tag] Changed Claude to start watching related public channels on its own, such as an incident channel a conversation depends on, instead of only when asked\n[Claude Tag] Fixed the admin Memory page not listing Slack channels Claude set up on its own even when they had saved memory; admins can now open, edit and delete that memory\n[Code Review] Fixed merging the base branch into a PR whose earlier review listed \u201cAdditional findings\u201d triggering a full re-review; these pushes now get the lighter follow-up review\n[Code Review] Fixed a whole REVIEW.md being ignored because of an @-mention, a code span wrapped across lines, or a backticked HTML tag; only lines linking to changed files are withheld\n[Code Review] Improved suggested fixes to say what the fix must keep working when other code depends on the behavior being changed\n[Code Review] Improved review comments that point to a second affected location to state that location\u2019s issue in a full sentence instead of a cut-off stub\n[Code Review] Fixed /ultrareview --post so a retry after a GitHub error posts the findings comment exactly once instead of never or twice; the comment now names the reviewed commit\n[Code Review] Fixed empty or content-identical pushes being re-reviewed on GitHub repositories whose owner or name contains a capital letter; these pushes are now skipped\n\n\nBug fixes and reliability improvements\n\n\nAdded fast mode in Claude Code Remote sessions (cloud and self-hosted runners): the host\u2019s fast-mode setting or /fast typed in the session applies where your organization allows it\nAdded mouse support to the /config panel in fullscreen mode: the wheel scrolls the settings list, a click on a setting\u2019s value changes it, and the row under the pointer is highlighted\nAdded claude self-hosted-runner --drain-marker-file : when that file exists at a SIGTERM drain, the runner reports its exit to the server as a host drain (telemetry only)\nAdded per-command allowed_domains to Bash, PowerShell and Monitor in auto mode with sandboxing: the hosts a command needs are reviewed with it and opened for it alone; other hosts are refused\nAdded omitClaudeMd to agent frontmatter and --agents JSON, letting custom and plugin subagents run without user, project and local CLAUDE.md files; managed policy files still load\nAdded --accept-command to claude plugin install and claude plugin update to accept exactly the command a previous --json run displayed, instead of -y\nAdded support for a multiplier above 1, up to 10, in the modelPricing managed setting and the Claude apps gateway pricing block, for marked-up internal chargeback rates\nAdded a spinner tip pointing Bedrock, Vertex AI, Foundry and LLM gateway users to the Claude desktop app; the claude.ai desktop app tip now suggests /desktop, which offers to download the app\nFixed a cached organization policy being reused after switching accounts, organizations, or API keys, and the policy not refreshing until the hourly check when the credential changes mid-session\nFixed the tool and command lists not updating when the organization policy finishes loading after startup or changes mid-session\nFixed an enterprise managed-mcp.json that can\u2019t be read or parsed being ignored: it now keeps exclusive MCP control (user, project and plugin servers don\u2019t load) and warns at startup\nFixed org policy being fetched through, and rejected by, third-party local proxies set via ANTHROPIC_UNIX_SOCKET; they are again treated like other custom gateways, including for Remote Control\nFixed cloud sessions rejecting every subagent tool call (\u201cupdatedInput \u2026 failed schema validation\u201d) when a workflow or agent approval was applied after the session\u2019s worker restarted\nFixed /fast off answering \u201cFast mode unavailable\u201d instead of turning fast mode off when the organization has fast mode disabled\nFixed sessions started with CLAUDE_CODE_SKIP_FAST_MODE_ORG_CHECK re-sending fast requests every turn after the API rejected fast mode; the rejection now stands and its reason is shown\nFixed fast mode under CLAUDE_CODE_RETRY_WATCHDOG failing the turn on a usage-credits limit, or retrying an overload at fast speed, instead of falling back to standard speed\nFixed Bash permission checks missing the file that fmt, column and similar commands read when it follows an option the checker doesn\u2019t recognize\nFixed Bash permission checks skipping files a wildcard expands to when the wildcard sits in a command\u2019s pattern or option value (for example grep -v dir/* file)\nFixed Bash permission checks so that shell variable declaration flags cannot misrepresent the command being run\nFixed Bash commands with two directory changes, a subshell, or a cd+git chain skipping the prompt under permissions.blockReadsOutsideWorkingDirectories in bypass and auto mode\nFixed a stale .git/config.lock breaking git checkout -b, git push -u and git config for the rest of a session after a sandboxed command failed to start (Linux)\nFixed settings file changes made outside the session going unnoticed on macOS machines whose system file-event service is saturated; the watcher now falls back to polling\nFixed resumed claude -p sessions whose tools all come from MCP servers failing with \u201cAt least one tool must have defer_loading=false\u201d\nFixed turns failing with \u201cAPI returned an empty or malformed response\u201d when an LLM gateway returns the non-streaming reply as text/plain\nFixed sustained high CPU usage and repeated tool-list requests when an MCP server sends list_changed notifications in a tight loop\nFixed MCP OAuth mishandling client registrations: denying consent forced a new one, one for another redirect URI was reused, and a concurrent write could delete a valid one or keep a mismatched one\nFixed tool search returning no match when Claude selects an MCP tool by its bare name instead of its full mcp__server__tool name\nFixed Ctrl+O cancelling pending MCP server reconnects, and /mcp sent from Remote Control failing while the transcript view is open\nFixed the Claude in Chrome prompt telling the model to load tools through ToolSearch when ToolSearch is unavailable\nFixed cross-session messages held by the receiving session\u2019s permission-mode policy leaving no trace: headless senders now get a delivery notice, and SendMessage results no longer imply it was read\nFixed Claude starting a second copy of a background command (such as a watch task or dev server) that was still running after the conversation was compacted\nFixed /model warning about losing the conversation cache when switching back to the model the conversation actually ran on\nFixed /reload-skills reporting a skill count that disagreed with the slash menu after /cd\nFixed /resume and /continue showing only 1-2 sessions in fullscreen mode on short terminals\nFixed /resume and /teleport keeping the previous conversation\u2019s file-read tracking, so Claude could edit files the resumed conversation had never read\nFixed --resume dropping the 1M context window ([1m]) when the resumed session\u2019s model family differs from the configured default model\nFixed artifacts attached with /artifacts disappearing from the session after --resume\nFixed background sessions (claude --bg, claude agents) not watching the artifacts they publish for republishes made elsewhere\nFixed custom agents, slash commands and output styles beyond the first not loading from a virtual drive that reports inode 0, such as an encrypted vault mounted as a Windows drive\nFixed self-hosted runner sessions silently losing all host config (settings, skills, plugins, MCP servers) when the host config directory exceeds 64 MiB; added --host-config-snapshot disk|memory\nFixed skills synced from claude.ai staying on disk indefinitely after signing out; copies not refreshed within cleanupPeriodDays now move to the recoverable trash at the next launch\nFixed spinner tips suggesting commands that aren\u2019t available for your account type or are disabled in your session\nFixed the /add-dir path input: the left and right arrow keys now move the cursor, and Enter adds only the typed path instead of also adding the highlighted completion\nFixed text fields outside the main prompt moving a leading ! to the end of what you typed (!foo came out as foo!)\nFixed the interactive /hooks menu crashing when a hook matcher is named after an inherited object property such as __proto__ or constructor\nFixed a fullscreen rendering glitch where text kept a stale background color after the box around it lost its background\nFixed Delete in st and Alt+arrow keys in rxvt-unicode not working in attached background sessions\nFixed the terminal\u2019s replies to capability queries (^[[?1;2c) appearing at the shell prompt or in an editor when Claude Code exits, is suspended, or opens an editor right after starting\nImproved terminal rendering performance: large diffs and long transcripts render faster, with fewer slow frames\nImproved startup time slightly by skipping a redundant validation of built-in model data on every launch\nImproved hook feedback: while a SessionStart, UserPromptSubmit, PreToolUse or SessionEnd hook runs, the spinner says so with elapsed time, and Esc cancels a prompt waiting on a SessionStart hook\nImproved the spinner status during long thinking: it now reads \u201cdeep in thought\u201d after 45s, and shows \u201cpicking the thought back up\u201d while recovering from the output-token limit\nImproved dynamic workflows to pause when you hit your usage limit and continue automatically when it resets, instead of dropping the affected agents\nImproved Remote Control to leave fewer empty sessions on claude.ai when setup fails on a flaky network\nImproved the Claude in Chrome message in cloud sessions when the browser can\u2019t be reached: it now says the computer may be asleep before it suggests an install\nImproved claude mcp serve: a running tool call now sends a progress update every 30 seconds, so clients show it is still running and idle timeouts don\u2019t abort a long command that prints nothing\nImproved Foundry and Claude Platform on AWS sessions: an alwaysLoad MCP server that finishes connecting mid-conversation is usable on the next turn without a tool-search round trip\nImproved Markdown files published as artifacts: they now render as styled document pages (title header, document typography, syntax-highlighted code)\nImproved Artifact tool publish errors: a publish with no file now says to write the page to a file first, and an unsupported file type is reported before a missing favicon\nImproved the Artifact tool\u2019s error when a page declares a capability its contract version lacks: it now lists every supported capability and notes when a newer contract version has it\nImproved artifact watching: a session can now watch up to 10 published artifacts at once for republishes made elsewhere, up from 5\nImproved PDF @-mentions to say \u201cpage count unknown\u201d instead of a page count guessed from the file size when pdfinfo cannot count the pages\nImproved /mobile to show a single QR code for claude.ai/mobile, which opens the right app store for your phone\nChanged auto mode so that a skill\u2019s or slash command\u2019s inline ! shell commands follow default-mode permission rules instead of the classifier; a command no rule decides runs as a reviewed tool call\nChanged auto mode so a subagent reports back to its caller through a dedicated hand-back call that the safety classifier reviews, instead of its last message being reviewed after the fact\nChanged Monitor watches to always have a deadline (at most 30 minutes; 10 in single-prompt -p runs) and notify Claude to re-arm, replacing the no-timeout persistent option\nChanged the IDE selection indicator in the prompt to a [\u29c9 \u2026] pill that wraps with the text instead of squeezing multi-line prompts; delete it with Backspace to leave the selection out\nChanged the default dynamic workflow size to small on Pro plans and lowered the medium size guideline from 15 to 10 agents\nChanged Claude apps gateway, Bedrock, Vertex AI, and Foundry sessions so that they no longer refresh a leftover claude.ai login that the session does not use\nUpdated the bundled claude-api skill to enable eager_input_streaming on streaming custom tools, and to start deliverable-shaped Managed Agents work with user.define_outcome\n[VSCode] Added an Attach Open File setting that, when turned off, stops the open file from being added to messages; selected text is still attached\n[VSCode] Fixed the Hooks and Permission rules dialogs reporting a save that landed as failed, and the Hooks dialog going blank under a plugin-only policy lock or showing color codes in save errors\n[VSCode] Fixed Hooks dialog saves: no duplicate hook on replace, a header name retyped in other capitals keeps its secret, and settings.local.json is gitignored before the save returns\n[VSCode] Fixed session history showing only the current session when the workspace is on a Windows mapped network drive or SUBST drive\n[VSCode] Fixed the session list\u2019s Active filter hiding open idle sessions when Open is also checked in the filter menu\n[VSCode] Fixed a new chat switching back to the previous chat when the session list refreshed\n[VSCode] Fixed open tabs and the side bar keeping the old config folder until a window reload after CLAUDE_CONFIG_DIR changed in the environmentVariables setting\n[VSCode] Fixed console windows flashing on Windows when the extension runs background commands such as git, ripgrep, and the sign-in status check\n[VSCode] Fixed the prompt cache clock\u2019s hover text appearing only after a delay, and the auto-compact icon showing the browser\u2019s own tooltip beside its popup\n[VSCode] Improved the Hooks dialog: a save refused because of the settings file itself now opens a popup with an \u201cOpen settings file\u201d button and the reason behind \u201cCopy error\u201d\n[VSCode] Changed the on state of toggle switches from Claude orange to the editor theme\u2019s button color\n[Claude Code on the web] Fixed a cloud session sometimes taking about ten minutes to respond after its process exited while the session still looked live; sending a message now restarts it right away\n[Claude Code on the web] Changed the Routines page on claude.ai/code to a new layout with Yours and Templates tabs and two-column routine cards that show run status, and removed its calendar view\n[Claude Code on the web] Added a Custom network access option to the Cloud environments editor in admin settings, with the same