Documentation 2026-11-10 21:57 UTC to 2026-11-11 22:59 UTC

220 220 

221> [!NOTE]221> [!NOTE]

222>222>

223> `effort` trades latency and token cost for reasoning depth within each response. [Extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) is a separate feature that produces `thinking` blocks in the output, and the `display` field on `ThinkingConfig` for [Python](https://code.claude.com/docs/en/agent-sdk/python.md#thinkingconfig) or [TypeScript](https://code.claude.com/docs/en/agent-sdk/typescript.md#thinkingconfig) controls whether you receive their text. They are independent: you can set `effort: "low"` with extended thinking enabled, or `effort: "max"` without it.223> `effort` trades latency and token cost for reasoning depth within each response. [Extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) produces `thinking` blocks in the output, and the `display` field on `ThinkingConfig` for [Python](https://code.claude.com/docs/en/agent-sdk/python.md#thinkingconfig) or [TypeScript](https://code.claude.com/docs/en/agent-sdk/typescript.md#thinkingconfig) controls whether you receive their text.

224 224 

225Use lower effort for agents doing simple, well-scoped tasks, such as listing files or running a single grep, to reduce cost and latency. Set `effort` in the top-level `query()` options for the [session](https://code.claude.com/docs/en/sub-agents.md#choose-an-effort-level), or per subagent with the `effort` field on [`AgentDefinition`](https://code.claude.com/docs/en/agent-sdk/subagents#agentdefinition-configuration) to override the session level.225Use lower effort for agents doing simple, well-scoped tasks, such as listing files or running a single grep, to reduce cost and latency. Set `effort` in the top-level `query()` options for the [session](https://code.claude.com/docs/en/sub-agents.md#choose-an-effort-level), or per subagent with the `effort` field on [`AgentDefinition`](https://code.claude.com/docs/en/agent-sdk/subagents#agentdefinition-configuration) to override the session level.

226 226 

1383| :- | :- | :- |1383| :- | :- | :- |

1384| `adaptive` | `type`, `display` | Claude adaptively decides when to think |1384| `adaptive` | `type`, `display` | Claude adaptively decides when to think |

1385| `enabled` | `type`, `budget_tokens`, `display` | Enable thinking with a specific token budget |1385| `enabled` | `type`, `budget_tokens`, `display` | Enable thinking with a specific token budget |

1386| `disabled` | `type` | Disable thinking |1386| `disabled` | `type` | Turn thinking off [where the model and provider allow it](https://code.claude.com/docs/en/model-config.md#where-you-cant-turn-thinking-off) |

1387 1387 

1388The optional `display` field controls whether thinking text is returned `"summarized"` or `"omitted"`. On Claude Opus 4.7 and later, the API default is `"omitted"`, so set `"summarized"` to receive thinking content in [`ThinkingBlock`](#thinkingblock) outputs. On providers such as Amazon Bedrock and Google Cloud's Agent Platform, Claude Code doesn't pass `display` for earlier models it recognizes. Those are Opus 4.6, Sonnet 4.6, Haiku 4.5, and earlier models, which use their default.1388The optional `display` field controls whether thinking text is returned `"summarized"` or `"omitted"`. On Claude Opus 4.7 and later, the API default is `"omitted"`, so set `"summarized"` to receive thinking content in [`ThinkingBlock`](#thinkingblock) outputs. On providers such as Amazon Bedrock and Google Cloud's Agent Platform, Claude Code doesn't pass `display` for earlier models it recognizes. Those are Opus 4.6, Sonnet 4.6, Haiku 4.5, and earlier models, which use their default.

1389 1389 

664| `rewindFiles(userMessageId, options?)` | Restores files to their state at the specified user message. Pass `{ dryRun: true }` to preview changes. Requires `enableFileCheckpointing: true`. See [File checkpointing](https://code.claude.com/docs/en/agent-sdk/file-checkpointing.md) |664| `rewindFiles(userMessageId, options?)` | Restores files to their state at the specified user message. Pass `{ dryRun: true }` to preview changes. Requires `enableFileCheckpointing: true`. See [File checkpointing](https://code.claude.com/docs/en/agent-sdk/file-checkpointing.md) |

665| `setPermissionMode()` | Changes the permission mode (only available in streaming input mode) |665| `setPermissionMode()` | Changes the permission mode (only available in streaming input mode) |

666| `setModel()` | Changes the model (only available in streaming input mode). Passing `undefined` or the string `"default"` resets to [Claude Code's default model](https://code.claude.com/docs/en/model-config.md). A session in `auto` mode [changes to `default` mode](https://code.claude.com/docs/en/agent-sdk/permissions#auto-mode-auto) if the new model doesn't support auto mode |666| `setModel()` | Changes the model (only available in streaming input mode). Passing `undefined` or the string `"default"` resets to [Claude Code's default model](https://code.claude.com/docs/en/model-config.md). A session in `auto` mode [changes to `default` mode](https://code.claude.com/docs/en/agent-sdk/permissions#auto-mode-auto) if the new model doesn't support auto mode |

667| `setMaxThinkingTokens()` | *Deprecated:* Use the `thinking` option instead. Changes the maximum thinking tokens. Passing `null` resets thinking to the session default: a mid-session override is cleared, and thinking stays off for sessions that have it disabled |667| `setMaxThinkingTokens()` | *Deprecated:* Use the `thinking` option instead. Changes the maximum thinking tokens. Pass `null` to undo earlier calls and restore the `thinking` or `maxThinkingTokens` option the session started with |

668| `applyFlagSettings(settings)` | Merges settings into the session's flag settings layer at runtime (only available in streaming input mode). See [`applyFlagSettings()`](#applyflagsettings) |668| `applyFlagSettings(settings)` | Merges settings into the session's flag settings layer at runtime (only available in streaming input mode). See [`applyFlagSettings()`](#applyflagsettings) |

669| `updateSettings(source, settings)` | Writes one allowlisted key to the project's local settings file or your user settings file, so the value persists for later sessions. See [`updateSettings()`](#updatesettings). Requires TypeScript SDK v0.3.257 or later, which bundles Claude Code v2.1.257 |669| `updateSettings(source, settings)` | Writes one allowlisted key to the project's local settings file or your user settings file, so the value persists for later sessions. See [`updateSettings()`](#updatesettings). Requires TypeScript SDK v0.3.257 or later, which bundles Claude Code v2.1.257 |

670| `initializationResult()` | Returns the full initialization result including supported commands, models, account info, and output style configuration |670| `initializationResult()` | Returns the full initialization result including supported commands, models, account info, and output style configuration |

5432type ThinkingConfig =5432type ThinkingConfig =

5433 | { type: "adaptive"; display?: ThinkingDisplay } // The model determines when and how much to reason (Opus 4.6+)5433 | { type: "adaptive"; display?: ThinkingDisplay } // The model determines when and how much to reason (Opus 4.6+)

5434 | { type: "enabled"; budgetTokens?: number; display?: ThinkingDisplay } // Fixed thinking token budget5434 | { type: "enabled"; budgetTokens?: number; display?: ThinkingDisplay } // Fixed thinking token budget

5435 | { type: "disabled" }; // No extended thinking5435 | { type: "disabled" }; // Turn thinking off where the model and provider allow it

5436```5436```

5437 5437 

5438For `disabled`, see [where you can't turn thinking off](https://code.claude.com/docs/en/model-config.md#where-you-cant-turn-thinking-off).

5439 

5438The optional `display` field controls whether thinking text is returned `"summarized"` or `"omitted"`. On Claude Opus 4.7 and later, the API default is `"omitted"`, so set `"summarized"` to receive thinking content in `thinking` blocks. On providers such as Amazon Bedrock and Google Cloud's Agent Platform, Claude Code doesn't pass `display` for earlier models it recognizes. Those are Opus 4.6, Sonnet 4.6, Haiku 4.5, and earlier models, which use their default.5440The optional `display` field controls whether thinking text is returned `"summarized"` or `"omitted"`. On Claude Opus 4.7 and later, the API default is `"omitted"`, so set `"summarized"` to receive thinking content in `thinking` blocks. On providers such as Amazon Bedrock and Google Cloud's Agent Platform, Claude Code doesn't pass `display` for earlier models it recognizes. Those are Opus 4.6, Sonnet 4.6, Haiku 4.5, and earlier models, which use their default.

5439 5441 

5440### `SpawnedProcess`5442### `SpawnedProcess`

128 128 

129### Browser tools in plan mode129### Browser tools in plan mode

130 130 

131In [plan mode](https://code.claude.com/docs/en/permission-modes.md#analyze-before-you-edit-with-plan-mode), a permission prompt appears before Claude records a GIF, opens a new tab, or runs a shortcut. In an interactive CLI session, if [bypass permissions mode is available](https://code.claude.com/docs/en/permission-modes.md#skip-all-checks-with-bypasspermissions-mode) and [feature-flag fetching](https://code.claude.com/docs/en/env-vars.md#features-that-need-feature-flag-fetching) is off, these calls run without a prompt.131In [plan mode](https://code.claude.com/docs/en/permission-modes.md#analyze-before-you-edit-with-plan-mode), a permission prompt appears before Claude records a GIF or opens a new tab. In an interactive CLI session, if [bypass permissions mode is available](https://code.claude.com/docs/en/permission-modes.md#skip-all-checks-with-bypasspermissions-mode) and [feature-flag fetching](https://code.claude.com/docs/en/env-vars.md#features-that-need-feature-flag-fetching) is off, these calls run without a prompt.

132 132 

133A `tabs_context_mcp` call also prompts when it sets `createIfEmpty`, and so does a `browser_batch` call that includes any of these actions.133A `tabs_context_mcp` call also prompts when it sets `createIfEmpty`, and so does a `browser_batch` call that includes any of these actions.

134 134 

146* **Auto memory**: the sweep doesn't delete the memory files in a project's [auto memory](https://code.claude.com/docs/en/memory.md#auto-memory) directory, `projects/<project>/memory/`. Claude Code removes that directory only if it has been empty for the whole retention period. Before v2.1.228, the sweep treated folders inside the memory directory as session data and could delete old files beneath it.146* **Auto memory**: the sweep doesn't delete the memory files in a project's [auto memory](https://code.claude.com/docs/en/memory.md#auto-memory) directory, `projects/<project>/memory/`. Claude Code removes that directory only if it has been empty for the whole retention period. Before v2.1.228, the sweep treated folders inside the memory directory as session data and could delete old files beneath it.

147* **Claude Desktop and Cowork transcripts**: Claude Code keeps the transcript of a session you started or most recently continued in Claude Desktop or Cowork at any age. To give these transcripts an age limit, set [`desktopSessionCleanupPeriodDays`](https://code.claude.com/docs/en/settings-reference#desktopsessioncleanupperioddays). Requires Claude Code v2.1.248 or later; earlier versions delete them after `cleanupPeriodDays`.147* **Claude Desktop and Cowork transcripts**: Claude Code keeps the transcript of a session you started or most recently continued in Claude Desktop or Cowork at any age. To give these transcripts an age limit, set [`desktopSessionCleanupPeriodDays`](https://code.claude.com/docs/en/settings-reference#desktopsessioncleanupperioddays). Requires Claude Code v2.1.248 or later; earlier versions delete them after `cleanupPeriodDays`.

