Claude Code 2.1.282

What changed

Claude Code 2.1.282 shipped on September 24, 2026. New features cover prose width, a check for ignored telemetry variables, and new managed and Claude apps gateway settings. Most of the fixes again target resume: more cases where continued or resumed conversations, or ones where a slash command was used mid-turn, dropped Claude’s earlier extended thinking. On the security side, project settings can no longer turn on telemetry export or borrow the anthropic-skills and claude-ai names.

New features

• maxProseWidth setting

  Caps the width of Claude’s prose in wide terminals, while tables and code blocks keep the full width.

• Ignored telemetry variable notice

  A startup notice, plus /status and claude doctor entries, list telemetry variables in a project’s settings files that were ignored or that turned telemetry off.

• allowClaudeInChromeWithManagedMcp managed setting

  Lets claude --chrome run alongside an exclusive managed-mcp.json. The error shown when Chrome is blocked now names this setting.

• store.readiness_grace_seconds in the Claude apps gateway

  Lets /readyz stay ready through a short Postgres outage such as a database failover.

• Scrollbar in the /feedback drafts list

  In fullscreen mode, the drafts list gets a scrollbar that appears while the mouse is over the list.

• [Cloud sessions] Repository additions

  • Claude GitHub App status in Settings › Connectors › GitHub: whether the app is installed and reachable for your account, plus steps to connect, install or reconnect
  • “Open repository” and “Open compare page” links in the repository menu of a cloud session whose repository is hosted on a Git server other than GitHub
  • Attaching a repository from a different GitHub owner, such as a fork’s upstream, to a running cloud session that already has one, including sessions started from Slack

Key improvements

• Resuming very large sessions

  Very large sessions resume faster, including ones that were never compacted.

• Windows resume error (EBADF)

  When a session can’t be resumed because its transcript file could not be read, the error now names possible causes and what to try.

• Claude Desktop unknown-model error

  Now suggests switching to a different model.

• Unusual Unicode in permission prompts

  Rendered more accurately.

• /artifacts list

  Titles line up in one column, details are dropped whole instead of cut mid-word, and the list supports PgUp/PgDn, Home/End, the mouse wheel and clicks.

• claude-api skill updates

  Pre-output refusal billing now links to the How refusals are billed docs, mid-stream refusals bill at normal rates, and pre-output refusals count against rate limits. The skill also recommends ant apply for keeping Managed Agents resources as version-controlled files.

• [Cloud sessions] Routines load time

  The Routines page and the sidebar’s Scheduled list load faster for accounts whose past sessions scheduled many check-in reminders.

Changed behavior

• Server-side classifier by default with telemetry off (changed)

  On a direct Anthropic API connection with telemetry off, auto mode now uses the server-side classifier by default. CLAUDE_CODE_AUTO_MODE_SERVER=0 opts out.

• OpenTelemetry variables in project settings (changed)

  Project and local settings now ignore OpenTelemetry variables that turn on export, set its endpoint, or capture content, like CLAUDE_CODE_ENABLE_TELEMETRY and OTEL_LOG_*.

• sandbox.excludedCommands (changed)

  Project and local settings entries are ignored when managed settings or --settings set allowUnsandboxedCommands: false, or managed settings set allowManagedDomainsOnly: true.

• Windows/WSL managed settings (changed)

  An admin policy that is present but invalid or unreadable (HKLM, managed-settings.json) keeps user-writable HKCU and WSL /etc/claude-code from applying.

• anthropic-skills and claude-ai namespaces (changed)

  • Skill(anthropic-skills:*) and Skill(claude-ai:*) allow rules cover only skills synced from claude.ai, not plugins or other skills that merely use such a name.
  • Skill folders, command files and workflow commands in these namespaces no longer load. A plugin so named still loads but yields name ties to synced skills.
  • MCP servers configured under these names list no skills or prompts (their tools still work). Rename the server in your MCP configuration to list them again.

• ultracode visuals (changed)

  /effort and the prompt input use plain styling (no ripple, border flourish or keyword glimmer), and the dynamic-workflows spinner tip is removed.

• Clawd in the start-up banner (changed)

  The mascot’s feet now sit under the corners of his body.

• [Claude Tag] Card width and default model (changed)

  The bordered cards Claude uses in Slack replies for plans, tables and details render wide by default. Newly connected Slack workspaces follow the current default model instead of keeping whichever model was the default when they were connected.