allowed-domains list the environment dialog on claude.ai/code offers\n[Claude Code on the web] Improved the Cloud environments admin page: it shows the default environment for Claude Tag and Claude Code, with a link to change it, and marks the recommended kind to create\n[Claude Tag] Fixed Claude in a channel where it stays active losing its working context about once an hour when the conversation is mostly in threads; thread activity now keeps it from being reset\n[Claude Tag] Fixed a thread that asked Claude to watch a pull request no longer hearing about CI failures, comments and reviews after Claude was restarted in that thread\n[Claude Tag] Fixed deleting the first message of a thread Claude had already replied in not ending Claude\u2019s work there; it now stops, as it did when a message with no replies was deleted\n[Claude Tag] Fixed Claude holding back a post because of an earlier instruction addressed to a different bot or assistant; only instructions addressed to Claude bind it, and it asks when unsure\n[Claude Tag] Fixed the reply-mode card Claude posts on joining a busy channel saying it \u201csees a lot of automated posts\u201d when the channel is only chatty or large; the card now names the real reason\n[Claude Tag] Improved the Environment picker in Claude Tag admin settings: options are labeled Anthropic-hosted or self-hosted, with links to edit that environment or create one\n[Code Review] Fixed a pull request in a repository reviewed once per PR sometimes getting no review when a commit arrived while its review was waiting to start; it now reviews the requested commit\n[Code Review] Fixed Code Review occasionally posting the same findings two or three times when GitHub reported an error for a review it had in fact created\n[Code Review] Fixed follow-up reviews re-posting a security finding a person had already resolved when a later push moved the lines it was anchored to\n[Code Review] Fixed reopening a finished /ultrareview cloud session in the Claude app starting the whole review over again unprompted\nWindows: Fixed PowerShell commands failing with \u201cExit code 1\u201d and no output when the session\u2019s temp output path reaches 260 characters\n\n\nFixed read-only git commands in Bash unexpectedly asking for permission after a session had been running for a while (regression in 2.1.269)\n\n\nAdded claude plugin eval: run a plugin\u2019s eval suite against Claude Code and get scored, reproducible results (JSON + HTML report); see claude plugin eval --help\nAdded /output-style [name] to list and switch output styles, including over Remote Control and in cloud and other headless sessions\nAdded a diff of the files a Bash command changed to the Bash tool result when the Bash tool handles file edits (setting bashEditDiffEnabled)\nAdded OTEL_METRICS_INCLUDE_REPOSITORY to tag OpenTelemetry metrics and events with vcs.* repository attributes; commit events get vcs.ref.head.* with OTEL_LOG_TOOL_DETAILS\nAdded CLAUDE_CODE_GATEWAY_MODEL_DISCOVERY_TIMEOUT_MS to extend the LLM gateway /v1/models discovery timeout (default 3s)\nAdded a spinner tip suggesting /focus for a view with just your prompt, a one-line work summary, and the response\nAdded CLAUDE_CODE_WORKFLOW_MAX_CONCURRENT_AGENTS (1\u2013256) to raise the Workflow tool\u2019s per-run concurrent agent limit for inference-bound fan-outs\nFixed the prompt cache being partially invalidated on the turn after a response was cut off at the output-token limit and automatically resumed\nFixed a case where resuming a session after interrupting Claude mid-thought could change how earlier context was re-sent, hurting prompt-cache reuse\nFixed F1/F2/F4 not working in kitty-protocol terminals and Delete in st, Alt+arrows acting as Escape in rxvt-unicode, and Shift+punctuation typing the unshifted key in WezTerm (regression in 2.1.247)\nFixed remote and headless sessions reporting \u201cwaiting for your input\u201d while background agents were still running (set CLAUDE_CODE_BG_TASKS_REPORT_RUNNING=0 to restore the old behavior)\nFixed the terminal\u2019s replies to capability queries (^[[?1;2c) appearing as stray text at startup in some terminals\nFixed rows at the top or bottom of the transcript going blank in fullscreen after resizing the terminal\nFixed a deny or ask permission rule starting with ! applying beyond the settings source that wrote it; such a rule now applies only within its own source, and a bare ! negation is ignored\nFixed the git status Claude is told after a compaction: it is now the current status, not the one from the start of the session\nFixed synced plugin MCP servers not connecting when a remote session resumes\nFixed resumed headless sessions losing a turn\u2019s replies when the model was switched or a request was retried mid-turn\nFixed terminal escape codes, line breaks and oversized text from a background task\u2019s on-disk record reaching the task list and task notifications when work is resumed\nFixed CMYK JPEG images failing to attach with \u201ccannot decode\u201d; they are now converted and resized like other JPEGs\nFixed the managed settings approval dialog not naming the collector for a gRPC telemetry endpoint set without a scheme\nFixed plugin headersHelper consent prompts showing a URL path that could be misread as a different host\nFixed plugin errors showing [redacted URL] in place of a relative Windows path with a folder name that starts with @\nFixed missing cursor in the permission-rule, auto-mode-rule, add-directory, session-rename and feedback-review text fields when the terminal\u2019s native cursor is enabled\nFixed repeated clicks on a /fork receipt, each under a second apart, never backgrounding the session right away while it waited for the current tool to finish\nFixed plugin LSP servers that reject shutdown params (e.g. rust-analyzer) being left running at session end; exit is now sent even if shutdown fails\nFixed the attribution reminder overriding a CLAUDE.md or memory rule against commit and pull request attribution; lines set by managed settings still apply\nFixed prompt suggestions being dropped for text in Japanese, Chinese, Thai and other languages written without spaces between words\nFixed synchronized output being assumed from the terminal\u2019s name in GNOME Terminal and Konsole versions that do not support it\nFixed permission_denials in --output-format stream-json results omitting Read, Edit and Write calls blocked by a path-scoped deny rule\nFixed sessions run through the SDK or the desktop app showing an unknown status in other sessions\u2019 agent list\nFixed /insights failing on Bedrock, Vertex, Foundry, and gateway deployments whose account can\u2019t reach the default Opus model by using the session model there instead\nFixed organization policy limits not loading for the session when another Claude Code process refreshed the login at the same moment\nFixed Claude Desktop sessions using Bedrock, Vertex, or a gateway not getting the contextual \u201cwhat Claude needs\u201d turn-end notification text\nFixed MCP servers reconnecting when an updated config only changed the order of the server URL\u2019s query parameters\nFixed the prompt box\u2019s top border splitting into extra lines when viewing a background agent whose name or description has line breaks or is wider than the terminal\nFixed sessions getting permanently stuck on \u201cPrompt is too long\u201d when auto-compaction had no complete earlier exchange to summarize (mostly Agent SDK sessions with very large prompts)\nFixed /goal runs silently stalling after API errors, network drops, or token limits: the goal now retries with backoff, or pauses and says why, including until a usage limit resets\nFixed prompt cache misses in cloud sessions by waiting briefly for server configuration before the first request\nFixed /btw answers that contained made-up tool calls and output: the side question is now told not to write them, and any that appear are flagged as not executed\nFixed CLAUDE_CODE_RESUME_INTERRUPTED_TURN re-running a turn that had failed with an API error over 6 hours earlier, or longer ago than CLAUDE_CODE_RESUME_INTERRUPTED_TURN_MAX_AGE_MS when set\nFixed organization plugins enabled through managed settings not loading in headless sessions and on Claude Desktop (once Desktop bundles this CLI version); they load from the next session\nFixed plugin archives extracted for a session being readable by other local users, extracted files keeping world-writable bits from the archive, and stale files surviving re-extraction\nFixed Edit() deny rules and the write-path check not applying to the file a Bash tee command writes; a Bash(tee:*) allow rule no longer covers destinations outside the working directories\nFixed stray characters like 22c, or a terminal\u2019s color or version reply, being typed into the prompt at startup over slow connections (ssh, browser terminals)\nFixed the terminal\u2019s block cursor showing under the interface in rxvt-unicode after leaving or re-entering fullscreen\nFixed the cursor block staying visible after returning from an external editor in fullscreen mode on rxvt-unicode\nFixed the interface being drawn twice after returning from an external editor (Ctrl+G) outside fullscreen mode\nFixed the interface being drawn twice in Konsole after returning from an external editor\nWindows: Fixed PowerShell tool commands sent to the background stopping when Claude Code exits\nImproved the /diff panel to open fully rendered in one step instead of showing a loading state first\nImproved prompt suggestion filtering for Japanese, Chinese and Korean text: mixed-script and single-word suggestions are kept, and meta or evaluative text is dropped as it is for English\nImproved the Skill tool\u2019s \u201cUnknown skill\u201d error to name the plugin skill\u2019s full name when a bare name matches exactly one plugin skill\nImproved keyboard support over SSH and in unrecognized terminals: terminals that answer the kitty keyboard query (such as foot and Alacritty 0.16+) now get Shift+Enter and Ctrl+Shift shortcuts\nImproved responsiveness in long sessions: transcript updates no longer re-process the whole conversation to build the collapsed tool-use summaries\nImproved first-party sessions with telemetry disabled: an alwaysLoad MCP server that finishes connecting mid-conversation is usable on the next turn without a tool-search round trip\nChanged /ultrareview --post to post the PR comment directly when the findings arrive and print the comment link, instead of starting a second cloud session to post it\nChanged artifact database reads that save into the session scratchpad so they no longer stop for working-folder approval\nChanged skills synced from claude.ai in cloud sessions to be named anthropic-skills:, matching Claude Desktop; the bare name still works when nothing else uses it\n[VSCode] Added an agent map: an \u201cN agents\u201d footer pill opens a map of the session\u2019s sub-agents with per-agent cards, Stop agent, and read-only transcripts\n[VSCode] Added a Hooks dialog to the command menu for viewing hooks and adding, editing, or removing them in user, project, and local settings; managed, plugin, and session hooks stay read-only\n[VSCode] Added live progress rows for running subagents under the tool-call groups in Focus view\n[VSCode] Added a Permission rules dialog that lists permission rules and adds or removes them in user, project, and local settings; startup-option, session-only, and managed rules stay read-only\n[VSCode] Added a Cancel button to the Switch account screen that returns to your session as the current account\n[VSCode] Fixed Focus view showing a turn started by a delivered plain-text prompt, such as a scheduled task\u2019s, as part of the previous turn\n[VSCode] Fixed the footer\u2019s prompt cache clock hiding its minutes when the panel is narrow\n[VSCode] Fixed the session list keeping sessions from the default folder when CLAUDE_CONFIG_DIR is set in a settings file or the environmentVariables setting\n[VSCode] Fixed a plan preview that finished loading late sometimes hiding its comment box or showing an older plan\n[VSCode] Fixed a plan preview accepting comments that went nowhere after its Claude tab closed\n[VSCode] Fixed the prompt cache clock and reopen notice for a session compacted after its last reply and then closed, which now reads as cold when reopened\n[VSCode] Fixed a session renamed in the extension while Remote Control is on keeping its old name on claude.ai/code\n[VSCode] Fixed the \u201cEnable Remote Control for all sessions\u201d toggle keeping its last position after the setting was reset to default from a terminal\n[VSCode] Fixed restored Claude tabs not counting as open in the session list after a window reload until clicked, and their row opening a second tab\n[VSCode] Fixed Switch account making a tab forget its dismissed usage-limit warnings when you sign back in as the same account\n[VSCode] Fixed a session rename being replaced by the generated name after a window reload when the session was renamed during a long turn\n[VSCode] Fixed the sidebar usage meter keeping a stale per-model weekly limit row after the account loses that limit\n[VSCode] Fixed a rare case where an @-mention sent with the keyboard shortcut while a new chat view was still starting could be inserted into the input long after the keystroke\n[VSCode] Fixed the session list jumping down when the Account & usage header appeared a moment after opening the Claude side bar\n[VSCode] Improved documents and messages written for someone other than the user: Claude now writes them for that audience and names it at the top of its reply\n[VSCode] Improved screen reader and keyboard accessibility in the slash-command menu, @-mention menu, output-style picker, Send/Stop button, permission and question cards, and onboarding checklist\n[VSCode] Changed the current-file chip in the message box: an X now removes it, replacing the Hide toggle\n[VSCode] Removed the Claude Code items from a session tab\u2019s right-click menu and the editor title bar\u2019s \u201d\u2026\u201d menu; they could not act on the tab the menu was opened on\n[Claude Code on the web] Added taking back a queued message in a cloud session before Claude reads it: remove it from the queue, or press Esc or Up, and the text returns to the message box\n[Claude Code on the web] Fixed /model default in a cloud session leaving every later message failing in organizations that restrict which models Claude Code can use\n[Claude Code on the web] Fixed one-off scheduled routines occasionally running a second time