148 148 

149 Claude Code deletes these transcripts after `cleanupPeriodDays` instead in either of these cases:149 Claude Code deletes these transcripts after `cleanupPeriodDays` instead in cases such as these:

150 150 

151 * [Managed settings](https://code.claude.com/docs/en/managed-settings.md) set `cleanupPeriodDays`151 * [Managed settings](https://code.claude.com/docs/en/managed-settings.md) set `cleanupPeriodDays`

152 * The [HIPAA configuration applies to your sessions](https://code.claude.com/docs/en/hipaa-setup.md#check-how-developers-sign-in-and-connect)152 * The [HIPAA configuration applies to your sessions](https://code.claude.com/docs/en/hipaa-setup.md#check-how-developers-sign-in-and-connect)

170* **Linux**: `/tmp/claude-<uid>/<project>/<session-id>/scratchpad/`, or the same shape under `$TMPDIR` when your system sets one170* **Linux**: `/tmp/claude-<uid>/<project>/<session-id>/scratchpad/`, or the same shape under `$TMPDIR` when your system sets one

171* **Windows**: `%TEMP%\claude\<project>\<session-id>\scratchpad\`171* **Windows**: `%TEMP%\claude\<project>\<session-id>\scratchpad\`

172 172 

173`<project>` is your working directory path with every character other than letters and digits replaced by `-`, such as `-Users-you-my-project`. If you set [`CLAUDE_CODE_TMPDIR`](https://code.claude.com/docs/en/env-vars), the tree moves under that directory instead. Hooks receive the current session's path as [`scratchpad_dir`](https://code.claude.com/docs/en/hooks#common-input-fields).173`<project>` is your working directory path with every character other than letters and digits replaced by `-`, such as `-Users-you-my-project`. On Windows, `<project>` is a 16-character hash of that string, such as `73295dabccd14c56`. In a Desktop app session, it can be the string itself. To get the exact path, ask Claude. If you set [`CLAUDE_CODE_TMPDIR`](https://code.claude.com/docs/en/env-vars), the tree moves under that directory instead. Hooks receive the current session's path as [`scratchpad_dir`](https://code.claude.com/docs/en/hooks#common-input-fields).

174 174 

175Scratchpad files last as long as the session's transcript: the [retention sweep](#cleaned-up-automatically) deletes the directory when it deletes the transcript, and [`claude purge`](#clear-local-data) doesn't touch the temp directory. Because the directory sits under the system temp location, your operating system can also clear it, such as on restart. To keep something Claude wrote there, ask Claude to move it into your project.175Scratchpad files last as long as the session's transcript: the [retention sweep](#cleaned-up-automatically) deletes the directory when it deletes the transcript, and [`claude purge`](#clear-local-data) doesn't touch the temp directory. Because the directory sits under the system temp location, your operating system can also clear it, such as on restart. To keep something Claude wrote there, ask Claude to move it into your project.

176 176 

65| `/bug [report]` | Report a bug or share your conversation. You choose how much session history to include and confirm on a consent screen before anything is sent. When you're signed in to Anthropic on a first-party connection, the report goes to Anthropic; on a third-party provider, or without Anthropic credentials, Claude Code writes the report to a [local archive under `~/.claude/feedback-bundles/`](https://code.claude.com/docs/en/data-usage#telemetry-services) that you forward yourself. In the [VS Code extension](https://code.claude.com/docs/en/vs-code.md#use-the-prompt-box), `/bug` opens the extension's own feedback dialog instead; requires Claude Code v2.1.229 or later. When you run it while Claude is responding, Claude Code opens the dialog immediately. Before v2.1.232, Claude Code queued the command until the turn finished. Alias: `/share`. Before v2.1.212, `/bug` and `/share` were aliases of `/feedback` |65| `/bug [report]` | Report a bug or share your conversation. You choose how much session history to include and confirm on a consent screen before anything is sent. When you're signed in to Anthropic on a first-party connection, the report goes to Anthropic; on a third-party provider, or without Anthropic credentials, Claude Code writes the report to a [local archive under `~/.claude/feedback-bundles/`](https://code.claude.com/docs/en/data-usage#telemetry-services) that you forward yourself. In the [VS Code extension](https://code.claude.com/docs/en/vs-code.md#use-the-prompt-box), `/bug` opens the extension's own feedback dialog instead; requires Claude Code v2.1.229 or later. When you run it while Claude is responding, Claude Code opens the dialog immediately. Before v2.1.232, Claude Code queued the command until the turn finished. Alias: `/share`. Before v2.1.212, `/bug` and `/share` were aliases of `/feedback` |

66| `/cd <path>` | Move this session to a new working directory, keeping the conversation. Type a partial path to see matching directory suggestions; press `Tab` to accept one. For what Claude Code applies from the new directory as soon as you move, and how `/cd` differs from `/add-dir`, see [Move the session to another directory](https://code.claude.com/docs/en/permissions.md#move-the-session-to-another-directory) |66| `/cd <path>` | Move this session to a new working directory, keeping the conversation. Type a partial path to see matching directory suggestions; press `Tab` to accept one. For what Claude Code applies from the new directory as soon as you move, and how `/cd` differs from `/add-dir`, see [Move the session to another directory](https://code.claude.com/docs/en/permissions.md#move-the-session-to-another-directory) |

67| `/chrome` | Configure [Claude in Chrome](https://code.claude.com/docs/en/chrome.md) settings |67| `/chrome` | Configure [Claude in Chrome](https://code.claude.com/docs/en/chrome.md) settings |

68| `/claude-api [migrate\|upgrade\|api-version\|managed-agents-onboard\|managed-agents-secure\|prompt-audit\|cost-optimize\|build-eval\|hillclimb\|preserved-thinking-migration]` | **[Skill](https://code.claude.com/docs/en/skills.md#bundled-skills).** Load [Claude API](https://platform.claude.com/docs/en/api/overview) and [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) reference material for your project's language. Also activates automatically when your code imports `anthropic` or `@anthropic-ai/sdk`. For what each subcommand does and the version it requires, see [Work on Claude API projects](https://code.claude.com/docs/en/skills.md#work-on-claude-api-projects) |68| `/claude-api [migrate\|upgrade\|api-version\|managed-agents-onboard\|managed-agents-secure\|prompt-audit\|cost-optimize\|latency-optimize\|build-eval\|hillclimb\|preserved-thinking-migration]` | **[Skill](https://code.claude.com/docs/en/skills.md#bundled-skills).** Load [Claude API](https://platform.claude.com/docs/en/api/overview) and [Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview) reference material for your project's language. Also activates automatically when your code imports `anthropic` or `@anthropic-ai/sdk`. For what each subcommand does and the version it requires, see [Work on Claude API projects](https://code.claude.com/docs/en/skills.md#work-on-claude-api-projects) |

69| `/claude-in-chrome [task]` | **[Skill](https://code.claude.com/docs/en/skills.md#bundled-skills).** Have Claude carry out a task in your browser, such as testing a page, filling a form, or reading console logs, through [Claude in Chrome](https://code.claude.com/docs/en/chrome.md). Available when Chrome integration is enabled for the session, for example with `claude --chrome`, or when Claude Code can offer to [install the extension](https://code.claude.com/docs/en/chrome.md#install-the-extension-when-claude-asks) |69| `/claude-in-chrome [task]` | **[Skill](https://code.claude.com/docs/en/skills.md#bundled-skills).** Have Claude carry out a task in your browser, such as testing a page, filling a form, or reading console logs, through [Claude in Chrome](https://code.claude.com/docs/en/chrome.md). Available when Chrome integration is enabled for the session, for example with `claude --chrome`, or when Claude Code can offer to [install the extension](https://code.claude.com/docs/en/chrome.md#install-the-extension-when-claude-asks) |

70| `/clear [name]` | Start a new conversation with empty context. Pass a name to label the previous conversation in the `/resume` picker. To free up context while continuing the same conversation, use `/compact` instead. Resume the previous conversation with `/resume`, or, in the same Claude Code process, restore it from [the rewind menu's previous-session entry](https://code.claude.com/docs/en/checkpointing.md#rewind-past-a-cleared-conversation). Aliases: `/reset`, `/new` |70| `/clear [name]` | Start a new conversation with empty context. Pass a name to label the previous conversation in the `/resume` picker. To free up context while continuing the same conversation, use `/compact` instead. Resume the previous conversation with `/resume`, or, in the same Claude Code process, restore it from [the rewind menu's previous-session entry](https://code.claude.com/docs/en/checkpointing.md#rewind-past-a-cleared-conversation). Aliases: `/reset`, `/new` |

71| `/code-review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [--max-findings n\|all\|default] [pr#\|branch\|path]` | **[Skill](https://code.claude.com/docs/en/skills.md#bundled-skills).** Review the current diff, or a PR number, branch, or path you pass, for correctness bugs. Depending on your model and effort level, the review also covers cleanup opportunities. Pass `--fix` to have Claude apply the findings, `--comment` to post them on the GitHub PR or GitLab merge request, or `ultra` to run a deep [cloud review](https://code.claude.com/docs/en/ultrareview.md). Posting to a GitLab merge request requires Claude Code v2.1.257 or later. With `ultra` on a `github.com` PR target, pass `--post` to preselect [posting the finished findings to the PR](https://code.claude.com/docs/en/ultrareview.md#post-findings-to-the-pull-request) in the launch dialog; `--post` requires Claude Code v2.1.227 or later. See [Review locally with `/code-review`](https://code.claude.com/docs/en/code-review-local) for the effort levels, targeting, and how it relates to `/simplify`. Alias: `/review` |71| `/code-review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [--max-findings n\|all\|default] [pr#\|branch\|path]` | **[Skill](https://code.claude.com/docs/en/skills.md#bundled-skills).** Review the current diff, or a PR number, branch, or path you pass, for correctness bugs. Depending on your model and effort level, the review also covers cleanup opportunities. Pass `--fix` to have Claude apply the findings, `--comment` to post them on the GitHub PR or GitLab merge request, or `ultra` to run a deep [cloud review](https://code.claude.com/docs/en/ultrareview.md). Posting to a GitLab merge request requires Claude Code v2.1.257 or later. With `ultra` on a `github.com` PR target, pass `--post` to preselect [posting the finished findings to the PR](https://code.claude.com/docs/en/ultrareview.md#post-findings-to-the-pull-request) in the launch dialog; `--post` requires Claude Code v2.1.227 or later. See [Review locally with `/code-review`](https://code.claude.com/docs/en/code-review-local) for the effort levels, targeting, and how it relates to `/simplify`. Alias: `/review` |

466claude --worktree feature-auth466claude --worktree feature-auth

467```467```

468 468 

469Run the same command with a different name in a second terminal to start an isolated parallel session. In a repository with no commits, the command fails with `Failed to resolve base branch "HEAD": git rev-parse failed`. See [Worktrees](https://code.claude.com/docs/en/worktrees.md) for cleanup, `.worktreeinclude`, and non-git VCS support. To monitor parallel sessions from one screen instead of separate terminals, see [background agents](https://code.claude.com/docs/en/agent-view.md).469Run the same command with a different name in a second terminal to start an isolated parallel session. See [Worktrees](https://code.claude.com/docs/en/worktrees.md) for cleanup, `.worktreeinclude`, and non-git VCS support. To monitor parallel sessions from one screen instead of separate terminals, see [background agents](https://code.claude.com/docs/en/agent-view.md).

