HomeDocs changes
Documentation changes
What the docs quietly changed.
Anthropic's published documentation, fetched on a schedule and diffed against the last copy. Every edit dated, including the ones that never made a release note.
Last poll read all 4 cleanly: Claude Code CLI, Claude Developer Platform, Claude Documentation, Model Context Protocol. A date here is when this site noticed, not when Anthropic edited.
Claude Developer Platform
820pages tracked
773 changed this week · last read 23 Oct 2026
Claude Code CLI
228pages tracked
147 changed this week · last read 23 Oct 2026
Claude Documentation
280pages tracked
77 changed this week · last read 23 Oct 2026
Model Context Protocol
357pages tracked
171 changed this week · last read 23 Oct 2026
100 on this page
ReadSourcePageWhatLines
9 Sep 2026
129 pages
+2461 −1114 lines
claude-docs
government/config/settings
±changed
+1−1
claude-docschanged
government/config/settings
A list of hostnames that tools in Claude Desktop may reach, for example to install packages or fetch web pages. This covers the tools Claude uses during Cowork tasks, web fetch in
A list of hostnames that tools in Claude Desktop may reach, for example to install packages or fetch web pages. This covers the tools Claude uses during Cowork tasks, web fetch in
+1 −1 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-docs
government/changelog
±changed
+4−0
claude-docschanged
government/changelog
<Update label="2026.09.09.1">
* Changed how the "Telemetry headers" setting and your connectors are delivered to Claude Desktop, ahead of Claude Desktop retiring the older formats: nothing changes in your set
</Update>
+4 −0 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
vs-code
±changed
+11−2
claude-codechanged
vs-code
New sessions receive AI-generated titles based on your first message. Hover over a session to reveal rename and archive actions: rename to give it a descriptive title, or archive t
* **Session titles**: new sessions receive AI-generated titles based on your first message.
* **Rename and archive**: hover over a session to reveal these actions. Rename to give it a descriptive title, or archive to move it to the **Archived sessions** group at the botto
By default, a session with no activity for 14 days moves to **Archived sessions** automatically, unless it is open, unread, or in a [group](#organize-sessions-into-groups). Automat
+11 −2 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
settings-reference
±changed
+40−13
claude-codechanged
settings-reference
| [`maxEffortLevel`](#maxeffortlevel) | Cap the [effort level](/docs/en/model-config#adjust-effort-level) for ever
| [`modelSettings`](#modelsettings) | Keep a saved [effort level](/docs/en/model-config#adjust-effort-level) per
| [`modelSettings`](#modelsettings) | Keep a saved [effort level](/docs/en/model-config#adjust-effort-level) per
You don't usually edit this key by hand. Run `/advisor` to open a picker that shows the current choice, the models that can advise, and **No advisor**. Claude Code saves your pick
You don't usually edit this key by hand. Run `/advisor` to open a picker that shows the current choice, the models that can advise, and **No advisor**. Claude Code saves your pick
+40 −13 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
remote-control
±changed
+3−1
claude-codechanged
remote-control
* **Forwarded dialogs expire**: Claude Code keeps permission prompts and `AskUserQuestion` questions open until you answer them. When Claude Code forwards another kind of dialog to
* **Forwarded dialogs expire**: Claude Code keeps permission prompts and `AskUserQuestion` questions open until you answer them. When Claude Code forwards another kind of dialog to
* **The Fable usage-credits consent prompt isn't forwarded**: Claude Code shows the mid-session [Fable usage-credits consent prompt](/docs/en/model-config#fable-and-usage-credits)
* `/advisor`, from v2.1.260: pass the model as an argument, for example `/advisor opus`, or pass `off` to turn the advisor off. Both forms apply to the current session only and l
+3 −1 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
model-config
±changed
+8−6
claude-codechanged
model-config
Typing `/model <name>` directly behaves like `Enter`. A model set with `/model` in [non-interactive mode](/docs/en/headless), with the `-p` flag, applies to the current session onl
Typing `/model <name>` directly behaves like `Enter`. If you set a model with `/model` in [non-interactive mode](/docs/en/headless), with the `-p` flag, your choice applies to the
Your organization can cap the [effort level](#adjust-effort-level) in two ways. On a Claude Enterprise plan, organization admins set per-role effort limits, described below. On any
The plan-mode Opus phase uses the same context window as the `opus` model setting. On subscription tiers where Opus is [automatically upgraded to 1M context](#extended-context), `o
+8 −6 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
managed-settings
±changed
+1−0
claude-codechanged
managed-settings
* [`maxEffortLevel`](/docs/en/settings-reference#maxeffortlevel), where the lowest cap in any admin source applies. If a developer sets a lower cap in their own settings or with `-
+1 −0 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
env-vars
±changed
+1−1
claude-codechanged
env-vars
| `CLAUDE_CODE_EFFORT_LEVEL` | Set the effort level for supported models. Values: `low`, `medium`, `high`, `xhigh`, `max`, or `auto` to use the model d
| `CLAUDE_CODE_EFFORT_LEVEL` | Set the effort level for supported models. Values: `low`, `medium`, `high`, `xhigh`, `max`, or `auto` to use the model d
+1 −1 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
commands
±changed
+1−1
claude-codechanged
commands
| `/advisor [model\|off]` | Enable or disable the [advisor tool](/docs/en/advisor), which consu
| `/advisor [model\|off]` | Enable or disable the [advisor tool](/docs/en/advisor), which consu
+1 −1 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
advisor
±changed
+14−3
claude-codechanged
advisor
The command confirms with `Advisor set to` followed by the advisor model name. Your selection is saved to `advisorModel` in your user settings and persists across sessions.