after a transient server error\n[Claude Code on the web] Fixed routine runs that use subagents sometimes being treated as finished too early, which could skip the retry after a real failure or start a duplicate run\n[Claude Code on the web] Fixed file links in cloud session transcripts opening a GitHub 404 when Claude was working from a subfolder of the repository\n[Claude Code on the web] Changed the Cloud environments admin page to list every environment instead of capping each table at five rows behind a Show more control that could be unreachable\n[Claude Code on the web] Changed claude.ai/code for Free-plan users to open the plans page with a path to upgrade, instead of a \u201cDisabled by org admin\u201d page with no way forward\n[Claude Tag] Added a confirmation dialog before Connect all or Disconnect on a GitHub installation in admin settings, to guard against accidental organization-wide changes\n[Claude Tag] Fixed threads occasionally going silent after a failed turn because the failure notice was dropped when Slack briefly rate-limited it; the notice is now retried\n[Claude Tag] Fixed Claude accepting a switch to a model your organization hasn\u2019t enabled and then quietly answering with a fallback model; it now declines and says an admin can enable it\n[Claude Tag] Fixed a table posting as raw pipe text when Claude attached files to the same message; the table now posts as a normal reply and the files follow with a plain caption\n[Claude Tag] Fixed @Claude !restart at the top level of a channel where Claude isn\u2019t active starting an unrelated conversation; it now privately says there is nothing to restart\n[Claude Tag] Fixed plugin rows in Slack access settings showing an unlabeled raw ID with no way to turn the plugin off; they now show its name and link to the bundle that manages it\n[Claude Tag] Fixed the shared-session banner and Share dialog on sessions started from Slack claiming the whole organization could open the link; they now name the Slack channel\u2019s audience\n[Claude Tag] Improved load time of the admin settings page and its Slack channel picker, most noticeably for organizations with many channels or several connected workspaces\n[Claude Tag] Improved scheduled routines in Slack channels: a routine run can now reply in an existing thread instead of always posting a new top-level channel message\n[Claude Tag] Improved the timestamp on Claude\u2019s live progress checklists to show each reader\u2019s local time and how long ago it was updated, instead of a fixed UTC time\n\n\nAdded to the Claude apps gateway: with pricing: set in gateway.yaml, signed-in Claude Code clients receive the same rates through managed settings, so /cost and telemetry match the spend meter\nAdded a startup warning for gateways when access_control.allow_cidrs is empty, and a one-time warning the first time a request arrives from a public address\nAdded the gatewayInternalNetworks managed setting, letting administrators allow /login to a Claude apps gateway on their organization\u2019s own public IPv4 block\nAdded claude self-hosted-runner --remove-session-state (default off): delete each session\u2019s per-session directories under /_sessions/ when the session ends\nAdded configDirectory to the output of claude auth status --json\nAdded --json to claude plugin install, uninstall, update, enable and disable, and errorDetails/noteDetails to each row of claude plugin list --json\nAdded browser-tab icons for published artifacts, chosen by Claude to match each page\nFixed every turn failing with HTTP 400 on third-party Anthropic-compatible endpoints (ANTHROPIC_BASE_URL) since 2.1.265: a regex in the Artifact tool\u2019s input schema that those endpoints reject\nFixed WebFetch hanging indefinitely on a server that keeps the response open without finishing; a fetch now fails after 300 seconds. Set CLAUDE_CODE_WEBFETCH_DEADLINE_MS to override the deadline (0 turns it off)\nFixed a respawned in-process teammate picking up tools or a system prompt from a same-named agent file in a folder you have not trusted\nFixed sustained high CPU usage: a busy loop in long-running idle sessions no longer pins a CPU core, and rapid terminal focus reports during a session recap no longer keep the CPU high\nFixed Claude sometimes replying \u201cyour message came through empty\u201d after an MCP tool call\nFixed deny and ask permission rules on symlinked directories (/etc, /tmp, /var on macOS; /bin on Linux) not applying when a path was given by its real location, and Bash commands ignoring deny rules written on a symlinked path spelling\nFixed a case where a Read or Edit deny rule did not apply when an env -C, eval or similar command the permission checker cannot analyze was on the same line\nFixed plugin and marketplace errors showing a token or password from a git source URL\nFixed /mcp and /plugin server details, claude mcp list/get, and MCP login errors showing secrets resolved from ${VAR} placeholders in MCP configs\nFixed prompt caching and extended thinking breaking mid-session for SDK sessions using excludeDynamicSections: the first message is no longer re-rendered each request\nFixed entitled users being told a model is restricted after restart or in the Desktop Code tab when a cached model-access denial was stale\nFixed a running session silently switching to the organization\u2019s default model when another Claude Code process refreshed a stale model-access entry\nFixed long-context 429s on Fable models showing the usage-credits consent prompt instead of the 1M-context message on Pro and Team plans\nFixed workload identity federation via a profile (as claude-code-action configures it): processes sharing the profile could fail mid-run with 401 \u2026 jti reused\nFixed MCP server OAuth sign-in failing with \u201cNo available ports for OAuth redirect\u201d when the local callback port range can\u2019t be bound\nFixed the conversation summary produced by /compact and auto-compact mangling text that contained $ sequences\nFixed resuming a conversation that ended with /compact: its restored-file notes now load in the same order on every resume\nFixed SDK prompt suggestions, side questions and /rename sending the conversation from before a compaction\nFixed @ file and / command suggestions not appearing after recalling a previous prompt with the up arrow and editing it\nFixed claude agents: pressing \u2190 again at a natural pace to go back to the agent list no longer gets ignored until you pause for over a second\nFixed claude agents session delete getting stuck when a worktree can\u2019t be removed: the message names the cause and next step, and for a git worktree ctrl+x again deletes the directory anyway\nFixed background agent and workflow rows in the agents panel expanding to many lines when their text contained line breaks\nFixed Claude in Slack sessions losing their Slack tools when org managed settings set an MCP allowlist\nFixed Claude in Chrome asking to allow the host \u201chttps\u201d when a navigation URL had a scheme but a host that could not be parsed\nFixed the spinner wrapping onto several lines when the current task\u2019s label is long; the label and the \u201cNext:\u201d task line now stay within one terminal row\nFixed the /bug and /feedback description field showing no cursor when the terminal\u2019s native cursor is enabled\nFixed Remote Control sessions served by claude remote-control showing a generated name instead of their session title in ListAgents\nFixed claude plugin validate rejecting plugin paths whose directory name begins with two dots, which the plugin loader accepts\nFixed plugins silently skipping a default monitors file or root SKILL.md that could not be checked\nFixed WebFetch\u2019s error for localhost and other dotless hostnames to explain why the URL is refused and suggest curl\nFixed PermissionRequest hooks not firing in --print mode\nFixed policy-helper warnings not printing on headless (-p) runs\nFixed /resume listing a /fork background session under its parent\u2019s name instead of its own \u2442 fork name\nFixed /remote-control and other claude.ai-gated commands to suggest /login when signed out instead of showing a Claude for Enterprise migration message\nFixed CLAUDE_CODE_SESSIONEND_HOOKS_TIMEOUT_MS not extending SessionEnd hooks that have no per-hook timeout (they were still cancelled after 1.5 seconds)\nFixed /autofix-pr and other cloud-session commands saying to retry or install the Claude GitHub App when no GitHub account is connected; they now point to /web-setup or the web connect page\nFixed cloud-session commands such as /teleport and /remote-env to explain when an organization policy turns them off, instead of answering \u201cUnknown command\u201d\nFixed Bash sandbox instructions over-stating confinement: no unenforced path lists when filesystem isolation is off, and strict mode no longer claims commands can never run unsandboxed\nImproved fullscreen mode: adding or removing a prompt line (Shift+Enter) now repaints as fast as typing a character instead of re-rendering the visible transcript\nImproved --continue / --resume: the conversation appears immediately instead of waiting for SessionStart hooks, and the first message no longer re-reads the whole transcript\nImproved responsiveness during tool-heavy turns by no longer redrawing the transcript for a hidden per-tool-batch reminder\nImproved startup time in projects with .claude/workflows/ scripts: listing them no longer parses each script\nImproved auto mode denials: the message Claude receives now names the rule that blocked the action and asks Claude to try a safer method and finish unrelated work before stopping to ask you\nImproved Claude in Chrome: long page reads now stay inline instead of being saved to a file and read back\nImproved the MEMORY.md truncation warning to say how many lines were cut and where the cut starts\nImproved the terminal permission prompt for artifacts: it now leads with the ask\u2019s question\nImproved the prompt footer: an editor or /diff selection now shows inside the prompt input, and fullscreen mode shows Remote Control status in the header instead of the footer\nImproved the \u201cUsage credits required for 1M context\u201d message to say that usage credits turned on mid-session take effect after restarting Claude Code\nImproved /plugin: installing, enabling or disabling a plugin now takes effect when you close the menu; /reload-plugins is no longer needed afterwards\nChanged the system prompt on Bedrock, Vertex and Foundry to deliver environment, model and settings details as attachments, matching first-party sessions\nChanged Bedrock, Vertex and Foundry sessions to keep the tool list byte-stable across a conversation (late-connecting tools load deferred instead of rewriting it), matching first-party sessions\nChanged the task-tracking tools (TaskCreate/Get/Update/List, TodoWrite) to be offered only on Claude 3.x, Opus 4.0\u20134.7, Sonnet 4.0\u20134.6, Haiku 4.5; set CLAUDE_CODE_ENABLE_TODO_TOOLS=1 elsewhere\nChanged the artifact data-edit permission prompt in the terminal to a card that shows the document count and who can open the artifact\nChanged local Cowork sessions set to skip all approvals: the Artifact tool now refuses a local file outside the session\u2019s folders, or behind a symlink, instead of reading it without asking\nChanged plain WebFetch deny and ask rules to no longer apply to Artifact tool reads and updates; use an Artifact rule (or WebFetch(domain:claude.ai)) to block or gate them\nChanged the \u201cN MCP servers need authentication\u201d startup notice to announce each server once instead of at every launch\n[VSCode] Fixed the session list, settings toggles, and chat tabs when CLAUDE_CONFIG_DIR is set in a settings file or the environmentVariables setting\n[VSCode] Fixed the model pill, model picker and command menu going blank in open tabs for a few seconds after a login, logout or account switch\n[VSCode] Fixed Auto disappearing from the mode picker in new-tab or just-reloaded conversations when a project or local setting overrides the model named in ~/.claude/settings.json\n[VSCode] Fixed session names reverting to the last prompt after a window reload when a SessionStart hook is configured\n[VSCode] Fixed the footer\u2019s model pill and Remote Control pill waiting for the new tab\u2019s Claude process to start when another tab in the window is already up\n[VSCode] Fixed a second Claude process running through its full startup when a session tab\u2019s launch arrived more than half a second after its config read\n[VSCode] Fixed resuming a session from the session list ignoring claudeCode.preferredLocation: \"sidebar\" (it always opened a panel), and programmatic opens resetting that setting to \u201cpanel\u201d\n[VSCode] Fixed Windows issues: the WSL install prompt no longer appears on machines without WSL installed, and IDE diagnostics are now returned correctly for Windows files when WSL is installed\n[VSCode] Fixed the custom style builder saving a User level style in a folder the CLI does not read when CLAUDE_CONFIG_DIR is set through settings\n[VSCode] Added Left and Right arrow keys to change where an always-allow permission rule is saved, for keyboard and screen reader users\n[VSCode] Added a \u201cClaude Code: Focus last message\u201d command that moves keyboard focus to the newest message in the conversation, for keyboard and screen reader users\n[VSCode] Changed the Manage plugins dialog to apply installs, enables, disables and uninstalls to open sessions without a restart\n[VSCode] Changed some artifact permission prompts to omit the \u201cdon\u2019t ask again\u201d choice, matching the terminal\n[Claude Code on the web] Fixed cloud sessions running longer than about six hours silently losing files saved to persisted session folders; saves now persist for up to a day\n[Claude Code on the web] Fixed \u201cInvalid effort level\u201d errors when a routine resumes a session, or a session starts with no set effort, in orgs where an admin caps a model\u2019s effort\n[Claude Code on the web] Improved routine creation from a conversation: when the new routine has no connectors, Claude now says so and how to add them instead of only confirming it\n[Claude Tag] Fixed the admin settings page hanging on a loading skeleton or going blank after a transient load failure; a section that fails to load now shows a Retry button\n[Claude Tag] Added a link from a Slack channel\u2019s configure page back to the organization\u2019s Claude in Slack admin settings\n[Claude Tag] Fixed a Slack Enterprise Grid channel losing its