Bug fixes

Resume & extended thinking

• Every request failing with a 400 error in conversations whose history holds web search results the API cannot decrypt (for example, from a turn answered through a third-party gateway).
• More cases of continued or resumed sessions (--continue, --resume) re-sending earlier messages in a changed form, which could make the API drop Claude’s earlier reasoning.
• Earlier extended thinking being dropped when /model, /rename, /artifacts or another immediate slash command was used while Claude was working.
• Continued or resumed conversations losing earlier extended thinking when relaunched with a --tools list that leaves out a built-in tool offered earlier in the conversation.
• Sessions failing on every turn with an “Invalid data in redacted_thinking block” API error; Claude Code now drops the conversation’s thinking blocks and retries once.
• Compaction failing when the summarization request is refused; it now retries on a fallback model.
• A failed turn (“Effort ‘xhigh’ isn’t available with thinking turned off”) after a safety-related model switch in sessions with thinking off and effort above high.

Models & usage credits

• An unanswered Fable usage-credits prompt switching models in SDK-hosted sessions such as Claude Desktop; the turn now ends instead, and Remote Control clients now see the model-switch notice.
• /model with a full Fable model id stopping at an API error instead of opening the usage-credits prompt when the plan needs usage credits that aren’t turned on yet.

Login & org policy

• Requests failing for up to a minute with an “another Claude Code process is refreshing it” login error after that other process was closed or killed mid-refresh.
• Sessions started while another Claude Code window was refreshing the sign-in (common with several VS Code windows) not retrying their organization policy fetch.

Permissions & managed settings

• CLAUDE.md and rules being read at startup through a repository symlink reaching macOS’s /Network via .. or a /.vol-style kernel path, or a rules link to macOS’s /home being listed.
• Bash permission rules with a mid-pattern :* being skipped in settings files while --allowedTools honored them; they now work from every source, with a startup warning on how they match.
• A command approved on a restored permission prompt running twice when a remote session’s worker restarted.
• Managed settings ignoring a mistyped value for boolean lock keys such as disableClaudeAiConnectors or allowManagedPermissionRulesOnly; the lock now applies and startup names the key.
• Managed permissions, autoMode, worktree and attribution settings being ignored entirely when one nested value was invalid; the rest of the block now still applies.
• Repository, user and --add-dir skills, commands and skills-directory plugin manifests pre-approving their own tools via allowed-tools under managed allowManagedPermissionRulesOnly.

Third-party providers

• Safeguard block messages on Amazon Bedrock and Bedrock Mantle not showing a request ID; block messages now also show the message ID.
• Vertex AI: web search not being offered for models Claude Code doesn’t recognize yet, such as newly released ones.

Tool calls

• Bash and PowerShell hiding a full disk quota behind “Exit code 1” and leaving large output files in temp.
• Tool input validation errors naming only an unknown, missing or mistyped parameter when other parameters in the same call were also invalid; those are now listed too.
• PDF page-read error messages: paths with accented or non-Latin characters now appear readably, and a folder named like “password” or “invalid” can no longer make the error name the wrong cause.

Rendering & input

• Pasted multi-line text being submitted line by line after the terminal’s bracketed paste mode was reset mid-session.
• The prompt’s example text flashing and disappearing at startup in projects with a SessionStart hook, and a blank screen flashing before the first frame when starting in fullscreen mode.
• Garbled, misplaced rows in the non-fullscreen renderer after the screen got shorter while still taller than the terminal, e.g. deleting a prompt line while a shell command streams output.
• A stale character left in the last column of a diff when a redrawn line’s CJK character or emoji wrapped to the next row.
• The cursor landing before the end of a prompt recalled from history when the prompt contains a tab.
• The send-now hint showing ctrl+enter on terminals that send it as a newline (Windows Terminal before 1.25); it now shows ctrl+x ctrl+s there.
• Screen-reader mode, quoted lists and very long lists dropping the blank lines at the top of a code block that opens a list item, directly or inside a quote.