470 470 

471## Plan before editing471## Plan before editing

472 472 

19 19 

20## What survives compaction20## What survives compaction

21 21 

22When a long session compacts, Claude Code summarizes the conversation history to fit the context window. As of v2.1.198, the summarization request inherits your session's [extended thinking](https://code.claude.com/docs/en/model-config.md#extended-thinking) configuration, so it reasons with thinking enabled when your session has it enabled and stays off otherwise. Thinking affects only how the summary is produced; your session settings are unchanged afterward. What happens to each kind of content depends on how it was loaded:22When a long session compacts, Claude Code summarizes the conversation history to fit the context window.

23 23 

24| Mechanism | After compaction |24| Mechanism | After compaction |

25| :- | :- |25| :- | :- |

243| `CLAUDE_CODE_DISABLE_BUNDLED_SKILLS` | Set to `1` to disable the [skills](https://code.claude.com/docs/en/skills.md) and workflows included with Claude Code: bundled skills and workflows are removed entirely, while built-in commands like `/init` stay typable but are hidden from the model. `/doctor` stays typable like the built-in commands; hide it with `DISABLE_DOCTOR_COMMAND` instead. Skills from plugins, `.claude/skills/`, and `.claude/commands/` are unaffected. Equivalent to the [`disableBundledSkills`](https://code.claude.com/docs/en/settings-reference#disablebundledskills) setting |243| `CLAUDE_CODE_DISABLE_BUNDLED_SKILLS` | Set to `1` to disable the [skills](https://code.claude.com/docs/en/skills.md) and workflows included with Claude Code: bundled skills and workflows are removed entirely, while built-in commands like `/init` stay typable but are hidden from the model. `/doctor` stays typable like the built-in commands; hide it with `DISABLE_DOCTOR_COMMAND` instead. Skills from plugins, `.claude/skills/`, and `.claude/commands/` are unaffected. Equivalent to the [`disableBundledSkills`](https://code.claude.com/docs/en/settings-reference#disablebundledskills) setting |

244| `CLAUDE_CODE_DISABLE_CFC_PROMPT` | For hosts that embed Claude Code and supply their own browser guidance. Set to `1` to keep the [Claude in Chrome](https://code.claude.com/docs/en/chrome.md) browser tools available while omitting the Chrome section of the system prompt and the `/claude-in-chrome` [bundled skill](https://code.claude.com/docs/en/skills.md#bundled-skills). In VS Code, the extension at v2.1.314 or later also leaves the Claude in Chrome instructions out of a session's messages. Requires Claude Code v2.1.257 or later |244| `CLAUDE_CODE_DISABLE_CFC_PROMPT` | For hosts that embed Claude Code and supply their own browser guidance. Set to `1` to keep the [Claude in Chrome](https://code.claude.com/docs/en/chrome.md) browser tools available while omitting the Chrome section of the system prompt and the `/claude-in-chrome` [bundled skill](https://code.claude.com/docs/en/skills.md#bundled-skills). In VS Code, the extension at v2.1.314 or later also leaves the Claude in Chrome instructions out of a session's messages. Requires Claude Code v2.1.257 or later |

245| `CLAUDE_CODE_DISABLE_CLAUDE_MDS` | Set to `1` to prevent loading any CLAUDE.md memory files into context, including user, project, and auto memory files |245| `CLAUDE_CODE_DISABLE_CLAUDE_MDS` | Set to `1` to prevent loading any CLAUDE.md memory files into context, including user, project, and auto memory files |

246| `CLAUDE_CODE_DISABLE_CRON` | Set to `1` to disable [session-scoped scheduled tasks](https://code.claude.com/docs/en/scheduled-tasks.md). The `/loop` skill and cron tools become unavailable and any already-scheduled tasks stop firing, including tasks that are already running mid-session |246| `CLAUDE_CODE_DISABLE_CRON` | Set to `1` to disable [session-scoped scheduled tasks](https://code.claude.com/docs/en/scheduled-tasks.md#disable-scheduled-tasks). The `/loop` skill and `CronCreate` become unavailable and any already-scheduled tasks stop firing, including tasks that are already running mid-session |

247| `CLAUDE_CODE_DISABLE_DANGEROUS_RM_TIMEOUT` | Set to `1` to turn off the `auto` and `bypassPermissions` [time limit and denials](https://code.claude.com/docs/en/permission-modes.md#time-limits-and-denials-in-auto-and-bypasspermissions-modes) on critical-path removals. These removals then go to the classifier in `auto` mode, and get an untimed prompt in `bypassPermissions` mode wherever Claude Code can show one. Set it in the environment that launches Claude Code, since Claude Code ignores a copy delivered through a settings `env` block. Requires Claude Code v2.1.281 or later |247| `CLAUDE_CODE_DISABLE_DANGEROUS_RM_TIMEOUT` | Set to `1` to turn off the `auto` and `bypassPermissions` [time limit and denials](https://code.claude.com/docs/en/permission-modes.md#time-limits-and-denials-in-auto-and-bypasspermissions-modes) on critical-path removals. These removals then go to the classifier in `auto` mode, and get an untimed prompt in `bypassPermissions` mode wherever Claude Code can show one. Set it in the environment that launches Claude Code, since Claude Code ignores a copy delivered through a settings `env` block. Requires Claude Code v2.1.281 or later |

248| `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS` | Set to `1` to strip pre-release `anthropic-beta` request headers, the body fields that pair with them, and beta tool-schema fields such as `defer_loading` and `eager_input_streaming` from API requests. Use this when a proxy gateway rejects requests with an `Unexpected value(s)` error for the `anthropic-beta` header or an `Extra inputs are not permitted` error. [Disable pre-release capabilities](https://code.claude.com/docs/en/llm-gateway-protocol.md#disable-pre-release-capabilities) lists what the variable removes, including [MCP tool search](https://code.claude.com/docs/en/mcp.md#scale-with-mcp-tool-search), and what Claude Code keeps sending |248| `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS` | Set to `1` to strip pre-release `anthropic-beta` request headers, the body fields that pair with them, and beta tool-schema fields such as `defer_loading` and `eager_input_streaming` from API requests. Use this when a proxy gateway rejects requests with an `Unexpected value(s)` error for the `anthropic-beta` header or an `Extra inputs are not permitted` error. [Disable pre-release capabilities](https://code.claude.com/docs/en/llm-gateway-protocol.md#disable-pre-release-capabilities) lists what the variable removes, including [MCP tool search](https://code.claude.com/docs/en/mcp.md#scale-with-mcp-tool-search), and what Claude Code keeps sending |

249| `CLAUDE_CODE_DISABLE_EXPLORE_PLAN_AGENTS` | Set to `1` to disable the built-in [Explore and Plan subagents](https://code.claude.com/docs/en/sub-agents.md#built-in-subagents). Claude explores with its search tools or the general-purpose subagent instead, and [plan mode](https://code.claude.com/docs/en/permission-modes.md#analyze-before-you-edit-with-plan-mode) reads files directly rather than launching Explore and Plan agents. Custom subagents named `Explore` or `Plan` are unaffected. To remove every built-in subagent type in the Agent SDK or non-interactive mode, use `CLAUDE_AGENT_SDK_DISABLE_BUILTIN_AGENTS` instead. Requires Claude Code v2.1.198 or later |249| `CLAUDE_CODE_DISABLE_EXPLORE_PLAN_AGENTS` | Set to `1` to disable the built-in [Explore and Plan subagents](https://code.claude.com/docs/en/sub-agents.md#built-in-subagents). Claude explores with its search tools or the general-purpose subagent instead, and [plan mode](https://code.claude.com/docs/en/permission-modes.md#analyze-before-you-edit-with-plan-mode) reads files directly rather than launching Explore and Plan agents. Custom subagents named `Explore` or `Plan` are unaffected. To remove every built-in subagent type in the Agent SDK or non-interactive mode, use `CLAUDE_AGENT_SDK_DISABLE_BUILTIN_AGENTS` instead. Requires Claude Code v2.1.198 or later |

2349**What to do:**2349**What to do:**

2350 2350 

2351* **Not signed in**: the message continues `: not signed in. Run /login, then retry.` Run `/login`, then send again2351* **Not signed in**: the message continues `: not signed in. Run /login, then retry.` Run `/login`, then send again

2352* **Your organization's refusal**: for example `: Feedback has not been enabled for your organization.` or `: Feedback collection is not available for organizations with custom data retention policies.` If the message adds `To report a bug, open an issue`, do so at [github.com/anthropics/claude-code/issues](https://github.com/anthropics/claude-code/issues). `/feedback` can then answer that it's [disabled by your organization's policy](#feedback-is-disabled-by-your-organizations-policy) for seven days. A queued draft stays queued and still [expires on schedule](https://code.claude.com/docs/en/tools-reference.md#discard-or-keep-a-draft)2352* **Your organization's refusal**: for example `: Feedback has not been enabled for your organization.` or `: Feedback collection is not available for organizations with custom data retention policies.` `/feedback` can then answer that it's [disabled by your organization's policy](#feedback-is-disabled-by-your-organizations-policy) for seven days. A queued draft stays queued and still [expires on schedule](https://code.claude.com/docs/en/tools-reference.md#discard-or-keep-a-draft)

2353* **Refused credentials**: the message continues `: the sign-in this session uses is not allowed to send feedback` or `: the server did not accept the credentials this session uses`. Run `/status` at the Claude Code prompt to see which login, token, or API key this session sends2353* **Refused credentials**: the message continues `: the sign-in this session uses is not allowed to send feedback` or `: the server did not accept the credentials this session uses`. Run `/status` at the Claude Code prompt to see which login, token, or API key this session sends

2354* **`(request timed out)` or `(couldn't reach the service)` straight after `Couldn't send feedback`**: send the report or the queued draft again. The dialog keeps your text, and a queued draft's message says `The draft is still queued.` If other requests are failing too, check your network connection and see [Unable to connect to API](#unable-to-connect-to-api). If it keeps failing, file the report at [github.com/anthropics/claude-code/issues](https://github.com/anthropics/claude-code/issues)2354* **`(request timed out)` or `(couldn't reach the service)` straight after `Couldn't send feedback`**: send the report or the queued draft again. The dialog keeps your text, and a queued draft's message says `The draft is still queued.` If other requests are failing too, check your network connection and see [Unable to connect to API](#unable-to-connect-to-api). If it keeps failing, file the report at [github.com/anthropics/claude-code/issues](https://github.com/anthropics/claude-code/issues)

2355* **`(server returned <status>)`, or no reason, straight after `Couldn't send feedback`**: the status is the service's response code. The service's own message follows it when one was sent, as in `(server returned 403: Access Forbidden)`. If it keeps failing, file the report at [github.com/anthropics/claude-code/issues](https://github.com/anthropics/claude-code/issues)2355* **`(server returned <status>)`, or no reason, straight after `Couldn't send feedback`**: the status is the service's response code. The service's own message follows it when one was sent, as in `(server returned 403: Access Forbidden)`. If it keeps failing, file the report at [github.com/anthropics/claude-code/issues](https://github.com/anthropics/claude-code/issues)

6622 6622 

6623**2. Update every host that shares the home directory**6623**2. Update every host that shares the home directory**

6624 6624 

6625If several hosts share this home directory, first update Claude Code on every host to the version the message names or later. An older version can write a moved `.config.json` back.6625If several hosts share this home directory, first update Claude Code on every host to the version the message names or later.