Claude settings (repository, environment, access) after a Slack admin moved it to another workspace\n[Claude Tag] Improved how Claude explains a blocked action: it now says whether a permission check, its own decision to confirm first, or missing access stopped it\n[Claude Tag] Improved reply speed: Claude now runs several read-only lookups (searching Slack, reading a thread, finding people) at once instead of one after another\n[Claude Tag] Improved formatting of comparisons: sentence-length comparisons now come as lists instead of wide tables that scroll sideways, and long table cells wrap\n[Claude Tag] Fixed @Claude !restart in a thread with its own session sometimes also posting a contradictory \u201cthis thread is handled by the channel session\u201d notice\n[Claude Tag] Improved the message shown when your Claude account is in a different organization than the Slack workspace: it now explains how to connect the workspace to your org\n[Claude Tag] Fixed Markdown links whose URL is wrapped in angle brackets showing as literal bracket text in Slack instead of a clickable link\n[Claude Tag] Fixed a workspace guest\u2019s top-level @mention in a channel where guests may use Claude sometimes getting a \u201cyour Slack account isn\u2019t connected\u201d reply instead of an answer\n[Claude Tag] Fixed a channel\u2019s long-running session being replaced with a fresh one mid-conversation; the scheduled refresh now waits until the channel and its threads are quiet\n[Claude Tag] Fixed channel-settings cards clicked more than once telling the proposing session the change was refused after it had already applied; the outcome is now sent once\n[Claude Tag] Changed memory in public channels: each channel now keeps its own notes, and Claude no longer recalls notes it saved in other public channels; workspace notes stay shared\n[Code Review] Added a note under the still-open findings list in follow-up reviews: resolving a finding\u2019s thread, not just replying to it, stops later reviews from counting it as open\n[Code Review] Fixed reviews sometimes ending as incomplete when one of the agents verifying a finding failed midway; the review now replaces that agent and reaches a verdict\n[Code Review] Fixed a push-triggered review that was queued behind a running review still posting after the pull request had been converted to draft\n[Code Review] Fixed reviews ignoring a directory\u2019s CLAUDE.md conventions when the PR edited a root file (e.g. README.md) that only shares a name with a file that CLAUDE.md lists\n\n\nAdded maxEffortLevel setting (top-level or per model under modelSettings): caps the effort level on every provider, including Bedrock, Vertex and Foundry; users can still pick a lower level\nAdded --system-prompt-snapshot off to render the system prompt fresh on every request instead of reusing the conversation\u2019s recorded prompt (for iterating on prompt text)\nFixed Cowork scheduled tasks in the cloud failing at startup for organizations whose managed settings require sandboxing\nFixed /context and other local command output rendering blank on mobile clients\nFixed shift+enter and option+backspace not working after reconnecting to a tmux or ssh session inside an agent view\nFixed the dim last-prompt header not appearing at the top of the conversation when scrolling up in fullscreen mode\nFixed Workflow agent() calls with large output schemas being refused in auto mode instead of being checked by the safety classifier\nFixed a case where a marketplace entry path containing a backslash could bypass the containment check for fetched marketplaces on macOS and Linux\nFixed expired AWS or Google Cloud credentials under a host app such as Claude Desktop retrying ten times with a generic \u201crequest failed\u201d before the re-authenticate error appeared\nFixed resuming a session after /compact or another slash command ran via -p --resume: a spurious \u201cContinue from where you left off.\u201d turn is no longer inserted\nFixed resuming a large session (transcript over 5 MB): parallel tool calls and their hook output are no longer dropped from the reloaded conversation\nFixed managed allowedHttpHookUrls, httpHookAllowedEnvVars and allowedChannelPlugins to admit nothing, not everything, when unreadable\nFixed /login on machines whose managed settings require Claude apps gateway sign-in: Esc now closes the dialog instead of doing nothing\nFixed artifact publishes cut off by a dropped connection mid-upload: they now retry once when Claude Code can tell the upload never completed, instead of reporting an unknown outcome\nFixed effort: frontmatter on custom commands, skills, and subagents being ignored on models whose default effort is still pinned (Opus 4.7, Opus 4.8, Fable 5)\nFixed artifact publish failing with an unhelpful error when the page file isn\u2019t valid UTF-8 or contains a replacement character (U+FFFD); the error now names the line and column to fix\nFixed claude agents @ directory menu not listing repositories created after the session started\nFixed Remote Control clients that join a Claude Desktop or VS Code session showing a stale permission mode until it was changed again\nFixed claude remote-control exiting and dropping every attached session when its server credential expires (about 30 days after start); the host now re-registers and keeps going\nFixed the usage-limit warning flickering on and off during a session when requests for different models or modes report different limit windows\nFixed earlier reasoning being dropped when an MCP server re-sends, or a built-in tool re-renders, a tool the model already loaded\nFixed a tool that disappears mid-conversation, from a disconnected MCP server or an upgrade, rewriting the tool list and discarding earlier thinking\nFixed a background worker forked from a conversation adding EnterWorktree to the conversation\u2019s tool block mid-session, which broke prompt-cache reuse\nFixed mid-session MCP and plugin tools being added to the tool list in sessions without ToolSearch, which broke prompt-cache reuse; supported models now receive them as deferred definitions\nFixed switching models with /model re-sending every tool definition (a prompt-cache miss); commit and PR attribution text now arrives as a conversation note that updates on model changes\nFixed resumed sessions rewriting the inline tool set when an MCP connector reconnects at a different moment than before\nFixed resumed sessions re-rendering tool descriptions instead of replaying the recorded ones when the first turn ran a tool\nFixed prompt-cache misses and dropped extended thinking when a claude.ai connector\u2019s tools change between a session and its resume\nFixed resumed sessions rewriting earlier MCP tool announcements (and dropping extended thinking) before their connectors reconnect\nFixed a prompt-cache break when a print-mode (-p) conversation is resumed interactively: the system prompt prefix no longer changes\nImproved the /diff panel: it no longer flashes \u201c0 files changed\u201d and a spinner before settling, and its empty state is centered in the panel\nImproved the Bash tool\u2019s description guidance so Claude describes what a command does in plain words instead of echoing the command\nImproved sandbox guidance so Claude suggests /copy when clipboard commands such as pbcopy fail inside the sandbox\nImproved --resume first-render time for sessions with many Bash tool calls\nImproved prompt input responsiveness: keystrokes no longer occasionally wait a frame behind spinner or streaming repaints\nImproved prompt-cache stability: subagents and sessions started with --system-prompt or --append-system-prompt now record the system prompt and tool definitions once instead of re-rendering them\nImproved Artifact tool publish errors: when a publish is refused, the message now says why and what to do about it\nSelf-hosted runner: Changed --use-anthropic-git-proxy to be reported to the server at registration and to print a warning for each session that still clones through the legacy git proxy\nGateway: Changed forward_user_identity upstreams to return a 429 as-is to a developer whose email was forwarded, instead of failing over to the next upstream, so the proxy\u2019s per-user limits hold\n[VSCode] Fixed the extension host hanging at 100% CPU when forking, editing an earlier message, or rewinding in a conversation whose saved transcript contains a cyclic parent link\n[VSCode] Fixed pasting a screenshot on WSL2/WSLg inserting raw image bytes into the chat input; the image is now attached when the clipboard provides it, otherwise the paste is ignored\n[VSCode] Fixed chat diff blocks always rendering with a dark editor theme; they now follow the active VS Code color theme, including high contrast\n[VSCode] Fixed mixed right-to-left and English text rendering in the wrong order while typing in the message input\n[VSCode] Fixed accepting an edit in the diff view on a file with Windows (CRLF) line endings failing with \u201cString not found in file\u201d\n[VSCode] Fixed @-mentions dropping files whose paths contain spaces\n[VSCode] Fixed the sessions list view failing to load in windows connected over Remote-SSH when the workspace folder exists only on the remote host\n[VSCode] Fixed runaway ripgrep processes when viewing files in large or symlink-heavy workspaces\n[Claude Code on the web] Fixed GitHub Enterprise Server sessions showing your GitHub account as disconnected once its token expired; PR and issue operations now refresh it automatically\n[Claude Code on the web] Fixed gh and GitHub API calls failing in organizations without the Claude GitHub App; they now use your connected GitHub account and say so when none is connected\n[Claude Tag] Added a \u201cUse a custom connector\u201d link to the preset connection forms in Claude Tag admin settings, so you can switch to a custom connection without starting over\n[Claude Tag] Fixed Claude replying \u201cThe API rejected the request as invalid\u201d when the organization has run out of usage credits; the reply now says so and explains how to add more\n[Claude Tag] Fixed thread requests to edit or delete a message Claude posted at the channel\u2019s top level being answered with a correction instead of reaching the session that posted it\n[Claude Tag] Fixed Connect on Tool access requests under Admin settings > Review requests failing with \u201cAuthorization failed\u201d or showing the requested access bundle as deleted\n\n\nFixed a 2.1.265 regression affecting LLM-gateway and proxy setups: the undocumented CLAUDE_CODE_USE_GATEWAY environment variable, previously ignored unless ANTHROPIC_BASE_URL and ANTHROPIC_AUTH_TOKEN were both set, began forcing Cloud-gateway sign-in on its own in 2.1.265, so configurations that set it alongside an API key, apiKeyHelper, or custom auth headers failed every request with \u201cNot signed in to the Cloud gateway\u201d. The variable on its own is ignored again; no configuration change is needed\n\n\nAdded user.email and user.groups to the telemetry Claude Desktop and Cowork send through a Claude apps gateway, matching terminal sessions\nAdded support for pointing --plugin-dir at a folder of plugins: each child folder with a manifest loads, and children added or removed while running are picked up\nAdded a 1 GB cap on tool results saved to disk; the in-conversation preview says when a saved file was truncated\nFixed resuming a foreground-spawned subagent changing its tool list and system prompt prefix, which broke prompt-cache reuse for that agent\nFixed agent teammates and resumed subagents moving SubagentStart hook context and preloaded skills out of the prompt prefix on later turns, which broke prompt-cache reuse\nFixed resume after the previous process died while a tool was running: the last prompt is no longer rewritten, and the interrupted tool call is kept and marked interrupted\nFixed /model opusplan[1m] being rejected with \u201cModel not found\u201d\nFixed syntax-highlighted code in permission prompts and messages sometimes omitting a character after a Ruby ?, Erlang $, or Perl $ sigil\nFixed the fullscreen transcript jumping by one row whenever the slash-command or @-file suggestion list opened or closed\nFixed a plugin path containing a backslash bypassing the symlink containment check on macOS and Linux\nFixed plugin directories whose names begin with two dots being wrongly refused as outside the plugin root\nFixed VS Code and SDK sessions occasionally requiring re-login when a session was closed while refreshing its token\nFixed Remote Control sessions sending the end-of-turn signal before the reply\u2019s last message, which could show a reply as finished in the Claude app before its last part arrived\nFixed background (--bg) sessions occasionally being retired mid-turn when a message arrived just before the idle timeout\nFixed Claude Code\u2019s own git status and diff probes running clean filters configured by a nested repository inside the working tree\nFixed the advisor tool and its instructions being re-decided per request from the request\u2019s model; the decision is now made once and announced in the conversation when it changes\nFixed artifact publish accepting connector tool names the connector doesn\u2019t expose; the publish is now refused when none of the declared tools exist, and warned when only some don\u2019t\nFixed /add-dir refusing to load a subdirectory\u2019s agents when managed settings lock only skills to plugins, and promising agents when only agents are locked\nFixed two-key keyboard shortcuts cancelling silently when the second key arrived more than a second later, as happens inside tmux; they now wait 3 seconds and show a notice when they time out\nFixed forked skills (context: fork) not streaming their kickoff prompt and, with --forward-subagent-text, their text turns as progress events in stream-json\nFixed a plugin\u2019s default component folder that the OS cannot check, such as a symlink loop, being silently skipped; it is now reported in /plugin with the error code\nFixed the Claude apps gateway\u2019s OTLP telemetry relay pausing all forwarding to a collector for 30 seconds after it rejected a few payloads as malformed or too large\nFixed /plugin Discover/Browse and claude plugin list --json --available showing no description or display name for marketplace plugins whose metadata lives only in their plugin.json\nFixed /login showing \u201cno gateway URL is configured\u201d when re-run in a session that signed in to a Claude apps gateway set by managed settings\nFixed /model claiming a model was \u201csaved as your default\u201d