Commands & plugins

• claude remote-control --debug failing with “Unknown argument: —debug”, although Remote Control’s own eligibility error says to run with --debug.
• /install-github-app saying “cancelled” and then still pushing the branch and saving the API key secret; leaving now stops the remaining steps and reports what was already done.
• /feedback, /bug and /share on Bedrock, Vertex and other third-party providers still saving the report file after you cancelled during the save.
• Plugin uninstall reporting success and deleting the plugin’s saved options when its settings file still enabled it or could not be read; it now stops and names the file.
• Plugin uninstall deleting a plugin’s saved options and secrets when the list of installed plugins could not be read after the removal; they are now kept and the uninstall says so.
• A key typed right after / in /skills moving the skill list instead of reaching the search box, and the terminal cursor jumping from the /skills search box to the skill list while typing, which could hide the caret and put IME input in the wrong place.
• Lists with a scrollbar, such as /skills and /mcp, being two columns narrower outside fullscreen mode, where the scrollbar can never appear.
• The agent panel footer wrapping onto two lines with long rebound keys, and its “Esc to collapse” hint ignoring a rebound collapse key.
• A doubled · separator in the /tasks dialog footer when the stop-all-agents shortcut is unbound in keybindings.json.
• Artifact publishes failing when Claude gave the version a label longer than 60 characters; the label is now shortened.

Vim mode

• >> indenting empty lines, r with a count longer than the line changing text, 2J joining one line too many, and a count on the last line (2dd, 2>>) shifting or deleting it.
• Cursor placement: after dd, dj, dG or a whole-line p/P it lands on the first non-blank, yy no longer moves it, and Esc after an emoji no longer leaves it inside the emoji.
• A count typed before . being ignored when repeating x, s, p, d or c, and whole-line p/P, o, O, J, >> and << acting on the wrong line when a line above wraps.
• The cursor left past the end of a prompt recalled from history or pulled back from the queue in normal mode, so x did nothing.

[VSCode]

• Long replies falling behind the stream: the panel no longer re-parses the whole reply on every update.
• The dictation mic button covering the message input’s scrollbar when the input is tall enough to scroll.
• Remote Control sessions started on this computer not opening from their Web entry in the session list; they now open the local conversation unless it’s running elsewhere.
• An editor tab’s sign-in screen hanging silently after the extension host restarts; it now shows the “stopped responding” notice too.

[Cloud sessions]

• The next run time shown for an hourly routine being 30 minutes off for people in half-hour-offset time zones such as India.

[Claude Tag]

• Auto-join channel patterns saved for one workspace being ignored on an Enterprise Grid org-wide install; Claude now joins matching new channels.
• Claude not responding in an Enterprise Grid channel shared between two workspaces of one organization when the channel’s Claude Tag version was saved from the other workspace.
• The earlier Claude in Slack app’s progress card for sessions on a GitHub Enterprise Server repository: it now names the repository and offers a working Create PR button.
• Slack threads whose model has been retired falling back to another model on every reply, slower and with a fallback note each time; the thread now moves to a working model.
• Files Claude uploads to Slack not being able to carry a caption containing a table; captions now render with the same formatting as replies.
• Claude’s threads in Slack’s Agents & tools view sometimes being listed under their first message instead of their name; later renames now update the list too.
• Removing a GitHub organization’s grant from an access bundle’s Repositories tab failing to save after that GitHub organization was disconnected.
• Claude always replying “Couldn’t check this channel just now” in a channel shared with a Grid workspace it isn’t added to; the notice now says which workspace needs the app.
• The cost and token totals in Claude’s reply footer reading many times too high after the session’s cloud worker restarted.

Notes

• Telemetry variables in project settings no longer apply — CLAUDE_CODE_ENABLE_TELEMETRY, OTEL_LOG_* and other export-related OpenTelemetry variables in project or local settings are ignored. The startup notice, /status and claude doctor list which ones were ignored.
• Skills and MCP servers named anthropic-skills or claude-ai need renaming — skill folders and command files in these namespaces no longer load, and MCP servers so named list no skills or prompts.
• Mid-pattern :* Bash rules in settings files now take effect — rules that were silently skipped now apply; check the startup warning to confirm they match what you intended.
• Auto mode uses the server-side classifier on direct API connections even with telemetry off — set CLAUDE_CODE_AUTO_MODE_SERVER=0 to opt out.
• An invalid Windows admin policy now blocks user settings — if HKLM or managed-settings.json is present but invalid or unreadable, HKCU and WSL /etc/claude-code settings don’t apply either.