6626 6626 

6627**3. Stop Claude Code**6627**3. Stop Claude Code**

6628 6628 

54 54 

55### Auto mode55### Auto mode

56 56 

57A [permission mode](#permission-mode) where a separate classifier model reviews actions instead of you, so Claude Code runs most of them without asking you. Claude Code still asks you before actions your explicit ask rules match. The classifier blocks scope escalation, untrusted infrastructure, and [prompt injection](#prompt-injection). Tool results are stripped from what it sees, so hostile content in a file or web page can't manipulate it directly.57A [permission mode](#permission-mode) where a separate classifier model reviews actions instead of you, so Claude Code runs most of them without asking you. Claude Code still asks you before actions your explicit ask rules match. The classifier blocks scope escalation, untrusted infrastructure, and [prompt injection](#prompt-injection).

58 58 

59* **Claude Code v2.1.283 or later**: auto mode is the [built-in starting permission mode](https://code.claude.com/docs/en/permission-modes.md#which-mode-a-session-starts-in) for interactive terminal and VS Code sessions59* **Claude Code v2.1.283 or later**: auto mode is the [built-in starting permission mode](https://code.claude.com/docs/en/permission-modes.md#which-mode-a-session-starts-in) for interactive terminal and VS Code sessions

60* **Earlier versions**: auto mode is the built-in starting permission mode only on Pro, Max, Ultra, and Team plans60* **Earlier versions**: auto mode is the built-in starting permission mode only on Pro, Max, Ultra, and Team plans

261 261 

262### Prompt injection262### Prompt injection

263 263 

264Hostile instructions embedded in a file, web page, or tool result that attempt to redirect Claude toward actions you never asked for. Claude Code's defenses include the permission system, command injection detection, and trust verification. [Auto mode](#auto-mode) adds a server-side probe that scans tool results for suspicious content and a classifier that reviews actions with tool results stripped, so injected text can't manipulate it directly.264Hostile instructions embedded in a file, web page, or tool result that attempt to redirect Claude toward actions you didn't ask for. Claude Code's defenses include the permission system, command injection detection, and trust verification. In [auto mode](#auto-mode), a server-side probe scans tool results for suspicious content, and a classifier reviews actions.

265 265 

266Learn more: [Protect against prompt injection](https://code.claude.com/docs/en/security.md#protect-against-prompt-injection)266Learn more: [Protect against prompt injection](https://code.claude.com/docs/en/security.md#protect-against-prompt-injection)

267 267 

2341 2341 

2342#### Annotate a result for the auto mode classifier2342#### Annotate a result for the auto mode classifier

2343 2343 

2344Return `classifierContext` to send a short note about the tool call's result to the [auto mode](https://code.claude.com/docs/en/permission-modes.md#eliminate-prompts-with-auto-mode) classifier rather than to Claude. The classifier [never receives tool results themselves](https://code.claude.com/docs/en/permission-modes.md#how-the-classifier-evaluates-actions), so this field is the supported way to tell it something about what a call returned before it reviews later actions. The field requires Claude Code v2.1.236 or later.2344Return `classifierContext` to send a short note about the tool call's result to the [auto mode](https://code.claude.com/docs/en/permission-modes.md#eliminate-prompts-with-auto-mode) classifier rather than to Claude. The note is for the classifier's review of later actions. The field requires Claude Code v2.1.236 or later.

2345 2345 

2346The example below tells the classifier where a query's output came from:2346The example below tells the classifier where a query's output came from:

2347 2347 

52* **Delivery controls**: [`policyHelper`](https://code.claude.com/docs/en/settings-reference#policyhelper), [`wslInheritsWindowsSettings`](https://code.claude.com/docs/en/settings-reference#wslinheritswindowssettings), and [`managedSourcesBehavior`](https://code.claude.com/docs/en/settings-reference#managedsourcesbehavior)52* **Delivery controls**: [`policyHelper`](https://code.claude.com/docs/en/settings-reference#policyhelper), [`wslInheritsWindowsSettings`](https://code.claude.com/docs/en/settings-reference#wslinheritswindowssettings), and [`managedSourcesBehavior`](https://code.claude.com/docs/en/settings-reference#managedsourcesbehavior)

53* **Gateway login keys**: [`forceLoginGatewayUrl`](https://code.claude.com/docs/en/settings-reference#forcelogingatewayurl), [`gatewayInternalNetworks`](https://code.claude.com/docs/en/settings-reference#gatewayinternalnetworks), and the `"gateway"` value of [`forceLoginMethod`](https://code.claude.com/docs/en/settings-reference#forceloginmethod)53* **Gateway login keys**: [`forceLoginGatewayUrl`](https://code.claude.com/docs/en/settings-reference#forcelogingatewayurl), [`gatewayInternalNetworks`](https://code.claude.com/docs/en/settings-reference#gatewayinternalnetworks), and the `"gateway"` value of [`forceLoginMethod`](https://code.claude.com/docs/en/settings-reference#forceloginmethod)

54 54 

55A managed settings file, an MDM profile, or the claude.ai console applies one policy to everyone it reaches. To give one group of developers a different policy, deploy a different file or profile to that group; the claude.ai console [can't target a group yet](https://code.claude.com/docs/en/server-managed-settings.md#current-limitations), while a self-hosted [Claude apps gateway](https://code.claude.com/docs/en/claude-apps-gateway) delivers managed settings per IdP group.55With a managed settings file or an MDM profile, everyone it reaches gets one policy. To give one group of developers a different policy, deploy a different file or profile to that group. On an Enterprise plan, you can also [give each group its own configuration](https://code.claude.com/docs/en/server-managed-settings.md#give-groups-different-settings) in server-managed settings. A self-hosted [Claude apps gateway](https://code.claude.com/docs/en/claude-apps-gateway) delivers managed settings per IdP group.

56 56 

57When more than one mechanism delivers a policy to the same machine, Claude Code by default uses one and ignores the others. [How Claude Code combines managed sources](#how-claude-code-combines-managed-sources) gives the order and the opt-in that applies every source.57When more than one mechanism delivers a policy to the same machine, Claude Code by default uses one and ignores the others. [How Claude Code combines managed sources](#how-claude-code-combines-managed-sources) gives the order and the opt-in that applies every source.

58 58 

168 168 

169See [Model is not a recognized model id](https://code.claude.com/docs/en/errors.md#model-is-not-a-recognized-model-id) and [Model not found](https://code.claude.com/docs/en/errors.md#model-not-found) for the messages.169See [Model is not a recognized model id](https://code.claude.com/docs/en/errors.md#model-is-not-a-recognized-model-id) and [Model not found](https://code.claude.com/docs/en/errors.md#model-not-found) for the messages.

170 170 

171If you set the model with the `--model` flag, the `ANTHROPIC_MODEL` environment variable, or the `model` setting, Claude Code doesn't reject a mistyped value up front. The session starts, and the first request produces [There's an issue with the selected model](https://code.claude.com/docs/en/errors.md#theres-an-issue-with-the-selected-model). On the Anthropic API, a value that doesn't start with `claude-` and nearly spells a model name or alias can also produce a [startup warning that suggests it](#fix-a-mistyped-model-name).171If you set the model with the `--model` flag, the `ANTHROPIC_MODEL` environment variable, or the `model` setting, Claude Code doesn't reject a mistyped value up front. The session starts, and the first request produces [There's an issue with the selected model](https://code.claude.com/docs/en/errors.md#theres-an-issue-with-the-selected-model). On the Anthropic API, you can see a [warning that suggests the closest name](#fix-a-mistyped-model-name) if the value nearly spells a model name or alias and doesn't start with `claude-`.

172 172 

173When the requested model has a scheduled retirement date or is automatically remapped to a newer version, Claude Code shows a warning that names the requested model. Interactive sessions show it as a startup notice. From v2.1.182, the same warning is written to stderr in [non-interactive mode](https://code.claude.com/docs/en/headless.md) when using the default text output format. The check also covers a `model` set in [subagent frontmatter](https://code.claude.com/docs/en/sub-agents.md). The stderr warning is suppressed for `--output-format json` and `stream-json`; read the actual model from the `modelUsage` field of the [result message](https://code.claude.com/docs/en/headless.md#get-structured-output) instead.173When the requested model has a scheduled retirement date or is automatically remapped to a newer version, Claude Code shows a warning that names the requested model. Interactive sessions show it as a startup notice. From v2.1.182, the same warning is written to stderr in [non-interactive mode](https://code.claude.com/docs/en/headless.md) when using the default text output format. The check also covers a `model` set in [subagent frontmatter](https://code.claude.com/docs/en/sub-agents.md). The stderr warning is suppressed for `--output-format json` and `stream-json`; read the actual model from the `modelUsage` field of the [result message](https://code.claude.com/docs/en/headless.md#get-structured-output) instead.

174 174 

195}195}

196```196```

197 197 

198#### Fix a mistyped model name

199 

200Either of these warnings means Claude Code doesn't recognize the model ID you set, and the ID nearly spells a model name or alias it does recognize:

201 

202```text theme={null}

203⚠ Model "<model>" isn't recognized · did you mean "<suggestion>"?

204Claude Code doesn't recognize the model "<model>". Did you mean "<suggestion>"?

205```

206 

207* **You mistyped the ID**: change it to the suggestion [wherever you set the model](#setting-your-model)

208* **The ID is deliberate**: leave it as it is. The session sends it as you wrote it. If the context window Claude Code assumes for it is wrong, [correct the window](#correct-the-window-for-a-gateway-or-custom-model-id)

209 

198#### Set a default model for new sessions210#### Set a default model for new sessions

199 211 

200Set `ANTHROPIC_DEFAULT_MODEL=<alias|name>` to choose the model your sessions start on by default. Requires Claude Code v2.1.236 or later.212Set `ANTHROPIC_DEFAULT_MODEL=<alias|name>` to choose the model your sessions start on by default. Requires Claude Code v2.1.236 or later.

772| :- | :- |784| :- | :- |

773| Toggle for the current session | Press `Option+T` on macOS or `Alt+T` on Windows and Linux |785| Toggle for the current session | Press `Option+T` on macOS or `Alt+T` on Windows and Linux |

774| Set the global default | Run `/config` and toggle thinking mode. Saved as `alwaysThinkingEnabled` in `~/.claude/settings.json` |786| Set the global default | Run `/config` and toggle thinking mode. Saved as `alwaysThinkingEnabled` in `~/.claude/settings.json` |

775| Disable through an environment variable | Set [`MAX_THINKING_TOKENS=0`](https://code.claude.com/docs/en/env-vars), which turns thinking off on the Anthropic API except on Opus 5.5, Sonnet 5.5, Haiku 5.5, and Fable models. On [third-party providers](https://code.claude.com/docs/en/third-party-integrations.md), Claude Code omits the `thinking` parameter instead, and adaptive-reasoning models may still think |787| Disable through an environment variable | Set [`MAX_THINKING_TOKENS=0`](https://code.claude.com/docs/en/env-vars) |

788 

789Claude Code collapses thinking output by default. Press `Ctrl+O` to toggle verbose mode and see the reasoning as gray italic text. Interactive sessions on the Anthropic API receive redacted thinking blocks by default, so set [`showThinkingSummaries: true`](https://code.claude.com/docs/en/settings-reference#showthinkingsummaries) in settings if you want the full summaries available when you expand. You are charged for all thinking tokens generated, even when collapsed or redacted.