The command confirms with `Advisor set to` followed by the advisor model name. Your selection is saved to `advisorModel` in your user settings and persists across sessions, except
The command also works where there is no terminal picker: in [non-interactive mode](/docs/en/headless) with `-p`, in the Agent SDK, in the desktop app, and over [Remote Control](/d
+14 −3 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-code
admin-setup
±changed
+8−1
claude-codechanged
admin-setup
| [Effort cap](/docs/en/settings-reference#maxeffortlevel) | Cap the [effort level](/docs/en/model-config#adjust-effort-level) for every model or
Organizations whose members authenticate through claude.ai or the Anthropic API can also govern models without deploying settings: [organization model restrictions](/docs/en/model-
If your members sign in through claude.ai or the Anthropic API and you're on a Claude Enterprise plan, you can also govern models from your organization's admin settings without de
* [Organization model restrictions](/docs/en/model-config#organization-model-restrictions): disable individual models. Enforced server-side.
+8 −1 lines
Recorded here 9 Sep 2026 21:07 UTC
claude-docs
third-party/claude-desktop/data-storage
±changed
+4−3
claude-docschanged
third-party/claude-desktop/data-storage
| Linux | `~/.config/Claude-3p/` | (under the application-data directory) |
| `claude-code-sessions/` | Code conversation history, in the same per-session layout.
| `claude-code-sessions/` | Code session records holding each session's working folder, settings, title, and sometimes a short summary. The con
To fully reset a device's Claude Desktop on 3P state, delete the application-data directory above and the `~/Claude/` user-files folder. To return to standard Claude Desktop withou
To fully reset a device's Claude Desktop on 3P state, delete the application-data directory above and the `~/Claude/` user-files folder. Code session transcripts are in Claude Code
+4 −3 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-docs
third-party/claude-desktop/connectors-m365
±changed
+3−3
claude-docschanged
third-party/claude-desktop/connectors-m365
* `MailboxSettings.Read` enables reading mail filters and automatic-reply settings.
* `MailboxSettings.Read` lets the connector read the user's mailbox time zone so that dates in requests follow the user's local time rather than UTC.