when the settings file couldn\u2019t be written; it now says the save failed and why\nFixed /clear from Remote Control waiting on SessionStart hooks and on open terminal dialogs before completing\nFixed the /config dialog changing height when switching between its tabs\nFixed resuming a workflow run after its container restarted; a resume whose run journal is missing now fails with a clear error instead of rerunning every agent\nFixed the claude-api skill\u2019s error-code reference: model access failures return 404 and unavailable beta headers return 400, not 403\nFixed non-interactive sessions (-p with stream-json input, Agent SDK, cloud sessions) resetting the shell working directory at each new user message; a cd now persists across turns\nFixed MCP servers configured as http that only speak the legacy HTTP+SSE transport never connecting; Claude Code now falls back to SSE as the MCP spec describes\nFixed some claude.ai connectors in cloud sessions showing as needing authentication even though they are connected in claude.ai (servers that answer an unsupported request with HTTP 401)\nFixed remote sessions keeping their sandbox container alive while a connector approval or sign-in link waits for you\nFixed resumed sessions showing long model-facing recovery instructions in \u201cbackground task didn\u2019t finish\u201d notices instead of a short status line\nWindows: Fixed Read, Write and Edit refusing every file (\u201csymlink resolution changed after permission was checked\u201d) when running inside an AppContainer or restricted-token sandbox\nImproved --worktree startup on large repositories: the new worktree is now checked out in parallel (git 2.32+)\nImproved /workflows agent detail: tool calls are marked running, failed or done, the subagent\u2019s task list is shown when it has one, and Enter unfolds the listed calls with their inputs and results\nImproved slash commands typed mid-prompt: matches now show in a list (Tab opens it outside fullscreen) instead of a single suggestion, and a plugin skill is now found by its bare name\nImproved remote MCP servers that need sign-in: Claude Code no longer registers an OAuth client with them until you actually authenticate\nImproved the time to resume long sessions that read many files\nImproved the error shown when an image over the size limits cannot be decoded: it now names the cause and how to fix it instead of only citing the limit\nImproved the Artifact tool\u2019s read of an artifact someone else wrote: the summary now treats the page as untrusted content and flags embedded instructions rather than relaying them\nUpdated the .claude folder permission option to say what it actually allows: editing files in the project\u2019s .claude folder (or ~/.claude) for the session\nChanged machines with forceLoginGatewayUrl in managed settings to be Claude apps gateway sessions from startup, like forceLoginMethod: \"gateway\"; a leftover claude.ai login or API key is not used\nChanged image processing to use the runtime\u2019s built-in image support; the CLI no longer extracts a native image module to the temp directory\nChanged plugin display metadata to prefer the marketplace entry over plugin.json on the Installed tab and claude plugin details, filling gaps from plugin.json\nChanged Claude apps gateway sessions to export OpenTelemetry directly to a collector the gateway\u2019s managed settings name in OTEL_EXPORTER_OTLP_ENDPOINT, instead of through the gateway\u2019s relay; sessions without a named collector still use the relay\n[VSCode] Added automatic archiving of sessions inactive for a set period (new \u201cArchive inactive sessions\u201d setting, default 14 days)\n[VSCode] Fixed the sidebar chat coming back blank after Reload Window or a restart when the conversation had been open for more than 10 minutes\n[VSCode] Fixed the timeline dot sitting below the text on the \u201cRemote Control is active\u201d message\n\n\nBug fixes and reliability improvements\n\n\nAdded an \u201cOrganization policy\u201d line to /status and claude doctor that says why your organization\u2019s policy could not be loaded, such as a proxy not passing the endpoint through\nAdded bashOutputMaxChars and taskOutputMaxChars settings to raise how much command and background-task output Claude receives inline before it is saved to a file, up to 128K characters\nAdded --append-subagent-system-prompt-file to read the subagent system prompt from a file, for prompts too large to pass on the command line\nAdded /skill-doctor to show which loaded skills go unused and what they cost in context, so you can prune them\nFixed typed or pasted characters occasionally landing out of order or being dropped during fast input or key repeat\nFixed /add-dir printing a false \u201ccouldn\u2019t be resolved\u201d error when the working directory is on a /net automount\nFixed the Bedrock setup wizard hanging when AWS or an AWS credential helper never responds (it now times out with a clear error), and its model checks failing behind a TLS-inspecting proxy\nFixed cloud sessions discarding a plugin synced from claude.ai when managed settings force-enable it in enabledPlugins, then falling back to a marketplace clone that could fail\nFixed being unable to delete the character immediately before an inline [Image #N] chip in the prompt input\nFixed resuming a session losing hook output and other context around parallel tool calls, which changed the resumed request\nFixed Remote Control showing a stale permission mode when a phone, browser, or claude.ai app attaches to a terminal session or after the mode changes in the terminal\nFixed Remote Control sessions showing as still working (stuck spinner and Stop button) after stopping a turn from a connected phone or browser, or after a local slash command like /clear\nFixed SDK and cloud sessions ignoring a Stop or interrupt sent just after the first prompt, before the turn had started; the turn now stops instead of running to completion\nFixed Remote Control uploading a session pulled with /teleport into the connected session, which appeared appended to the original on phone and web\nFixed Remote Control\u2019s inbound event stream failing behind TLS-inspecting corporate proxies on native Windows\nFixed Remote Control sessions showing the default effort level on claude.ai when the effort comes from settings\nFixed gcpAuthRefresh opening a browser at startup when the Google credential check was slow, even though the credential was still valid\nFixed claude.ai connectors staying absent for the whole session when the startup connector fetch timed out \u2014 the CLI now retries in the background\nFixed sustained high CPU usage when a background agent could not be resumed and its wake-up was retried in a tight loop\nFixed feature flags gated to a newer version occasionally applying to an older Claude Code version running on the same machine\nFixed /usage and the VS Code usage panel dropping a model-specific weekly limit row when the usage endpoint is rate limited or when opened right after startup\nFixed claude -p --resume adopting a malformed session ID recorded in the transcript; it now resumes under a fresh session ID instead\nFixed the terminal progress indicator (iTerm2, Ghostty, ConEmu) showing the session as finished while a background workflow or agent was still running\nFixed a rare layout glitch where a box could render with the wrong height after its container switched between row and column direction\nFixed Claude apps gateway client IP when a trusted proxy appends a port to X-Forwarded-For; with an access list set, an unreadable entry now gets 403\nFixed Claude apps gateway telling Claude Desktop to export OpenTelemetry as JSON even when the terminal CLI uses protobuf, so protobuf-only collectors rejected Desktop\u2019s data\nFixed Desktop and web showing a session as busy while it only watches an artifact for updates\nFixed Claude in Chrome file_upload failing with \u201cpaths: expected array, received undefined\u201d in local Cowork sessions run from the Claude Desktop app\nFixed SendMessage to an offline Remote Control session on another machine reading as delivered; the result now says delivery is queued until that machine reconnects\nFixed plugin install hints from CLIs run in background Bash commands: they are now detected, and the raw tag no longer leaks into the conversation\nFixed in-process agent-team teammates re-sending their first-turn tool and skill announcements on the second turn, which changed the request prefix and missed the prompt cache\nImproved the /model picker and the VS Code model pill to show a model\u2019s name instead of its raw Bedrock, Vertex AI, or LLM gateway ID when Claude Code recognizes it\nImproved startup on Google Vertex AI when GOOGLE_APPLICATION_CREDENTIALS is set: API client creation no longer re-runs Google Cloud project discovery or spawns extra gcloud processes\nImproved streaming performance: already-rendered blocks are no longer re-checked by layout on each update\nImproved the dangerous-rm safety prompt to also catch rm -rf on positional parameters and inside double-quoted sh -c scripts\nImproved handling when the API sends no response headers: the retry now waits up to API_TIMEOUT_MS (10 minutes by default) instead of another 3 minutes, and the messages say what to change\nChanged a Claude apps gateway 403 on the managed settings load (at startup or after /login) to say Claude Code may not be enabled for the organization, instead of advising a new sign-in\nChanged machines whose managed settings pin forceLoginMethod: \"gateway\" to ignore a leftover API key or claude.ai login and ask for /login; Bedrock, Vertex AI, and Foundry sessions are unaffected\nChanged auto mode to treat a link that packs content into a public diagram renderer\u2019s URL as an upload to that site: no longer auto-approved unless you asked for it\nChanged the prompt\u2019s word-editing keys to match Bash: Ctrl+W deletes back to whitespace, Alt+F and Alt+D stop at word end, punctuation separates words; keybindingFlavor no longer has any effect\nChanged /context token counting to use a local estimate when the token-counting API is unavailable, instead of extra small-model requests\n[VSCode] Added a \u201cBuild a custom style\u201d walkthrough to the Output styles menu that writes a custom output style file and lists it right away\n[VSCode] Added an Add server form and a Remove action to the MCP servers dialog, so MCP servers can be added and removed without leaving the IDE\n[VSCode] Added a hollow ring in the session list for sessions open in a terminal, another VS Code window, or Claude Desktop, so they no longer look closed\n[VSCode] Added a fold button to permission and question prompts so the conversation behind them can be read without dismissing them; the space beside the prompt now scrolls the conversation\n[VSCode] Added \u201cArchive session\u201d to the session list\u2019s right-click menu and gave Unarchive its own icon\n[VSCode] Fixed a session teleported from Claude Code on the web treating a question that was cut off when the cloud session shut down as declined\n[VSCode] Fixed the session tab\u2019s Rename box opening empty for a tab restored with the window; it now starts with the current name\n[VSCode] Fixed collapsed sections in the session list panel briefly showing expanded each time the panel loaded\n[VSCode] Fixed Focus view showing a tool call as still running after Claude had moved on, such as while a question waited for your answer\n[VSCode] Fixed the session list\u2019s active-row highlight going stale when an unfocused Claude tab\u2019s session ID is corrected\n[VSCode] Fixed Cmd/Ctrl+Shift+T reopen and deep-link opens placing the Claude tab outside the Claude editor group when a Claude tab has focus\n[VSCode] Fixed the session tab\u2019s \u201cAdd to group\u201d putting a session opened from Claude Code on the Web in two groups; it now moves the entry the session list shows\n[VSCode] Fixed the model picker showing models an organization has since disabled until the window was reloaded twice\n[VSCode] Fixed a tab opened from the session list jumping back to that session, and a tab opened from a Web session restarting its teleport or staying empty, after VS Code reloads the tab\u2019s view\n[VSCode] Fixed /btw side-question history from earlier sessions being overwritten when a question is asked right after a window reload or while a settings file has errors\n[VSCode] Fixed the pending question card not reappearing after the Claude panel reloads when signed in with a Claude.ai or Console account\n[VSCode] Fixed claude.ai-only features staying visible in a window\u2019s other Claude panels after one panel picked up a third-party provider from a settings file\n[VSCode] Fixed the sign-in screen appearing despite the Disable Login Prompt setting when Claude Code reports no login or a request fails for lack of one\n[VSCode] Fixed the next queued permission prompt keeping text typed on the previous prompt and accepting an immediate second click\n[VSCode] Fixed install-plugin links opening the Claude sidebar without the install dialog in a window where only the session list had been shown\n[VSCode] Fixed the sidebar usage meter staying empty on a new window until the Account & usage dialog was opened, and a 0% usage limit being left out of the meter\n[VSCode] Fixed \u201cStart new session in this group\u201d losing the group after New conversation, and a missing unread dot for a session that finished before the sidebar\u2019s unread list loaded\n[VSCode] Fixed the editor tab badge showing unread during a running turn or missing on a tab opened from the session list, and \u201cAdd Session Tab to Group\u201d doing nothing for an archived session\n[VSCode] Fixed \u201cEnable Remote Control for all sessions\u201d so flipping it also applies right away to sessions open in other VS Code windows\n[VSCode] Fixed the session list\u2019s Open filter for sessions continued from claude.ai whose tab was still recorded under the web session, and labeled the filter menu\u2019s sections for screen readers\n[VSCode] Changed the model picker to one flat list of every model, with rows kept for older model spellings listed last\n\n\nAdded a diff panel that opens beside the conversation in fullscreen mode and shows your uncommitted changes as Claude edits; toggle it with /diff\nAdded a likely cause for prompt-cache misses (e.g. tool definitions or system prompt changed, idle past the TTL) to /cost and the status line\u2019s prompt_cache field\nAdded /reload-plugins to headless sessions, so it appears in the