790 

791#### Where you can't turn thinking off

776 792 

777You can't turn thinking off on Opus 5.5, Sonnet 5.5, Haiku 5.5, or the Fable models. The session toggle and the `/config` row show `Thinking can't be turned off` for these models instead of offering the switch, and a saved `alwaysThinkingEnabled: false` or `MAX_THINKING_TOKENS=0` has no effect there. On these models, the model decides per step how much to think based on the effort level. The saved setting applies again when you switch to a model that accepts it.793You can't turn thinking off on Opus 5.5, Sonnet 5.5, Haiku 5.5, or the Fable models. The session toggle and the `/config` row show `Thinking can't be turned off` for these models instead of offering the switch, and a saved `alwaysThinkingEnabled: false` or `MAX_THINKING_TOKENS=0` has no effect there. On these models, the model decides per step how much to think based on the effort level. The saved setting applies again when you switch to a model that accepts it.

778 794 

779Claude Code collapses thinking output by default. Press `Ctrl+O` to toggle verbose mode and see the reasoning as gray italic text. Interactive sessions on the Anthropic API receive redacted thinking blocks by default, so set `showThinkingSummaries: true` in [settings](https://code.claude.com/docs/en/settings.md) if you want the full summaries available when you expand. You are charged for all thinking tokens generated, even when collapsed or redacted.795To stop or reduce thinking, use the step for your provider:

796 

797* **Anthropic API**: you can turn thinking off on models such as `claude-opus-5`, `claude-sonnet-5`, and `claude-haiku-4-5`. [Aliases](#model-aliases) such as `sonnet` select models where you can't, so [set the model](#setting-your-model) by its ID.

798* **[Third-party providers](https://code.claude.com/docs/en/third-party-integrations.md)**: on providers such as Amazon Bedrock and Microsoft Foundry, models with [adaptive reasoning](#adaptive-reasoning-and-fixed-thinking-budgets) may still think with thinking off. [Lower the effort level](#set-the-effort-level) to reduce thinking.

780 799 

781### Extended context800### Extended context

782 801 

900 919 

901Once Claude Code recognizes the ID or uses the window you gave it, the warning no longer appears when you start a session on that ID.920Once Claude Code recognizes the ID or uses the window you gave it, the warning no longer appears when you start a session on that ID.

902 921 

903#### Fix a mistyped model name

904 

905Either of these warnings means Claude Code doesn't recognize the model ID you set, and the ID nearly spells a model name or alias it does recognize:

906 

907```text theme={null}

908⚠ Model "<model>" isn't recognized · did you mean "<suggestion>"?

909Claude Code doesn't recognize the model "<model>". Did you mean "<suggestion>"?

910```

911 

912* **You mistyped the ID**: change it to the suggestion [wherever you set the model](#setting-your-model)

913* **The ID is deliberate**: leave it as it is. The session sends it as you wrote it. If the context window Claude Code assumes for it is wrong, [correct the window](#correct-the-window-for-a-gateway-or-custom-model-id)

914 

915#### Which IDs `CLAUDE_CODE_MAX_CONTEXT_TOKENS` applies to922#### Which IDs `CLAUDE_CODE_MAX_CONTEXT_TOKENS` applies to

916 923 

917On an [LLM gateway](https://code.claude.com/docs/en/llm-gateway) or other custom deployment, Claude Code can assume a context window for the model ID that differs from the model's real window, whether or not it resolves the ID to a Claude model. Set [`CLAUDE_CODE_MAX_CONTEXT_TOKENS`](https://code.claude.com/docs/en/env-vars) to the window Claude Code should assume instead.924On an [LLM gateway](https://code.claude.com/docs/en/llm-gateway) or other custom deployment, Claude Code can assume a context window for the model ID that differs from the model's real window, whether or not it resolves the ID to a Claude model. Set [`CLAUDE_CODE_MAX_CONTEXT_TOKENS`](https://code.claude.com/docs/en/env-vars) to the window Claude Code should assume instead.

484* Opening a pull request against a different repository or organization, forking with `gh repo fork`, or pushing to a third-party repository, unless you named that external target484* Opening a pull request against a different repository or organization, forking with `gh repo fork`, or pushing to a third-party repository, unless you named that external target

485* Content from a sensitive local store, or from a file whose name, path, or type marks it as sensitive, entering a commit, a push, PR or issue text, a gist or paste, or a package publish, unless you named both the source and the destination. Session transcripts and conversation logs, credential and configuration dot-folders such as SSH keys, cloud credentials, browser profiles, and shell history, and user-data exports all count, and the repository being private doesn't clear it485* Content from a sensitive local store, or from a file whose name, path, or type marks it as sensitive, entering a commit, a push, PR or issue text, a gist or paste, or a package publish, unless you named both the source and the destination. Session transcripts and conversation logs, credential and configuration dot-folders such as SSH keys, cloud credentials, browser profiles, and shell history, and user-data exports all count, and the repository being private doesn't clear it

486* Writing to Claude Code session transcripts, the `.jsonl` history files under `~/.claude/projects/` or your configured config directory, whether directly or through a shell command. The rule also covers the metadata lines Claude Code appends to each transcript entry for its own checks. Reading a transcript isn't blocked486* Writing to Claude Code session transcripts, the `.jsonl` history files under `~/.claude/projects/` or your configured config directory, whether directly or through a shell command. The rule also covers the metadata lines Claude Code appends to each transcript entry for its own checks. Reading a transcript isn't blocked

487* A recursive forced delete such as `rm -rf "$VAR"` or `Remove-Item -Recurse -Force $dir` whose target is a shell variable that isn't assigned anywhere in the conversation the classifier sees, or a glob rooted at such a variable. The value came only from earlier command output, which the classifier never receives, so the classifier can't verify the deletion target against the other deletion rules. The block clears when you name the exact path being deleted, or when Claude re-runs the delete with the resolved literal path written into the command. Deletes whose target the classifier can resolve aren't affected.487* A recursive forced delete such as `rm -rf "$VAR"` or `Remove-Item -Recurse -Force $dir` whose target is a shell variable that isn't assigned in the conversation the classifier sees, or a glob rooted at such a variable. Without that assignment, the classifier can't tell what the command would delete.

488 488 

489 A glob directly under the variable, as in `rm -rf "$VAR"/*`, is a [critical path](#critical-paths) instead. `Remove-Item` targets that are a bare `*` or end in `/*` or `\*` never reach the classifier: Claude Code [denies them outright](#remove-item-in-powershell).489 The block clears when you name the exact path being deleted, or when Claude re-runs the delete with the resolved literal path written into the command. Deletes whose target the classifier can resolve aren't affected.

490 

491 A glob directly under the variable, as in `rm -rf "$VAR"/*`, is a [critical path](#critical-paths) instead. For a `Remove-Item` target that is a bare `*` or ends in `/*` or `\*`, see [Remove-Item in PowerShell](#remove-item-in-powershell): Claude Code denies that command outright.

490 492 

491Several of these categories depend on [environment](https://code.claude.com/docs/en/auto-mode-config.md#define-trusted-infrastructure) entries, such as sensitive remote targets and protected IaC scopes, that you can narrow to concrete names.493Several of these categories depend on [environment](https://code.claude.com/docs/en/auto-mode-config.md#define-trusted-infrastructure) entries, such as sensitive remote targets and protected IaC scopes, that you can narrow to concrete names.

492 494 

336 336 

337### Assign release channels to user groups337### Assign release channels to user groups

338 338 

339To run stable and early-access channels, host two marketplaces that point at different refs of the same plugins. Then give each user group its own marketplace through either separate endpoint-managed settings or a gateway policy. Server-managed settings from the admin console [apply to every user in your organization](https://code.claude.com/docs/en/server-managed-settings.md#current-limitations), so they can't assign different settings to different groups.339To run stable and early-access channels, host two marketplaces that point at different refs of the same plugins. Then give each user group its own marketplace through separate endpoint-managed settings or a gateway policy. On an Enterprise plan, you can also use server-managed settings from the admin console.

340 340 

341* Deploy separate [endpoint-managed settings](https://code.claude.com/docs/en/managed-settings.md#delivery-mechanisms), such as a managed settings file or an MDM profile, to each group's devices. To check whether the per-group file or profile applies on a device that also has an organization-wide source, see [How Claude Code combines managed sources](https://code.claude.com/docs/en/managed-settings.md#precedence-within-the-managed-tier).341* Deploy separate [endpoint-managed settings](https://code.claude.com/docs/en/managed-settings.md#delivery-mechanisms), such as a managed settings file or an MDM profile, to each group's devices. To check whether the per-group file or profile applies on a device that also has an organization-wide source, see [How Claude Code combines managed sources](https://code.claude.com/docs/en/managed-settings.md#precedence-within-the-managed-tier).

342* Define one [Claude apps gateway policy](https://code.claude.com/docs/en/claude-apps-gateway-config.md#managed) per group. By default, the gateway applies the first policy whose match rule fits a user, so order the policies so that each user reaches their group's policy. That policy's `extraKnownMarketplaces` map doesn't merge with any other policy's, so list every marketplace the group needs in it, not only its channel marketplace.342* Define one [Claude apps gateway policy](https://code.claude.com/docs/en/claude-apps-gateway-config.md#managed) per group. By default, the gateway applies the first policy whose match rule fits a user, so order the policies so that each user reaches their group's policy. That policy's `extraKnownMarketplaces` map doesn't merge with any other policy's, so list every marketplace the group needs in it, not only its channel marketplace.

343* On an Enterprise plan, [give each group its own configuration](https://code.claude.com/docs/en/server-managed-settings.md#give-groups-different-settings) in the admin console. A group's configuration doesn't merge with the organization-wide one, so put every setting the group needs in the group's own configuration, including each marketplace.

343 344 

344With either mechanism, the stable group receives this configuration:345With any of these mechanisms, the stable group receives this setting:

345 346 

346```json theme={null}347```json theme={null}

347{348{

383 384 

384These requests from security reviews have no dedicated key in the current settings schema. The nearest existing controls are:385These requests from security reviews have no dedicated key in the current settings schema. The nearest existing controls are:

385 386 

386* **Per-user or per-group targeting**: every plugin key applies to every user who receives the settings. Server-managed settings deliver one configuration per organization. For per-group policy, use separate endpoint-managed settings or gateway policies, as under [Assign release channels to user groups](#assign-release-channels-to-user-groups).387* **Per-user or per-group targeting**: every plugin key applies to every user who receives the settings. For per-group policy, deliver each group its own settings, as under [Assign release channels to user groups](#assign-release-channels-to-user-groups).

387* **Restricting entries inside an allowed marketplace**: the allowlist matches marketplace sources. To block one plugin from an allowed marketplace, set it to `false` in managed `enabledPlugins`.388* **Restricting entries inside an allowed marketplace**: the allowlist matches marketplace sources. To block one plugin from an allowed marketplace, set it to `false` in managed `enabledPlugins`.

388* **Hiding `/plugin`**: no key disables the command. The nearest equivalent combines an allowlist naming only your marketplace, managed `enabledPlugins` entries for the plugins you supply, and `disableSideloadFlags`.389* **Hiding `/plugin`**: no key disables the command. The nearest equivalent combines an allowlist naming only your marketplace, managed `enabledPlugins` entries for the plugins you supply, and `disableSideloadFlags`.

389* **Gating `--plugin-dir` through the allowlist**: the allowlist doesn't cover `--plugin-dir`. `disableSideloadFlags` does.390* **Gating `--plugin-dir` through the allowlist**: the allowlist doesn't cover `--plugin-dir`. `disableSideloadFlags` does.