With only read scopes granted, the connector is read-only. To let Claude take actions in Microsoft 365 (sending mail, managing drafts, labels, filters, and calendar events, working
With only read scopes granted, the connector is read-only. To let Claude take actions in Microsoft 365 (sending mail, managing drafts, labels, and calendar events, working with fil
| `MailboxSettings.ReadWrite` | Manage labels, mail filters, and automatic replies |
+3 −3 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-docs
government/security/security-and-data-handling
±changed
+23−7
claude-docschanged
government/security/security-and-data-handling
The sandbox sees the workspace folders the user has attached to the session, its own scratch area, and read-only reference material bundled by the application (such as skill an
The sandbox sees the workspace folders the user has attached to the session, its own scratch area, and read-only reference material bundled by the application (such as skill an
By default, Claude for Government prompts the user for file writes outside the attached workspace, connector actions, and each web search. In Cowork, shell commands run without a p
By default, Claude for Government prompts the user for connector actions, for each web search, and, in Cowork, when Claude asks to add another folder to the session. In Chat and Co
Yes. Administrators can set each built-in and connector tool to **ask** (approval required on every call), **allow** (pre-approved), or **blocked** (removed entirely), and user
+23 −7 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-docs
government/deploy-desktop/configure
±changed
+21−12
claude-docschanged
government/deploy-desktop/configure
Use the host exactly as it was provided to your agency. An alias that your agency sets up on its own, such as a DNS record, redirect, or reverse proxy under your own domain that fo
With any alias of this kind, the app still offers **Sign in with your organization**, but sign-in fails as soon as the user chooses it; see [Troubleshooting](#troubleshooting). The
No other keys are needed to connect the app; Claude for Government supplies everything else per user after sign-in. If your agency distributes Claude Desktop updates itself, [Autom
+21 −12 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-docs
government/connectors/microsoft-365
±changed
+12−12
claude-docschanged
government/connectors/microsoft-365
| Permission | What it lets Claude do |
| --------------------------------- | --------------------------------------------------------------------------- |
| `Mail.Read` (Default) | Read the member's mail |
| `Mail.Read.Shared` (Default) | Read mail in mailboxes shared with the member |
| `Calendars.Read` (Default) | Read the member's calendar events |
+12 −12 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-docs
government/config/plugins-and-connectors
±changed
+1−1
claude-docschanged
government/config/plugins-and-connectors
On Claude Desktop, that tool policy shapes what members experience. A tool you switch off is blocked, so Claude cannot use it. A tool you switch on is available and asks the member
On Claude Desktop, that tool policy shapes what members experience. A tool you switch off is blocked, so Claude cannot use it. A tool you switch on is available and asks the member
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
sub-agents
±changed
+9−12
claude-codechanged
sub-agents
Set `permissionMode` to choose the permission mode a subagent runs in. Use the modes' config values, so Manual mode is `default`. If you leave it unset, the subagent inherits the m
Set `permissionMode` to choose the permission mode a subagent runs in. Use the modes' config values, so Manual mode is `default`. If you leave it unset, the subagent inherits the m
The main conversation's permission mode decides whether Claude Code uses the value you set:
+9 −12 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
settings-reference
±changed
+2−2
claude-codechanged
settings-reference
List the tool uses Claude Code blocks. Use it for files that hold API keys, secrets, or environment values: Claude Code excludes matching files from file discovery and search resul
List the tool uses Claude Code blocks. Use it for files that hold API keys, secrets, or environment values: Claude Code excludes matching files from file discovery and search resul
Tool names accept glob patterns, so `"*"` denies every tool and `"mcp__*"` denies every MCP tool. Claude Code ignores a deny rule for the [`EndConversation`](/docs/en/tools-referen
Tool names accept glob patterns, so `"*"` denies every tool and `"mcp__*"` denies every MCP tool. Claude Code ignores a deny rule for the [`EndConversation`](/docs/en/tools-referen
+2 −2 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
settings-example
±changed
+6−5
claude-codechanged
settings-example
One team's shared settings, committed to the repository so everyone who clones it gets the same permissions, hooks, telemetry, and plugin marketplace. Save a file like this at `.cl
One team's shared settings, committed to the repository so everyone who clones it gets the same permissions, hooks, telemetry, and plugin marketplace. Save a file like this at `.cl
* **Rules match the command and path as written.** `Bash(git push *)` doesn't match [`git -C . push`](/docs/en/permissions#bash-rule-limits). `Read(./.env)` on its own stops the fi
// Always confirm before pushing
// Confirm before git push commands
+6 −5 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
server-managed-settings
±changed
+1−1
claude-codechanged
server-managed-settings
This example enforces a permission deny list, prevents users from bypassing permissions, and restricts permission rules to those defined in managed settings:
This example enforces a permission deny list, prevents users from bypassing permissions, and restricts permission rules to those defined in managed settings. The `Bash(curl *)`
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
self-hosted-environments-reference
±changed
+11−7
claude-codechanged
self-hosted-environments-reference
| `--configure-git` | `SELF_HOSTED_RUNNER_CONFIGURE_GIT=1` | off | Write global git identity and enable Anthropic com
| `--configure-git` | `SELF_HOSTED_RUNNER_CONFIGURE_GIT=1` | off | At startup, write global git identity, enable Anth
| `--kill-session-after-min <n>` | `SELF_HOSTED_RUNNER_MAX_LIFETIME_MS` | `0` | Terminate a session child once it has lived N minu
| `--kill-session-after-min <n>` | `SELF_HOSTED_RUNNER_MAX_LIFETIME_MS` | `0` | Limit a session to N minutes wall-clock, as a safe
| `SELF_HOSTED_RUNNER_MAX_LIFETIME_GRACE_MS` | `900000` | Bounds how long a `--kill-session-after-min` kill is deferred while waiting for an in-flight turn to finish
+11 −7 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
self-hosted-environments-deploy
±changed
+16−4
claude-codechanged
self-hosted-environments-deploy
* `push.negotiate = true`, so git asks your git host which commits it already has before packing a push. Requires Claude Code v2.1.257 or later.