Claude Code Desktop and SDK command lists\nAdded a text form of /advisor (/advisor, /advisor , /advisor off) for the desktop app, Remote Control, and other headless (-p/Agent SDK) sessions\nAdded oidc.scope_on_refresh to the Claude apps gateway for IdPs that return an id_token on refresh only when asked for openid again\nAdded Claude apps gateway support for newer Claude Desktop keys in desktop policy blocks, including userPluginMarketplacesEnabled and userPluginUploadsEnabled\nFixed Edit/Write/Read permission rules whose path contains parentheses being dropped as invalid or ignored by the Bash sandbox, which left \u201cread-only\u201d folders writable\nFixed one file permission rule with an uncompilable pattern (e.g. an unclosed [) making every file edit fail with Invalid regular expression; such a deny rule now guards the literal path it spells\nFixed Bash permission checks auto-approving zsh commands that hide a command substitution in a REPORTTIME, REPORTMEMORY or DIRSTACKSIZE assignment; these now prompt for approval\nFixed Bedrock model discovery, token counting and AWS SSO/STS credential calls failing with \u201cunable to get local issuer certificate\u201d when the corporate root CA is only in the OS certificate store\nFixed permissions.blockReadsOutsideWorkingDirectories on macOS hiding the user\u2019s git config from sandboxed git and hiding a worktree-isolated sub-agent\u2019s own checkout\nFixed managed settings not loading for claude.ai Enterprise/Team users who also had a leftover API key from an earlier /login\nFixed /status listing a signed-in claude.ai account and a configured API key as if both were in effect; the credential not in use is now marked\nFixed managed skillOverrides entries keyed on a bundled skill\u2019s alias (e.g. checkup for /doctor) not applying, and Skill(name) deny rules not covering a nested skill listed as :name\nFixed model: fable agents ignoring the [1m] tag on an ANTHROPIC_DEFAULT_FABLE_MODEL pin and silently running with a 200K context window\nFixed the /model picker not showing Fable 5.1 for organizations that can use it, which was only accepted when typed as /model claude-fable-5-1\nFixed prompt caching on Claude Fable 5.1 not covering the context attached after tool results, so it was re-sent as uncached input on every tool-call turn\nFixed model switching staying blocked for the rest of the session after a plugin hook load failure; each switch now re-checks and the refusal names the cause\nFixed model switching being blocked for the session when an organization-managed plugin\u2019s marketplace could not be loaded\nFixed SDK-provided MCP servers (e.g. Desktop connectors) sometimes missing from the first turn and only appearing on the next one\nFixed Claude in Chrome tools failing with \u201cNot connected\u201d mid-task in cloud-hosted claude.ai sessions when a connector was added or removed\nFixed flags, joined emoji and accented letters splitting across wrapped lines, and stale text staying on screen when a flag or joined emoji falls in the terminal\u2019s last two columns (now shown as \u2026)\nFixed Remote Control accepting a model pick that is not a valid model name; it is now refused with an error instead of failing on the next message\nFixed /rewind and --rewind-files reporting success when checkpoint backup files were missing and nothing was actually restored\nFixed /rewind leaving stale file-read tracking from the rewound-away turns, which caused \u201cFile unchanged since last read\u201d stubs and full-file re-injection after external edits\nFixed -p --resume/--continue (as used by the desktop app) failing on every retry once a session\u2019s worktree directory lost its git metadata; it now fails once, then resumes without the worktree\nFixed a subagent that resumed another agent via SendMessage never being woken by that agent\u2019s completion (the notification went to the main conversation instead)\nFixed agent teams: an in-process teammate\u2019s transcript losing messages, or going blank, during long API retry waits (e.g. under CLAUDE_CODE_RETRY_WATCHDOG) as retry notices evicted real messages\nFixed a session that moved to the background appearing twice in ListAgents (once as a phantom \u201cinteractive\u201d twin with the same name) and receiving SendMessage deliveries in the viewer\nFixed intermittent \u201ctask output swap refused\u201d errors when many sessions share a project directory\nFixed Ctrl+Z in fullscreen leaving the shell on the alternate screen, drawn over the paused interface\nFixed Workflow tool subagents being restarted as stalled while a long context compaction was still in progress\nFixed plugins from a URL marketplace failing to install with \u201cmarketplace entry path does not stay inside the marketplace directory\u201d when a host app (e.g. Claude Desktop) stores it as a directory\nFixed an extra browser tab opening when an artifact is published in a session you\u2019re driving from claude.ai, the desktop app, or mobile (Remote Control)\nFixed the Artifact tool\u2019s first call failing with an \u201cInvalid tool parameters\u201d validation error in some Cowork sessions\nFixed IDE line selections being dropped when running a skill or slash command (the \u201cN lines selected\u201d context now reaches Claude)\nFixed repository detection for GitLab projects in nested subgroups (e.g. gitlab.com/group/subgroup/project)\nFixed owner/repo#123 issue references in rendered output linking to github.com when working in a GitLab repository; they now link to the gitlab.com issue\nGlob/Grep: Fixed the search path being probed on disk before the permission check; a missing path is now reported after permission is decided, as Read does\nReverted the 2.1.259 change applying Read() deny rules to Bash arguments; it denied npm run build under a Read(./**/build/**) rule in every mode and made cd \u2026 && grep prompt even in auto mode\nImproved structured output: Workflow agent({schema}) rejects a JSON Schema that can never be satisfied up front, and retry-cap errors now include the last validation failure\nImproved deleting a background session whose worktree has unpushed commits: the message now names the branch and commit count, and deleting again discards the worktree\nImproved the Claude apps gateway\u2019s refresh-failure log to name the step that failed\nImproved idle CPU usage of non-interactive (-p / SDK) sessions\nImproved the Claude apps gateway on Amazon Bedrock: input tokens for an aborted request are now counted with AWS\u2019s free CountTokens API (grant bedrock:CountTokens) instead of a one-token request\nImproved the settings error for rules such as Edit(C:\\dir\\(name)\\**), where \\( is read as an escaped parenthesis rather than a path separator, to suggest an unambiguous spelling\nImproved auto-compact for 1M-context models: Opus and Fable sessions now compact shortly before the 1M-token limit, and recovery compaction on very large contexts no longer times out at 10 minutes\nImproved /ultrareview and claude ultrareview to wait up to 45 minutes (previously 30) for long-running cloud reviews\nImproved /effort on Claude Fable 5.1 so changing effort mid-session no longer invalidates the prompt cache\nUpdated the bundled claude-api skill so its Go, Java, and C# samples use current-generation model IDs, and clarified that cheaper worker or sub-agent models should be current-generation too\nChanged ctrl+l / cmd+k in fullscreen mode to clear the transcript view like a terminal clear; scroll up to see earlier messages\nChanged permission rules with text after the closing parenthesis (e.g. Bash(ls) x), which never matched anything, to be reported as invalid settings instead of being silently ignored\nChanged server-managed settings so a managed CLAUDE.md (claudeMd) no longer triggers the security approval dialog; hooks, shell-command, sandbox, and unsafe env settings still require approval\nChanged Claude in Chrome to follow your organization\u2019s Claude in Chrome admin setting; when an admin turns it off, --chrome, /chrome and the browser tools are unavailable\nChanged Claude apps gateway to send orgPluginSettings in the list form read by Claude Desktop 1.15200.0 and later; older desktops ignore it\nChanged Claude apps gateway to also refuse to start, naming the field, when a desktop policy misspells a field in a nested object of a managedMcpServers or orgPluginSettings entry\nChanged commands typed at the ! bash-mode prompt to run outside the sandbox even when strict sandbox mode (sandbox.allowUnsandboxedCommands: false) is on, like typing into your own terminal\nChanged self-hosted runner --kill-session-after-min to release a session that is only waiting on its user (paused, resumable on the next message) instead of killing it and reporting a failure\nRemoved the one-hour time limit on background commands started by subagents; they now run until they exit or are stopped, matching the main session\n[VSCode] Added the selected effort level to the footer model pill, fixed a stale effort level after switching models, and returned the footer pills to their earlier compact size\n[VSCode] Added Open and Closed to the session list\u2019s status filter menu\n[VSCode] Fixed the welcome screen disappearing in a new session when Remote Control turns on automatically\n[VSCode] Fixed the session history picker loading a session a second time when it is already open in another tab; it now switches to that tab\n[VSCode] Fixed the session tab\u2019s Rename command silently doing nothing while the tab\u2019s view was reloading; it now always applies\n[VSCode] Fixed a half-finished message, an empty tool card or an extra \u201cThought for\u201d line staying on screen after Claude Code retried a dropped response\n[VSCode] Fixed \u201cEnable Remote Control for all sessions\u201d not applying to a session tab that was still starting when the toggle was flipped\n\n\nAdded managedMcpServers managed setting: organizations can provide HTTP/SSE MCP servers to every user (same entry shape as .mcp.json); entries that name a command to run are skipped\nAdded --permission-prompts none for unattended headless hosts: anything that would prompt is denied automatically while the active permission mode (including auto mode) keeps deciding\nAdded recognition of glab mr create/merge/close/reopen/note/update so GitLab merge requests show as MR !N in the collapsed tool summary and refresh the footer MR badge\nAdded --json to claude plugin validate for a machine-readable validation report\nFixed concurrent sessions silently reverting each other\u2019s ~/.claude.json changes \u2014 workspace trust no longer resets and MCP/project state is no longer lost when running many sessions at once\nFixed a conversation whose thinking was rejected once being rejected again on every later turn\nFixed Bash Read() deny rules not covering files given as option values (--ignore-revs-file=.env, -f.env, @file), git diff/git grep file operands, or cd DIR && cat FILE compounds; grep -r/cp -r over a directory holding a denied file now asks\nFixed the prompt cache being invalidated when the OAuth token refreshed in sessions with telemetry disabled\nFixed fullscreen mode showing a blank conversation after a long turn with hundreds of tool calls\nFixed auto mode running a turn on a model it doesn\u2019t support when a command or skill\u2019s frontmatter model: named one; the turn now keeps the session model\nFixed CLAUDE_CODE_MAX_CONTEXT_TOKENS being ignored for Vertex-style model IDs (@YYYYMMDD suffix) of model versions Claude Code doesn\u2019t recognize\nFixed the live output preview of a running shell command hiding its newest lines when an earlier line wrapped\nFixed a background GitHub connection check that ran on every launch for claude.ai users; the result is now remembered across launches\nFixed --resume failing (and --continue opening an empty conversation) when a saved session contains an attachment entry with no payload\nFixed frontmatter model: on custom commands and skills being ignored in interactive sessions\nFixed Artifact publishing failing once with an \u201cunexpected parameter note\u201d error in conversations continued from an older version\nFixed managed forceRemoteSettingsRefresh being ignored at startup when a policy helper configured by MDM or the managed settings file had already run\nFixed worktree isolation refusing hook-created worktrees on machines where git rev-parse fails with a message other than \u201cnot a git repository\u201d\nFixed OpenTelemetry metrics and events from cloud sessions missing the user.email, organization.id, and user.account_uuid attributes\nFixed MCP servers that disconnect while their tools are being listed at startup showing as connected with no tools instead of reporting the error\nFixed the file edit permission dialog sometimes showing a changed line cut short with no indication\nFixed repository detection dropping a known repo identity after a transient git probe failure\nFixed managed settings silently going unenforced when the managed-settings file, a drop-in, the MDM plist, or the HKLM value cannot be parsed: Claude Code now refuses to start and names the source\nFixed Stop not actually stopping background agents and workflows in remote-control sessions: killed tasks now stay visible and re-stoppable until their processes exit\nFixed resuming a workflow run while its previous stopped run was still exiting, which could run duplicate copies of its agents\nFixed marketplace repo URLs on github.com with a trailing slash or dangling ?