31>31>

32> **macOS, Linux, WSL:**32> **macOS, Linux, WSL:**

33>33>

34> ```bash theme={null}34> ```bash theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null}

35> curl -fsSL https://claude.ai/install.sh | bash35> curl -fsSL https://claude.ai/install.sh | bash

36> ```36> ```

37>37>

39>39>

40> **Windows PowerShell:**40> **Windows PowerShell:**

41>41>

42> ```powershell theme={null}42> ```powershell theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null}

43> irm https://claude.ai/install.ps1 | iex43> irm https://claude.ai/install.ps1 | iex

44> ```44> ```

45>45>

46> **Windows CMD:**46> **Windows CMD:**

47>47>

48> ```batch theme={null}48> ```batch theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null}

49> curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd49> curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

50> ```50> ```

51>51>

63 63 

64> [tab] Homebrew64> [tab] Homebrew

65>65>

66> ```bash theme={null}66> ```bash theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null}

67> brew install --cask claude-code67> brew install --cask claude-code

68> ```68> ```

69>69>

75 75 

76> [tab] WinGet76> [tab] WinGet

77>77>

78> ```powershell theme={null}78> ```powershell theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null}

79> winget install Anthropic.ClaudeCode79> winget install Anthropic.ClaudeCode

80> ```80> ```

81>81>

257 257 

258| Command | What it does | Example |258| Command | What it does | Example |

259| - | - | - |259| - | - | - |

260| `/clear` | Clear conversation history | `/clear` |260| `/clear` | Start a new conversation. Resume the previous one with `/resume`. | `/clear` |

261| `/help` | Show available commands | `/help` |261| `/help` | Show available commands | `/help` |

262| `/exit` or Ctrl+D twice | Exit Claude Code | `/exit` |262| `/exit` or Ctrl+D twice | Exit Claude Code | `/exit` |

263 263 

739 739 

740The sandbox doesn't run on native Windows, so with `failIfUnavailable` set, Claude Code exits at startup on those machines. If your fleet includes Windows hosts, you can:740The sandbox doesn't run on native Windows, so with `failIfUnavailable` set, Claude Code exits at startup on those machines. If your fleet includes Windows hosts, you can:

741 741 

742* **Deliver the configuration by operating system**: deploy it through your MDM or as a [managed settings file](https://code.claude.com/docs/en/managed-settings.md#delivery-mechanisms) on macOS and Linux machines only. [Server-managed settings](https://code.claude.com/docs/en/server-managed-settings.md#current-limitations) apply to all users in the organization742* **Deliver the configuration by operating system**: deploy it through your MDM or as a [managed settings file](https://code.claude.com/docs/en/managed-settings.md#delivery-mechanisms) on macOS and Linux machines only. [Server-managed settings](https://code.claude.com/docs/en/server-managed-settings.md#give-groups-different-settings) follow the member, not the device.

743* **Move Windows users into a supported environment**: have them run Claude Code inside WSL2 or a container743* **Move Windows users into a supported environment**: have them run Claude Code inside WSL2 or a container

744 744 

745### Keep developers from widening the policy745### Keep developers from widening the policy

229 229 

230## Disable session-scoped scheduled tasks230## Disable session-scoped scheduled tasks

231 231 

232Set `CLAUDE_CODE_DISABLE_CRON=1` in your environment to disable the scheduler entirely. The cron tools and `/loop` become unavailable, and any already-scheduled tasks stop firing. See [Environment variables](https://code.claude.com/docs/en/env-vars.md) for the full list of disable flags.232Set `CLAUDE_CODE_DISABLE_CRON=1` in your environment to disable the scheduler entirely. `CronCreate` and `/loop` become unavailable, and any already-scheduled tasks stop firing. While the project's `.claude/scheduled_tasks.json` file holds tasks, `CronList` and `CronDelete` stay available, so you can still ask Claude to list or cancel them. See [Environment variables](https://code.claude.com/docs/en/env-vars.md) for the full list of disable flags.

233 233 

234## Limitations234## Limitations

235 235 

183 183 

184* **A repository the session pushes results to**: the runner fails the session, and on a non-zero exit surfaces the tail of the script's stderr to the user. Write the reason to stderr as one short line, write it last, and then exit non-zero. Don't print secrets to stderr, and remove `set -x` before you deploy the hook.184* **A repository the session pushes results to**: the runner fails the session, and on a non-zero exit surfaces the tail of the script's stderr to the user. Write the reason to stderr as one short line, write it last, and then exit non-zero. Don't print secrets to stderr, and remove `set -x` before you deploy the hook.

185 185 

186 If git's `couldn't find remote ref` appears in roughly the last 4 KB of the hook's stderr, the runner reports a deleted branch instead. The session's error message then says the branch no longer exists on the remote and asks the user to restore it. If that branch exists, see [A session fails with `no longer exists on the remote` or `couldn't find remote ref`](https://code.claude.com/docs/en/self-hosted-environments-deploy#no-longer-exists-on-the-remote).186 If git's `couldn't find remote ref` appears in roughly the last 4 KB of the hook's stderr, the runner reports a missing branch instead. If that branch exists, see [A session fails with `couldn't be found in the remote repository` or `couldn't find remote ref`](https://code.claude.com/docs/en/self-hosted-environments-deploy#no-longer-exists-on-the-remote).

187* **A repository the session only reads from**, such as a repository added to a running session: the runner skips it and continues with the remaining repositories. The runner posts a `Skipped` step to the session and removes whatever the hook left at the checkout path.187* **A repository the session only reads from**, such as a repository added to a running session: the runner skips it and continues with the remaining repositories. The runner posts a `Skipped` step to the session and removes whatever the hook left at the checkout path.

188 188 

189 When the `Skipped` step mentions the checkout hook, find the failure detail on the `[runner:warn] checkout hook failed` line in the runner's log. When the step says a source ref `no longer exists on remote`, see [A session fails with `no longer exists on the remote` or `couldn't find remote ref`](https://code.claude.com/docs/en/self-hosted-environments-deploy#no-longer-exists-on-the-remote). If skipping leaves the session with no repository at all, the runner fails the session anyway.189 When the `Skipped` step mentions the checkout hook, find the failure detail on the `[runner:warn] checkout hook failed` line in the runner's log. When the step says a source ref `no longer exists on remote`, see [A session fails with `couldn't be found in the remote repository` or `couldn't find remote ref`](https://code.claude.com/docs/en/self-hosted-environments-deploy#no-longer-exists-on-the-remote). If skipping leaves the session with no repository at all, the runner fails the session anyway.

190 190 

191When the hook succeeds, the runner removes the checkout path after the session ends.191When the hook succeeds, the runner removes the checkout path after the session ends.

192 192 

916* **Runner logs `Poll failed` lines containing `rejecting the malformed poll response`**: the runner received a work-poll response whose body isn't the queue's expected JSON. One cause is an intercepting proxy or a captive portal answering with its own page. If the runner connects through a proxy, configure the proxy to pass responses from `api.anthropic.com` through unaltered.916* **Runner logs `Poll failed` lines containing `rejecting the malformed poll response`**: the runner received a work-poll response whose body isn't the queue's expected JSON. One cause is an intercepting proxy or a captive portal answering with its own page. If the runner connects through a proxy, configure the proxy to pass responses from `api.anthropic.com` through unaltered.

917 917 

918 The runner rejects the malformed response, counts it under the `transport` kind of the `claude_code_self_hosted_runner_poll_errors_total` [metric](https://code.claude.com/docs/en/self-hosted-environments-reference.md#prometheus-metrics), and retries on the failed-poll schedule described in [Session lifecycle](https://code.claude.com/docs/en/self-hosted-environments#session-lifecycle). The runner keeps serving its live sessions. Before v2.1.246, the runner read such a response as an empty work queue, which could end its live sessions or make it exit.918 The runner rejects the malformed response, counts it under the `transport` kind of the `claude_code_self_hosted_runner_poll_errors_total` [metric](https://code.claude.com/docs/en/self-hosted-environments-reference.md#prometheus-metrics), and retries on the failed-poll schedule described in [Session lifecycle](https://code.claude.com/docs/en/self-hosted-environments#session-lifecycle). The runner keeps serving its live sessions. Before v2.1.246, the runner read such a response as an empty work queue, which could end its live sessions or make it exit.

919* **A session fails with `no longer exists on the remote` or `couldn't find remote ref`**: the runner reads git's `couldn't find remote ref` as a deleted branch. The branch can still exist. Look for it on your git host, under exactly the name in the message, then check each cause that fits:919* **A session fails with `couldn't be found in the remote repository` or `couldn't find remote ref`**: look for the branch on your git host. Search for exactly the name in the message.

920 920 

921 * **The branch is gone**: restore it, as the message asks.921 The session's error includes `Retrying will fail the same way until you fix the missing branch described below.` When the error names the branch, the text under `Last output before exit:` reads:

922 

923 ```text theme={null}

924 The branch '<ref>' couldn't be found in the remote repository <repo>. If it was never pushed, push it. If it was deleted after a merge, restore it from its pull request. Then send a new message, or start a new session from a branch that has been pushed. (couldn't find remote ref)

925 ```

926 

927 * **The branch isn't on the host**: push it if it was never pushed, or restore it from its pull request if it was deleted after a merge.

922 * **The branch exists and you don't administer the runner**: send the runner's administrator the full message and tell them the branch exists.928 * **The branch exists and you don't administer the runner**: send the runner's administrator the full message and tell them the branch exists.

923 * **The branch exists and the runner has a `checkout` hook**: a git command in your hook can print `couldn't find remote ref` to stderr before the hook exits non-zero. Find that command, and correct the ref it asks for or the remote it fetches from. The runner passes your hook the revision the session asked for as [`CLAUDE_RUNNER_REPO_REF`](https://code.claude.com/docs/en/self-hosted-environments-configuration#checkout).929 * **The branch exists and the runner has a `checkout` hook**: a git command in your hook can print `couldn't find remote ref` to stderr before the hook exits non-zero. Find that command, and correct the ref it asks for or the remote it fetches from. The runner passes your hook the revision the session asked for as [`CLAUDE_RUNNER_REPO_REF`](https://code.claude.com/docs/en/self-hosted-environments-configuration#checkout).

924 * **The branch exists and the runner has no `checkout` hook**: the runner fetches the branch itself. If a [`--git-host-rewrite`](#rewrite-git-urls-for-private-networks) applies to the repository, confirm that the host it rewrites to has the branch under the same name, and push the branch there if it doesn't.930 * **The branch exists and the runner has no `checkout` hook**: the runner fetches the branch itself. If a [`--git-host-rewrite`](#rewrite-git-urls-for-private-networks) applies to the repository, confirm that the host it rewrites to has the branch under the same name, and push the branch there if it doesn't.

925 * **The branch exists and the runner is older than v2.1.301**: upgrade the runner as described in [Pin the version](#pin-the-version), so that a different failure shows its own error. Before v2.1.301, other failures with `couldn't find remote ref` in their error output were also reported as a deleted branch.931 * **The branch exists and the runner is older than v2.1.301**: upgrade the runner as described in [Pin the version](#pin-the-version), so that a different failure shows its own error. Before v2.1.301, other failures with `couldn't find remote ref` in their error output were also reported as a deleted branch.

926 932 

927 Once the runner is upgraded or the cause is fixed, retry the session or start a new one. If none of these checks finds the cause, [report an issue](#report-an-issue).933 Once the runner is upgraded or the cause is fixed, send the session a new message, or start a new session from a branch that has been pushed. If none of these checks finds the cause, [report an issue](#report-an-issue). Before v2.1.314, the message said the branch `no longer exists on the remote`. Follow the same checks for that message.

928 934 

929 For the source the session pushes results to, a deleted branch fails the session. For a git source the session only reads from, the runner skips that source and continues on the remaining ones. The runner fails the session when skipping would leave it with no repository at all. Before v2.1.228, such a session started in an empty directory.935 For the source the session pushes results to, a deleted branch fails the session. For a git source the session only reads from, the runner skips that source and continues on the remaining ones. The runner fails the session when skipping would leave it with no repository at all. Before v2.1.228, such a session started in an empty directory.

930* **Claude Tag sometimes doesn't post its reply in Slack**: the reply reminder is a [hook the control plane supplies](https://code.claude.com/docs/en/self-hosted-environments-configuration.md#how-each-session’s-config-is-assembled). Confirm the runner is Claude Code v2.1.229 or later with `claude --version`, since older runners ignore supplied hooks; confirm `python3 --version` succeeds inside the runner image, since the hooks run `python3`; and confirm no settings file turns hooks off, whether `disableAllHooks` in `managed-settings.json`, the runner user's `~/.claude/settings.json`, or the repository's `.claude/settings.json` or `.claude/settings.local.json`, or `allowManagedHooksOnly` or a [`strictPluginOnlyCustomization`](https://code.claude.com/docs/en/settings-reference#strictpluginonlycustomization) that covers hooks in managed settings.936* **Claude Tag sometimes doesn't post its reply in Slack**: the reply reminder is a [hook the control plane supplies](https://code.claude.com/docs/en/self-hosted-environments-configuration.md#how-each-session’s-config-is-assembled). Confirm the runner is Claude Code v2.1.229 or later with `claude --version`, since older runners ignore supplied hooks; confirm `python3 --version` succeeds inside the runner image, since the hooks run `python3`; and confirm no settings file turns hooks off, whether `disableAllHooks` in `managed-settings.json`, the runner user's `~/.claude/settings.json`, or the repository's `.claude/settings.json` or `.claude/settings.local.json`, or `allowManagedHooksOnly` or a [`strictPluginOnlyCustomization`](https://code.claude.com/docs/en/settings-reference#strictpluginonlycustomization) that covers hooks in managed settings.

125| `SELF_HOSTED_RUNNER_MAX_LIFETIME_GRACE_MS` | `900000` | How long the runner waits after a session reaches its `--kill-session-after-min` limit, for a running turn to finish or the release to complete, before it terminates the session |125| `SELF_HOSTED_RUNNER_MAX_LIFETIME_GRACE_MS` | `900000` | How long the runner waits after a session reaches its `--kill-session-after-min` limit, for a running turn to finish or the release to complete, before it terminates the session |

126| `SELF_HOSTED_RUNNER_POST_TURN_SETTLE_MS` | `7000` | Cap on how long the runner counts a session as busy for the `--drain-wait-sec` drain after a turn finishes, while the session's process reports the turn's end to Anthropic. `0` or an unusable value falls back to the default, so the hold can't be turned off. Requires Claude Code v2.1.275 or later. |126| `SELF_HOSTED_RUNNER_POST_TURN_SETTLE_MS` | `7000` | Cap on how long the runner counts a session as busy for the `--drain-wait-sec` drain after a turn finishes, while the session's process reports the turn's end to Anthropic. `0` or an unusable value falls back to the default, so the hold can't be turned off. Requires Claude Code v2.1.275 or later. |

127| `SELF_HOSTED_RUNNER_SIGKILL_GRACE_MS` | `30000` | How long the runner waits for the OS to deliver `SIGKILL` to a child stuck in uninterruptible I/O before exiting itself. Floored at `--post-session-hook-timeout-sec` plus 15 seconds, and 30 more when `--push-outcome-on-release` is set, so the effective minimum is 75 seconds at defaults. |127| `SELF_HOSTED_RUNNER_SIGKILL_GRACE_MS` | `30000` | How long the runner waits for the OS to deliver `SIGKILL` to a child stuck in uninterruptible I/O before exiting itself. Floored at `--post-session-hook-timeout-sec` plus 15 seconds, and 30 more when `--push-outcome-on-release` is set, so the effective minimum is 75 seconds at defaults. |

128| `CLAUDE_RUNNER_DISABLE_RENEWAL_WARNING` | unset | When `1`, the runner doesn't post `The server has refused to renew this session's access to the model` in a session. After each refusal, it waits until the next scheduled renewal to ask again. Requires Claude Code v2.1.319 or later. |

128| `CLAUDE_RUNNER_FETCH_DEPTH` | `50` | Git fetch depth for fresh clones. Set a positive integer, or `full` or `0` for a complete fetch. For a clone that's already in the workspace, see [Reuse a pre-warmed checkout](https://code.claude.com/docs/en/self-hosted-environments-deploy.md#reuse-a-pre-warmed-checkout). |129| `CLAUDE_RUNNER_FETCH_DEPTH` | `50` | Git fetch depth for fresh clones. Set a positive integer, or `full` or `0` for a complete fetch. For a clone that's already in the workspace, see [Reuse a pre-warmed checkout](https://code.claude.com/docs/en/self-hosted-environments-deploy.md#reuse-a-pre-warmed-checkout). |

129| `CLAUDE_RUNNER_FETCH_SERVER_PROGRESS_CAP_MS` | `600000` | How long in milliseconds, per attempt, a git fetch may wait for its first data while the git server's own progress numbers keep rising, as when the server prepares the pack for a large repository. `0` or `off` turns the wait off: such a fetch is then cut off after two minutes without data. Any other whole number is clamped to between `120000` and `1800000`, 2 to 30 minutes. Requires Claude Code v2.1.295 or later. |130| `CLAUDE_RUNNER_FETCH_SERVER_PROGRESS_CAP_MS` | `600000` | How long in milliseconds, per attempt, a git fetch may wait for its first data while the git server's own progress numbers keep rising, as when the server prepares the pack for a large repository. `0` or `off` turns the wait off: such a fetch is then cut off after two minutes without data. Any other whole number is clamped to between `120000` and `1800000`, 2 to 30 minutes. Requires Claude Code v2.1.295 or later. |

130| `CLAUDE_RUNNER_SKIP_GIT_VERIFY` | unset | When `1`, skip the `.git` presence check after a `checkout` hook runs. Set this when your hook materializes a non-git source. |131| `CLAUDE_RUNNER_SKIP_GIT_VERIFY` | unset | When `1`, skip the `.git` presence check after a `checkout` hook runs. Set this when your hook materializes a non-git source. |

108 108 

109In a running session, `/status` shows the same line after a failed fetch, and for some skipped-fetch causes, such as a third-party provider variable or a custom `ANTHROPIC_BASE_URL` exported in the user's shell.109In a running session, `/status` shows the same line after a failed fetch, and for some skipped-fetch causes, such as a third-party provider variable or a custom `ANTHROPIC_BASE_URL` exported in the user's shell.

110 110 

111### Give groups different settings

112 

113On an Enterprise plan, you can give each group of members its own configuration instead of one organization-wide configuration. If your organization has [HIPAA enabled](https://code.claude.com/docs/en/hipaa-setup.md), group configurations aren't available yet. Each group must already exist, either synced from your identity provider through SCIM or as a custom group you manage in the admin console.

114 

115Each member receives at most one configuration, the same on every device:

116 

117* **A member of a group with a configuration**: the configuration ranked highest among their groups

118* **Anyone else**, including a member whose only group with a configuration was deleted: the catch-all configuration, if you set one

119 

120Configurations don't merge with each other or with the organization-wide configuration, so each one must hold every setting you want that group to have. To apply a restriction to every member, put it in every group configuration and in the catch-all.

121 

122Members keep receiving the organization-wide configuration until the last of these steps, so you can set up and test everything first:

123 

124**1. Add a configuration for each group**

125 

126In [**Organization settings > Claude Code > Managed settings**](https://claude.ai/admin-settings/claude-code), click **Add configuration**, pick the group, and choose a starting point. To loosen policy for a group, pick a group that your identity provider syncs, so that no one can add themselves to it in the admin console.

127 

128To keep what members have today and change a few settings, start from a copy of your organization-wide configuration. Edit the configuration and save it.

129 

130**2. Set the catch-all**

131 

132Click **Set** on **Everyone else** and add the JSON for members who match no group configuration. To leave them with what they have today, paste in your organization-wide configuration. If you skip this step, their local sessions receive no server-managed settings, and any [endpoint-managed settings](https://code.claude.com/docs/en/managed-settings.md#delivery-mechanisms) on the device still apply.

133 

134**3. Rank the groups**

135 

136Drag the configurations into order. A member of several groups receives only the highest-ranked configuration, so put your most restrictive or most specific groups at the top.

137 

138**4. Check who gets what**

139 

140Under **Check effective settings** at the bottom of the page, search for a member to see which configuration they would receive. Check a member from each group, a member of several groups, and a member who matches no group configuration.

141 

142**5. Switch to group-based mode**

143 

144Click **Switch to group-based configurations** in the banner at the top of the page and confirm. Members receive their configurations within about an hour, apart from the [keys that apply only at the next launch](#fetch-and-caching-behavior). To go back, switch to organization-wide mode from the same banner: the organization-wide configuration applies again, if you have one, and your group configurations are kept.

145 

146A group's model limits hold only in local sessions. Cloud and remote Cowork sessions use the [`availableModels`](https://code.claude.com/docs/en/model-config#restrict-model-selection) list in the catch-all. If you set no catch-all, it comes from your organization-wide configuration, even after you switch to group-based mode. If you leave the key out of the catch-all, members aren't limited by this list in those sessions.

147 

148If you use a group's configuration to set where API requests go, that applies only to Claude Code, not to other tools that call the API.

149 

111### Access control150### Access control

112 151 

113The following roles can manage server-managed settings:152The following roles can manage server-managed settings:

115* **Primary Owner**154* **Primary Owner**

116* **Owner**155* **Owner**

117 156 

118Restrict access to trusted personnel, as settings changes apply to all users in the organization.157Restrict access to trusted personnel, as settings changes can apply to all users in the organization.

119 158 

120### Managed-only settings159### Managed-only settings

121 160 

125 164 

126Server-managed settings have the following limitations:165Server-managed settings have the following limitations:

127 166 

128* Settings apply uniformly to all users in the organization. Per-group configurations are not yet supported.167* On a Team plan, settings apply uniformly to all users in the organization. [Group configurations](#give-groups-different-settings) require an Enterprise plan.

129* You can't distribute a [`managed-mcp.json`](https://code.claude.com/docs/en/managed-mcp) file through server-managed settings. Deliver the `allowedMcpServers` and `deniedMcpServers` policy keys there instead. On Claude Code v2.1.259 or later, you can also provide remote servers with [`managedMcpServers`](https://code.claude.com/docs/en/managed-mcp#provide-servers-through-managed-settings), which accepts `http` and `sse` servers only and doesn't take exclusive control the way the file does.168* You can't distribute a [`managed-mcp.json`](https://code.claude.com/docs/en/managed-mcp) file through server-managed settings. Deliver the `allowedMcpServers` and `deniedMcpServers` policy keys there instead. On Claude Code v2.1.259 or later, you can also provide remote servers with [`managedMcpServers`](https://code.claude.com/docs/en/managed-mcp#provide-servers-through-managed-settings), which accepts `http` and `sse` servers only and doesn't take exclusive control the way the file does.

130 169 

131 Claude Code reads a `managed-mcp.json` deployed at its [system path](https://code.claude.com/docs/en/managed-mcp.md#exclusive-control-with-managed-mcp-json) separately from the managed settings tier, so the file still applies when server-managed settings are in effect.170 Claude Code reads a `managed-mcp.json` deployed at its [system path](https://code.claude.com/docs/en/managed-mcp.md#exclusive-control-with-managed-mcp-json) separately from the managed settings tier, so the file still applies when server-managed settings are in effect.