/# producing an unusable .git clone URL\nFixed blocking Stop hooks causing the turn after a block to lose the model\u2019s reasoning from that turn and, on some models, miss the prompt cache\nFixed remote (claude.ai) sessions taking 60 seconds to start a turn after a browser-hosted MCP server\u2019s page had gone away\nFixed worktree-isolated sessions refusing common Bash loops, xargs pipelines and launcher-wrapped commands that cannot reach the main checkout\nImproved terminal resize and first-render performance for long responses by reusing text measurements\nImproved /workflows agent detail: JSON outcomes are pretty-printed with syntax colors and real line breaks, and long outcomes fold behind an expand toggle\nImproved headless/SDK session start: the first turn begins up to 50 ms sooner when MCP servers finish connecting\nImproved /install-github-app to explain it is GitHub-only and point to the GitLab CI/CD docs when run inside a GitLab repository\nImproved nested background subagent results to be saved in the parent subagent\u2019s transcript, so resumed subagents keep them and shared transcripts show the delivery\nChanged allowedMcpServers to govern only servers users add: a literal managed-mcp.json server your allowlist used to filter out now loads on upgrade; use deniedMcpServers to keep it off\n[VSCode] Added an Active quick filter and a status filter menu (Needs input, Working, Completed) to the session list sidebar\nFixed remote and scheduled sessions doing nothing after a connector-tool permission prompt was approved while the session was paused\n\n\nFixed Claude Code failing to launch on macOS 12 (Monterey), a regression introduced in 2.1.255\nFixed remote and scheduled sessions failing with \u201cuser messages must have non-empty content\u201d after a re-sent permission approval could not be applied\n\n\nAdded Claude Fable 5.1 (claude-fable-5-1), now the default Fable model \u2014 1M context, 10/10/50 per Mtok with $0.25/Mtok cache reads\nAdded \u201cTime format\u201d (timeFormat) and timeZone settings: 12-hour, 24-hour, 24-hour UTC, or a strftime pattern for the turn-end clock and transcript-view timestamps\nAdded a Containment Escape rule to auto mode so cloud metadata-credential fetches, egress evasion, and cross-tenant reach are no longer auto-approved unless your environment marks them expected\nAdded CLAUDE_CODE_SUBAGENT_MODEL_FORCE to apply CLAUDE_CODE_SUBAGENT_MODEL (or the main model) to every subagent, ignoring per-spawn and agent-definition model overrides\nAdded s in /effort to change effort for the current session only, matching /model\nAdded a /doctor warning for stale sandbox mask files left by a killed session\nAdded a one-time prompt in auto mode before the first file read outside the working directories, with the option to block such reads (permissions.blockReadsOutsideWorkingDirectories)\nAdded support for a gateway-supplied description on discovered /model picker entries (CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY); entries without one still read \u201cFrom gateway\u201d\nFixed settings in a .claude/ folder created after startup not being picked up until restart\nFixed sessions dispatched from an agent view opened with \u2190 always starting in the original session\u2019s permission mode, overriding the target directory\u2019s defaultMode and the agent\u2019s permissionMode\nFixed keybindings.json rebinds of Ctrl+G being ignored in claude agents; its Ctrl+S / Ctrl+T are now rebindable via the new Agents context\nFixed background sessions failing to start on macOS npm installs during a self-update, and on Windows when a stale daemon lock file pointed at a reused process id\nFixed the working spinner stopping while a response streams behind a slash-command panel\nFixed a background session\u2019s state.json detail repeating its own dispatch prompt after a scheduled wake-up\nFixed claude agents keeping a background session you re-prompted buried in Completed after it finished again; Completed now orders by the latest finish\nFixed claude --bg from a directory that was just deleted reporting \u201cbackgrounded\u201d and leaving a crashed session row; it now prints the reason and exits 1\nFixed Remote Control connecting mid-session re-sending the Bash tool definition, causing a prompt-cache miss\nFixed a doubly-listed custom Authorization header overriding the configured credential on Bedrock, Mantle, Vertex, and WIF, and the Vertex setup wizard picking up a leftover Anthropic profile from ~/.config/anthropic\nFixed Claude apps gateway sending stray host Authorization or profile headers to Foundry, Vertex, and Bedrock, and Foundry Entra ID upstreams not starting when ANTHROPIC_FOUNDRY_API_KEY is set\nFixed a leftover Anthropic API key or auth token being sent alongside your Foundry subscription key in API-key mode\nFixed /schedule routines whose prompt was saved without a message role and then ran with nothing to do\nFixed claude agents not saying that a background session is waiting for you to approve a message from another session, or who sent it\nFixed a prompt stashed with Ctrl+S inside an opened background session being lost when the session went idle or was stopped and then reopened\nFixed telemetry (OTEL) settings pushed through server-managed settings being ignored on warm starts, including desktop-app Code sessions\nFixed a teammate permission request being answered twice when the leader\u2019s mailbox write was briefly locked\nFixed a phantom duplicate slash-command row rendering below the in-flight turn while a command\u2019s auto-continued response streamed\nFixed policyHelper timeoutMs and refreshIntervalMs values above the timer maximum (2147483647) causing failures or re-runs every millisecond; they are now clamped\nFixed the token counter freezing or crawling after switching to another subagent\u2019s transcript, and made background subagents\u2019 and teammates\u2019 counters update live while a response streams\nFixed sandbox network hosts written with a trailing dot (example.com.): a deniedDomains entry didn\u2019t block the host inside the sandbox, and \u201cdon\u2019t ask again\u201d for such a host kept prompting\nFixed dismissing the Remote Control consent prompt (Esc, or n at claude remote-control) counting as consent, so the next request connected without asking\nFixed /mcp reconnect and enable still connecting a settings-file MCP server that a managed MCP allow/deny list or strictPluginOnlyCustomization loaded after startup should block\nFixed claude mcp remove leaving a remote server\u2019s stored OAuth credentials behind when strictPluginOnlyCustomization locks MCP to plugin-only servers\nFixed Remote Control (claude remote-control) sessions started from the Claude app ignoring the selected model and running on the machine\u2019s default instead\nFixed --disallowedTools and session deny rules being dropped after the first settings reload when allowManagedPermissionRulesOnly is enabled\nFixed --resume listing a backgrounded conversation twice and --continue reopening its stalled pre-background copy; --continue now also opens finished background sessions\nFixed fullscreen mode not letting you click ! shell command output to expand it\nFixed background sessions left running an older Claude Code binary piling up across auto-updates instead of being retired\nFixed claude agents --json briefly switching the terminal to raw mode and undoing another program\u2019s terminal settings on exit\nFixed Proactive output style sessions busy-looping with filler messages and repeated log reads instead of idling while a background command or Monitor they started is still running\nFixed subagents stopping when a response was cut off mid-stream by a computer sleep, dropped connection, or server error; they now automatically continue instead of ending with an incomplete response\nFixed \u2190 doing nothing in the /btw panel inside a claude agents session: it now returns to the agents list (even mid-answer), and the panel comes back when you reopen the session\nFixed sessions with an advisor model set missing the prompt cache on background requests (compaction, /recap, prompt suggestions) and re-sending the full conversation uncached each time\nFixed claude -p exiting about 5 seconds after its final result while a Monitor the model armed was still running; it now waits for the watch to fire or time out\nFixed a permissions.ask rule being skipped in auto mode when the matching command ran inside a compound command or subshell, letting it run without the confirmation prompt\nFixed plugins being able to read files outside their own directory through a declared command, agent, skill, hooks or other component path that is a symlink; such paths are now refused with an error\nFixed /add-dir rejecting a directory inside the current working directory; it now loads that directory\u2019s skills, commands, and agents like --add-dir does at startup\nFixed the main agent not being told when you resume a subagent you had stopped from its transcript view\nFixed a crash when pasting ANSI-colored text (e.g. a CI log) into dialogs like /feedback\nFixed claude mcp add/remove hanging or exhausting memory when the project\u2019s .mcp.json is a FIFO or a device-file symlink; it now fails fast with an actionable message\nFixed unbounded memory growth when non-JSONL data is piped into claude -p --input-format stream-json; it now fails fast with a clear error\nFixed backgrounding a turn (\u2190 or Ctrl+B) while a subagent or other tool was running occasionally making the background session treat that tool as rejected instead of re-running it\nFixed Bash Read()/Edit() deny rules not applying to < file redirects and reader commands like tac and egrep; a deny rule on any argument or redirect target now refuses the command\nFixed resuming or messaging a subagent whose transcript had grown past 5 MB (for example after reading many images) failing with \u201cNo transcript found\u201d\nFixed worktree-isolated sessions refusing Bash loops, $VAR reads, \"$(\u2026)\" and heredocs that never touch git as \u201ctoo complex to verify that it stays inside the worktree\u201d\nFixed /model and /effort showing a prompt-cache warning after rewinding a conversation back to empty\nFixed prompt-cache misses on every turn in long screenshot-heavy sessions once images exceeded the per-request size cap\nFixed the Edit permission prompt\u2019s diff view rendering emoji and multi-code-point characters with incorrect widths\nFixed WebSocket MCP server connection failures being logged as \u201c[object ErrorEvent]\u201d instead of the underlying error\nFixed background sessions failing to open with \u201cCouldn\u2019t start the background service\u201d while another Claude Code process was downloading an npm update; the start now waits for it\nFixed background commands that detach from their shell (for example under timeout or setsid) surviving a task stop or Claude Code exit\nFixed Claude not being told when you stop a background command from the tasks panel or a connected client\nFixed stopping a background subagent leaving its monitors running\nFixed sandboxed git commands in a linked worktree losing write access to the repository\u2019s common .git directory after cd into a subdirectory\nFixed Bedrock and Bedrock Mantle requests going silent during long hidden-thinking phases on Opus 4.7 and later, which let idle timeouts cut the connection; the stream now carries progress events\nFixed launching Claude Code after a Claude apps gateway expired or revoked your session: it now says the session ended and offers /login instead of reporting a network error\nFixed cloud sessions losing git/GitHub credentials for the rest of the session when the session\u2019s network proxy failed to start at launch; it now retries in the background and recovers\nFixed leftover cc-daemon-* folders in the system temp directory after an interrupted background daemon start; the cleanupPeriodDays retention sweep now removes them\nFixed Bash permission checks auto-approving certain [[ ]] conditionals that zsh parses differently from bash; these commands now prompt for approval\nFixed the managed-settings approval prompt showing the generic warning instead of its telemetry wording when the settings also turn detailed tracing or raw API body logging off, or trace export on\nFixed agent-team teammates in tmux/iTerm2 panes sometimes staying open after acknowledging a shutdown request\nFixed the keyless Console sign-in (\u201cSign in with your Console account\u201d) not applying your organization\u2019s server-managed settings, and /status not showing the Organization for that sign-in\nImproved rendering performance: less re-render work per turn in long conversations, streaming no longer slows down as the reply grows, and background-agent updates no longer re-render the whole screen\nImproved prompt input responsiveness by reducing per-keystroke rendering work\nImproved policy helper diagnostics \u2014 refresh failures now show in /status, declining the managed-settings dialog prints why Claude Code exited, and helper timeouts are reported as timeouts\nImproved /code-review --comment to post findings on GitLab merge requests via glab mr note instead of reporting the target as unsupported\nImproved notifications: an MCP elicitation or permission ask queued under another dialog now sends its idle desktop notification at the same delay as a visible ask\nImproved verbose/transcript output: async hook completion notices that arrive together now appear on one line instead of one line per hook\nImproved claude self-hosted-runner --configure-git to also enable git push negotiation, so the first push of a new branch from a stale clone uploads only the new commits instead of the whole tree\nImproved liveness reporting to SDK hosts while a response is held open by gateway keep-alives, so long waits under a raised CLAUDE_STREAM_IDLE_TIMEOUT_MS are not mistaken for a hung session\nImproved MCP connection and OAuth debug/error logs so credentials carried in a server\u2019s URL or request headers are redacted\nImproved /fork to keep the original conversation\u2019s prompt cache in the new background session: its worktree briefing now arrives as a message instead of a system-prompt change\nImproved emoji autocomplete to accept the remaining GitHub/Slack shortcode aliases (:satisfied:, :telephone:, :collision:, \u2026)\nChanged --effort to lift a new model\u2019s default-effort hold for that session only rather than permanently; an effort picked on claude.ai for a Remote Control session now applies during the hold\nChanged a policyHelper in MDM or managed-settings.json shadowed at launch by cached server-managed settings to run (or exit) as soon as the fetch reports them removed, not at the next launch\nChanged managedSourcesBehavior: \"merge\" to take sandbox.credentials.awsPairs and sandbox.ripgrep whole from the highest managed source that sets them instead of combining the sources\u2019 values\nChanged gateway model discovery (CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY=1) to run even when CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC is set, since it only queries your gateway\nChanged claude --resume --bg to continue that session under its own ID when nothing is running it, instead of silently starting a copy; a copy is now announced\nChanged /btw history browsing from \u2190/\u2192 to Shift+\u2190/Shift+\u2192 (or [/]), stepping through your recent side questions and back to the live answer\nChanged defaultMode: \"bypassPermissions\" in .claude/settings.json or .claude/settings.local.json to be ignored, like \"auto\"; set it in user or managed settings, or pass --permission-mode\nChanged fable and best in Claude apps gateway sessions to keep resolving to Fable 5 for now, since gateways not yet configured for Fable 5.1 reject it; pick Fable 5.1 in /model to use it\nChanged --add-dir, /add-dir, and additionalDirectories to refuse network paths (UNC shares, /net/ automounts) with a message before touching them; on Windows use a mapped drive letter\nChanged Claude apps gateway sign-in and token refresh requests to verify the gateway\u2019s pinned TLS certificate, as the managed settings fetch already does\nChanged Cowork and claude.ai cloud sessions: reading an artifact that isn\u2019t yours now always asks you first, even in auto mode\nRemoved the Ctrl+E command explanation on Bash and PowerShell permission prompts\n[VSCode] Added collapsible ACCOUNT & USAGE and SESSION MANAGER section headers to the session list panel, with the account email, the usage meter, and a View details link opening the usage dialog\n[VSCode] Added a model pill to the input footer that shows the current model and opens the model picker, with an Effort row and a \u201cMore models\u201d page\n[VSCode] Added a collapse toggle to the Ungrouped section of the session list\n[VSCode] Added output style selection to the command menu, including custom styles\n[VSCode] Fixed third-party provider deployments (Bedrock, Vertex, and others) still showing claude.ai-only features (remote sessions, dictation, usage) and calling claude.ai with a leftover login\n[VSCode] Fixed the session list panel\u2019s usage meter staying blank after the panel loads; it now shows the last known usage immediately\n[VSCode] Fixed the \u201cEnable Remote Control for all sessions\u201d toggle so turning it on or off applies to sessions that are already open, not only to new ones\n[VSCode] Fixed screen reader announcements: a control character before a fence or heading no longer drops visible lines from speech, and bold markers spanning a heading are no longer mis-paired\n[VSCode] Changed the action menu to list slash commands in a filterable \u201cSlash commands\u201d dialog instead of inline; picking one runs it; the MCP servers dialog gained the same filter box\n[VSCode] Changed \u201cDelete session\u201d to \u201cArchive session\u201d: archived sessions move to a collapsible \u201cArchived sessions\u201d group at the bottom of the list with an Unarchive action\n\n\nFixed Bash commands failing with \u201ctask output swap refused (tasks dir moved or linked)\u201d on some Macs\nFixed \u201calways allow\u201d not saving in a project that has no .claude/settings.local.json yet\nFixed Remote Control sessions hosted by Claude Desktop or VS Code stalling for minutes after a tool finished when the connection to claude.ai was degraded\nFixed background task notifications with very large failure output (for example git errors on a full disk) making the conversation exceed the API request size limit\n\n\nAdded PreModelSwitch and PostModelSwitch hook events (block, confirm, or annotate a model switch); SessionStart resume hooks now receive session staleness and the estimated re-cache cost\nAdded live streaming of a foreground subagent\u2019s tool calls and results to Remote Control clients (background subagents, the default, still show status only)\nAdded a Spend limit bar to /usage and a rate_limits.spend_limit status line field for developers behind a Claude apps gateway with spend limits\nAdded a per-session prompt-cache line to /cost (hit ratio, misses, tokens re-cached, warm/cold) and a matching prompt_cache object for status line scripts\nAdded attach, logs, stop, respawn, and rm to claude --help; the --resume message for a running background session now names the exact claude attach command\nFixed file tools (Read, Write, Edit) following a symlink swapped inside the working directory after the permission check, which could read or write outside the approved location\nFixed plugin commands declared in a marketplace entry being able to point outside the plugin directory; such paths are now rejected with a path-traversal error\nFixed project settings being able to enable detailed beta tracing or raw API body logging, and a lower-scope beta tracing endpoint bypassing an OTLP collector pinned by managed settings or a host app\nFixed the Workflow tool reading (and quoting in errors) a scriptPath outside what the session may read before the permission check ran\nFixed Grep and Glob not applying Read(...) deny rules to files reached through a symlinked search path\nFixed conversations getting stuck on \u201ctext content blocks must be non-empty\u201d errors after a turn where the model produced only thinking\nFixed the first launch on a fresh install starting in default mode instead of auto mode for accounts whose startup default is auto mode\nFixed Opus 5 requests failing with \u201ceffort \u2026 is not supported when thinking is disabled\u201d when effort was xhigh/max and thinking was turned off; effort is now sent as high in that case\nFixed replying to a message Claude Desktop delivered from another session: SendMessage to that session id now delivers through Claude Desktop instead of failing with \u201cnot reachable\u201d\nFixed TUI lag with many parallel subagents: per-second progress ticks now replace their predecessor instead of piling up in the transcript\nFixed agent teams: a teammate\u2019s final answer not reaching the team lead \u2014 it now arrives in the idle notification instead of a content-free \u201cavailable\u201d notice\nFixed background subagents being unable to reply to a message from an unnamed sibling or parent agent (from was the agent type, which is not an address)\nFixed managed-settings disableAutoMode arriving mid-session not moving an already-running auto-mode session back to default mode\nFixed a \u201cswitch to Opus 1M for 5x more context\u201d tip that appeared even when the current Opus model already has a 1M context window\nFixed Claude apps gateway sessions treating a stored Anthropic profile (e.g. a Console sign-in) as active: listing it in /status and retrying gateway 401s with it, though requests never use it\nFixed cloud sessions telling Claude the model had changed when the host was only setting the session\u2019s initial model\nFixed Remote Control reporting a failure when an organization\u2019s policy disables it; it now shows a single quiet notice instead\nFixed /mcp reconnect on Remote Control showing a generic withheld-detail error instead of the real remedy when a server was disabled in another session\nFixed --input-format stream-json: client-injected assistant tool calls sent without a message id were merged into the first one and their results lost, including when resuming older sessions\nFixed session transcripts being silently overwritten when a directory change relocated a session onto an existing same-ID transcript\nFixed background sessions and their subagents being unable to edit files inside a git worktree they created with git worktree add\nFixed background sessions occasionally starting without any plugin skills (and staying that way) when another Claude Code process was refreshing the plugin marketplace at the same moment\nFixed selecting text in an opened background session inside tmux over SSH: it now copies to the tmux buffer like a foreground session instead of falling back to OSC 52\nFixed SDK and cloud sessions hanging indefinitely when an SDK MCP server\u2019s handshake acknowledgment was lost; the wait now times out after 70 seconds and marks only that server failed\nFixed self-hosted runner leaving a stuck session\u2019s Bash tool processes running after the session was force-stopped\nFixed /usage-credits for Team and Enterprise members whose admin set the org\u2019s usage-credit limit to $0: it now offers to ask the admin instead of saying a cap was reached\nFixed --worktree --tmux with a merge-request number on a gitlab.com origin trying a doomed GitHub-style fetch first instead of fetching the GitLab ref directly\nFixed Ctrl+G failing with \u201cEmacs quit unexpectedly\u201d in background sessions for editors that open /dev/tty, such as emacs -nw and micro\nFixed an additionalDirectories entry containing a null byte crashing startup, or breaking /add-dir and later settings updates when it came from an SDK host, IDE, or hook; it is now skipped\nFixed the MCP server menu\u2019s copy shortcut: it now says how the sign-in URL was copied instead of always claiming success\nFixed italic text (such as the session recap line) rendering as highlighted blocks in GNU screen and in tmux sessions using a screen terminal type\nFixed claude mcp add --header and claude mcp add-json help text naming the wrong transports\nFixed claude ultrareview and /ultrareview waiting the full 30 minutes when the cloud session fails to start; they now stop early and report the reason\nFixed Bash permission checks auto-approving commands that assign an arithmetic expression to an integer shell variable (e.g. OPTIND=1/0, RANDOM=2+2); these now prompt for approval\nFixed backgrounded sessions (\u2190, /background, --bg) losing a Vertex/Bedrock gateway (ANTHROPIC_*_BASE_URL + CLAUDE_CODE_SKIP_*_AUTH) exported in the shell, so every request failed\nFixed claude --bg --model fable on Max plans stopping to ask for usage credits while the interactive session on the same account still had Fable allowance\nFixed the one-time \u201cmake auto mode your default\u201d offer appearing in unattended sessions (e.g. agent-team teammate panes), where a stray keypress could accept it unread\nFixed the managed-settings approval prompt re-appearing after signing in again to the same Claude apps gateway when the settings are unchanged\nFixed disabled /bug and /share reporting that /feedback was disabled; tips, /help, and refusal messages no longer suggest /feedback when an org policy or env var turns it off\nFixed cloud session creation advising GitHub setup after a transient GitHub connection failure \u2014 the message now says to retry instead\nImproved CPU usage during turns in interactive sessions by cutting redundant UI re-renders\nImproved install size: the native binary is about 5 MB smaller\nImproved cloud sessions: when the session\u2019s network proxy drops a connection during a Bash command, the tool result now names the host and reason instead of only \u201cconnection reset\u201d\nImproved /schedule to explain that MCP servers configured in Claude Code can\u2019t be attached to cloud routines, instead of a bare \u201cNo MCP connectors\u201d message\nImproved framing of messages from your own subagents: Claude is told the sender is a worker inside this session, not an unrelated Claude session\nImproved the prompt placeholder to read \u201cMessage @name\u2026\u201d while viewing a background subagent or fork transcript opened from the subagent panel or /tasks\nImproved sanitization of MCP server names in error messages, menus, and command results\nImproved Amazon Bedrock session start under CLAUDE_CODE_PROVIDER_MANAGED_BY_HOST (e.g. Claude Desktop): a session given a Bedrock model ID or ARN no longer waits for inference-profile discovery\nImproved the managed settings approval dialog to list only the settings that changed since you last approved them\nImproved retry when the model\u2019s tool call is malformed: the broken output is now dropped from the retry context, including on Bedrock, Vertex, and Foundry\nChanged /radio to be available on Bedrock, Vertex AI, Foundry, and Claude Platform on AWS, and when telemetry is disabled\nChanged Claude in Chrome so browser actions always go through Claude Code\u2019s permission checks, including in sessions with telemetry disabled, which previously used the Chrome extension\u2019s own prompts\nChanged CLAUDE_CODE_SUBAGENT_MODEL to set the default subagent model rather than override everything: an agent definition\u2019s model: and an explicit per-spawn model now take precedence over it\nChanged the default commit trailer to Co-Authored-By: Claude Code when the active model isn\u2019t a recognized Claude model (e.g. third-party models behind a custom ANTHROPIC_BASE_URL)\nChanged the default model for seat-based Enterprise subscriptions to Opus 5, matching other premium plans\nChanged /effort to save your default effort level per model, so each model keeps its own setting when you switch\nChanged analytics to no longer turn off before sign-in solely because managed settings force gateway login (or cannot be read); they stay off once signed in to the gateway or via DISABLE_TELEMETRY\nChanged the footer PR badge on Bedrock, Vertex, and Foundry, and when telemetry is off, to call the GitHub API directly (via gh auth token, GH_TOKEN, or GITHUB_TOKEN) instead of gh pr view\nChanged how Bash command output files are created and read back when commands run in the sandbox, so a sandboxed command cannot redirect or replace them\nChanged plugin/LSP install suggestions and the auto-mode default offer to wait until you\u2019ve sent or cleared what you\u2019re typing, so the Enter that sends your prompt can\u2019t answer them\nChanged server-managed settings that terminate sandbox TLS, route sandbox traffic through your own proxy, inject credentials, or weaken sandbox isolation to require approval before they apply\nChanged ANTHROPIC_CUSTOM_HEADERS from managed or project settings to require approval when it sets a credential, org/tenant, routing, or API-behavior header (e.g. Authorization, Host)\nChanged project-level .claude/settings.json env to no longer set CLAUDE_CONFIG_DIR, CLAUDE_CODE_TMPDIR, or TMPDIR/TMP/TEMP; set them in your shell, user, or managed settings instead\nRemoved syntax highlighting for six rarely used languages (1c, gml, isbl, mathematica, maxima, sqf); the binary is 2.5 MB smaller\n[VSCode] Fixed the sign-in screen\u2019s \u201cBedrock, Foundry, or Vertex\u201d button opening the docs at the top of the page instead of the third-party provider setup section\n[VSCode] Changed the Remote Control banner to a footer pill (shown while Remote Control is on or has failed) that opens the session on claude.ai/code; turn it on or off with /remote-control\n\n\nBug fixes and reliability improvements\n\n\nAdded --restricted (or CLAUDE_CODE_RESTRICTED=1): removes the built-in tools that run commands or code and WebFetch (unless named in --tools), keeps file tools inside the working directory, refuses bypassPermissions, and ignores user, project and local settings files\nAdded experimental.cacheTtl (\"5m\" or \"1h\") to agent frontmatter: a per-agent prompt cache TTL used when no subagent TTL setting is configured\nAdded claude self-hosted-runner --client-label