245* `cleanupPeriodDays` shortens retention of session transcripts and other local data to seven days245* `cleanupPeriodDays` shortens retention of session transcripts and other local data to seven days

246* `companyAnnouncements` shows a message at startup246* `companyAnnouncements` shows a message at startup

247 247 

248Administrators deploy a file like this as `managed-settings.json`, or the same JSON through MDM or [server-managed settings](https://code.claude.com/docs/en/server-managed-settings.md). One deployed file applies to every machine or account it reaches. To give a group different values, deploy a different file or profile to that group, since [server-managed settings don't support per-group policy yet](https://code.claude.com/docs/en/server-managed-settings.md#current-limitations).248Administrators deploy a file like this as `managed-settings.json`, or the same JSON through MDM or [server-managed settings](https://code.claude.com/docs/en/server-managed-settings.md). One deployed file applies to every machine or account it reaches. To give a group different values, deploy a different file or profile to that group. On an Enterprise plan, you can instead [give the group its own configuration](https://code.claude.com/docs/en/server-managed-settings.md#give-groups-different-settings) in server-managed settings.

249 249 

250> [tab] Copyable settings file250> [tab] Copyable settings file

251>251>

5530 5530 

5531Set how many days Claude Code keeps [session transcripts and other application data](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically) before deleting them. Claude Code runs the deletion as a background sweep after a session starts, as long as it can safely determine the retention period. The sweep deletes transcripts without showing a message, so a session you haven't used for longer than the retention period no longer appears in the [`/resume`](https://code.claude.com/docs/en/sessions#resume-a-session) picker.5531Set how many days Claude Code keeps [session transcripts and other application data](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically) before deleting them. Claude Code runs the deletion as a background sweep after a session starts, as long as it can safely determine the retention period. The sweep deletes transcripts without showing a message, so a session you haven't used for longer than the retention period no longer appears in the [`/resume`](https://code.claude.com/docs/en/sessions#resume-a-session) picker.

5532 5532 

5533By default, Claude Code keeps transcripts of sessions you started or most recently continued in Claude Desktop or Cowork past this period, even if you set this key yourself. This period [still applies to them in some cases](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically), such as under the HIPAA configuration. Before v2.1.248, Claude Code deleted them after this period.

5534 

5533* **Scope**: [`Any file`](#scopes)5535* **Scope**: [`Any file`](#scopes)

5534* **Type**: number of days, a whole number, minimum `1`5536* **Type**: number of days, a whole number, minimum `1`

5535* **Default**: `30`5537* **Default**: `30`

5544 5546 

5545### `desktopSessionCleanupPeriodDays`5547### `desktopSessionCleanupPeriodDays`

5546 5548 

5547Set an age limit in days for the transcripts of sessions you started or most recently continued in Claude Desktop or Cowork. Without this key, Claude Code [keeps those transcripts at any age](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically). Claude Code deletes each one once it's older than both this limit and [`cleanupPeriodDays`](#cleanupperioddays), so with `cleanupPeriodDays` at its default of 30, a value of `7` still keeps them 30 days. [Cleaned up automatically](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically) lists the cases where `cleanupPeriodDays` applies instead and Claude Code ignores this key. Requires Claude Code v2.1.248 or later.5549Set an age limit in days for the transcripts of sessions you started or most recently continued in Claude Desktop or Cowork. Without this key, Claude Code [keeps those transcripts at any age](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically), except in cases such as a session you deleted in Claude Desktop. Requires Claude Code v2.1.248 or later.

5550 

5551Claude Code deletes each transcript once it's older than both this limit and [`cleanupPeriodDays`](#cleanupperioddays), so with `cleanupPeriodDays` at its default of 30, a value of `7` still keeps them 30 days. [Cleaned up automatically](https://code.claude.com/docs/en/claude-directory.md#cleaned-up-automatically) lists cases where `cleanupPeriodDays` applies instead and Claude Code ignores this key.

5548 5552 

5549* **Scope**: [`User or managed`](#scopes). Claude Code also reads the key from a file you pass with `--settings`, and ignores it in project and local settings.5553* **Scope**: [`User or managed`](#scopes). Claude Code also reads the key from a file you pass with `--settings`, and ignores it in project and local settings.

5550* **Type**: number of days, a whole number, minimum `0`5554* **Type**: number of days, a whole number, minimum `0`

91| `managed-agents-secure [<agent-id>]` | Read the settings of one Managed Agent and report where they match the [Secure your Managed Agents deployment](https://platform.claude.com/docs/en/managed-agents/secure-deployment) guide or differ from it. Requires the [`ant` CLI](https://platform.claude.com/docs/en/cli-sdks-libraries/cli/quickstart) | v2.1.304 or later |91| `managed-agents-secure [<agent-id>]` | Read the settings of one Managed Agent and report where they match the [Secure your Managed Agents deployment](https://platform.claude.com/docs/en/managed-agents/secure-deployment) guide or differ from it. Requires the [`ant` CLI](https://platform.claude.com/docs/en/cli-sdks-libraries/cli/quickstart) | v2.1.304 or later |

92| `prompt-audit` | Flag instructions written for older models in your prompts, skills, and tool descriptions and propose fixes as a diff | v2.1.221 or later |92| `prompt-audit` | Flag instructions written for older models in your prompts, skills, and tool descriptions and propose fixes as a diff | v2.1.221 or later |

93| `cost-optimize` | Profile where your project's Claude API spend goes and propose savings from options such as prompt caching, trimming unneeded input and output tokens, batch processing, effort, and model choice, one change at a time | v2.1.247 or later |93| `cost-optimize` | Profile where your project's Claude API spend goes and propose savings from options such as prompt caching, trimming unneeded input and output tokens, batch processing, effort, and model choice, one change at a time | v2.1.247 or later |

94| `latency-optimize` | Find where the time goes in your existing Claude API code and propose changes that reduce how long people wait | v2.1.319 or later |

94| `build-eval` | Build an eval set for your Claude-powered app | v2.1.259 or later |95| `build-eval` | Build an eval set for your Claude-powered app | v2.1.259 or later |

95| `hillclimb` | Iteratively improve your app against an existing eval | v2.1.259 or later |96| `hillclimb` | Iteratively improve your app against an existing eval | v2.1.259 or later |

96| `preserved-thinking-migration` | Find the edits your integration makes to earlier turns, its system prompt, or its tool list that invalidate [preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking) blocks, measure how much reasoning each one drops, and propose fixes one at a time, re-measuring after each change | v2.1.282 or later |97| `preserved-thinking-migration` | Find the edits your integration makes to earlier turns, its system prompt, or its tool list that invalidate [preserved thinking](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking) blocks, measure how much reasoning each one drops, and propose fixes one at a time, re-measuring after each change | v2.1.282 or later |

197The time limit counts from the moment the command enters the background:197The time limit counts from the moment the command enters the background:

198 198 

199* A command that Claude starts in the background gets 30 minutes, or the `timeout` Claude passes with `run_in_background`, up to a maximum of 2 hours199* A command that Claude starts in the background gets 30 minutes, or the `timeout` Claude passes with `run_in_background`, up to a maximum of 2 hours

200* A command that starts in the foreground and then moves to the background, for example at its timeout, gets 30 minutes from the move200* A command that starts in the foreground and then moves to the background at its timeout gets 30 minutes from the move

201 201 

202In a run with the `-p` flag whose prompt you pass as text rather than with `--input-format stream-json`, both defaults are 10 minutes instead of 30, because the run [waits for background commands after its result](https://code.claude.com/docs/en/headless.md#background-tasks-at-exit).202In a run with the `-p` flag whose prompt you pass as text rather than with `--input-format stream-json`, both defaults are 10 minutes instead of 30, because the run [waits for background commands after its result](https://code.claude.com/docs/en/headless.md#background-tasks-at-exit).

203 203 

399 399 

400If the reload interrupted Claude mid-step, Claude continues that step when the conversation comes back, and a notice in the chat marks the continuation. Requires Claude Code v2.1.274 or later. If the step was interrupted more than an hour ago or the session is open elsewhere, the conversation comes back idle instead.400If the reload interrupted Claude mid-step, Claude continues that step when the conversation comes back, and a notice in the chat marks the continuation. Requires Claude Code v2.1.274 or later. If the step was interrupted more than an hour ago or the session is open elsewhere, the conversation comes back idle instead.

401 401 

402When a notice in the chat includes `plan mode is now off, so Claude did not continue on its own`, the step was interrupted in plan mode. To keep planning, select **Plan** in the mode indicator before you send your next message.

403 

402#### Turn off continuation after a reload404#### Turn off continuation after a reload

403 405 

404To stop Claude from continuing an interrupted step after a reload, open the [Continue After Reload setting](vscode://settings/claudeCode.continueAfterReload) and uncheck it.406To stop Claude from continuing an interrupted step after a reload, open the [Continue After Reload setting](vscode://settings/claudeCode.continueAfterReload) and uncheck it.

626 628 

627## VS Code commands and shortcuts629## VS Code commands and shortcuts

628 630 

629To open the Command Palette, press `Cmd+Shift+P` on Mac or `Ctrl+Shift+P` on Windows/Linux. Type "Claude Code" there to see all available VS Code commands for the Claude Code extension.631To open the Command Palette, press `Cmd+Shift+P` on Mac or `Ctrl+Shift+P` on Windows/Linux. Type "Claude Code" there to see the extension's Command Palette commands.

630 632 

631Some shortcuts depend on which panel is "focused" (receiving keyboard input). When your cursor is in a code file, the editor is focused. When your cursor is in Claude's prompt box, Claude is focused. Use `Cmd+Esc` / `Ctrl+Esc` to toggle between them.633Some shortcuts depend on which panel is "focused" (receiving keyboard input). When your cursor is in a code file, the editor is focused. When your cursor is in Claude's prompt box, Claude is focused. Use `Cmd+Esc` / `Ctrl+Esc` to toggle between them.

632 634 

651| Rename Session Tab | - | Rename the session in the active Claude tab. Requires Claude Code v2.1.257 or later |653| Rename Session Tab | - | Rename the session in the active Claude tab. Requires Claude Code v2.1.257 or later |

652| Add Session Tab to Group | - | Add the session in the active Claude tab to a [session group](#organize-sessions-into-groups) you pick or create. Requires Claude Code v2.1.257 or later |654| Add Session Tab to Group | - | Add the session in the active Claude tab to a [session group](#organize-sessions-into-groups) you pick or create. Requires Claude Code v2.1.257 or later |

653| Mark Session as Unread | - | Mark the session in the active Claude tab as unread in the sessions list. Requires Claude Code v2.1.257 or later |655| Mark Session as Unread | - | Mark the session in the active Claude tab as unread in the sessions list. Requires Claude Code v2.1.257 or later |

656| `claude-vscode.toggleDictation` | `Cmd+D` (Mac) / `Ctrl+D` (Windows/Linux) | Start or stop [voice dictation](https://code.claude.com/docs/en/voice-dictation.md). The shortcut requires Claude to be focused. This command isn't in the Command Palette |

654| Show Logs | - | View extension debug logs |657| Show Logs | - | View extension debug logs |

655| Logout | - | Sign out of your Anthropic account |658| Logout | - | Sign out of your Anthropic account |

656 659