* `core.hooksPath` pointing at a runner-managed hooks directory. Its `commit-msg` and `prepare-commit-msg` hooks add a `Co-authored-by:` trailer for the session's creator to each c
* **For the first `n` minutes**: the runner serves its sessions normally and keeps enforcing `--startup-timeout-min` and `--kill-session-after-min`. If you also set [`--release-idl
* **For the first `n` minutes**: the runner serves its sessions normally and keeps enforcing `--startup-timeout-min` and `--kill-session-after-min`. If you also set [`--release-idl
Anthropic calls connector tools, such as GitHub, Slack, Linear, and the other claude.ai connectors, from its own infrastructure rather than from your runner, so when Claude uses a
+16 −4 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
self-hosted-environments-configuration
±changed
+9−7
claude-codechanged
self-hosted-environments-configuration
* `completed`: a clean exit, including a session archived or deleted while the child was still connected.
* `failed`: a child crash or a setup failure after spawn.
* `interrupted`: an idle release, startup timeout, server deassign, drain, or watchdog kill.
* `abandoned`: reserved for sessions another runner claimed; the hook doesn't currently fire in that case.
+9 −7 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
security
±changed
+1−1
claude-codechanged
security
* **Network command approval**: Commands that fetch content from the web such as `curl` and `wget` are not auto-approved by default. In Manual mode they prompt like any other non-r
* **Network command approval**: Commands that fetch content from the web such as `curl` and `wget` are not auto-approved by default. In Manual mode they prompt like any other non-r
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
remote-control
±changed
+1−4
claude-codechanged
remote-control
* **The message doesn't say to contact your organization admin**: your organization has a HIPAA configuration that is incompatible with Remote Control, and `/status` lists `HIPAA`
### "Remote Control isn't available for your organization due to its compliance policy"
Your organization has a data retention or compliance configuration that is incompatible with Remote Control; the parenthetical at the end of the message names it. In this state the
+1 −4 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
permissions
±changed
+20−4
claude-codechanged
permissions
Write the command you want Claude to run without asking, and replace the parts that vary with `*`. With this configuration, Claude Code runs npm scripts and git commits without ask
Write the command you want Claude to run without asking, and replace the parts that vary with `*`. With this configuration, Claude Code runs npm scripts and git commits without ask
Bash rules match the whole command text, with `*` standing in for any text. [Wildcard patterns](#wildcard-patterns) shows which commands each rule shape matches and where to put th
Bash rules match the whole command text, with `*` standing in for any text. [Wildcard patterns](#wildcard-patterns) shows which commands each rule shape matches and where to put th
+20 −4 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
permission-modes
±changed
+2−2
claude-codechanged
permission-modes
* Pushing to any branch of the repository you're working in, including the default branch. A non-default branch whose name marks it as a deploy or publication target, such as `prod
* Pushing to any branch of the repository you're working in, including the default branch. A non-default branch whose name marks it as a deploy or publication target, such as `prod
Pushing to any branch of the repository you're working in and creating a pull request that matches your request run without a prompt, unless the push or pull request falls under th
Pushing to any branch of the repository you're working in and creating a pull request that matches your request run without a prompt, unless the push or pull request falls under th
+2 −2 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
mcp
±changed
+7−3
claude-codechanged
mcp
Some services still expose only an SSE endpoint. Use the same command as the HTTP transport, with `--transport sse`:
Some services still expose only an SSE endpoint. Add these with the same `claude mcp add --transport http <name> <url>` command as [an HTTP server](#option-1-add-a-remote-http-serv
On an earlier version, or to connect over SSE directly, pass `--transport sse` instead:
A URL means the server is remote. For an `https://` endpoint, add it with `--transport http`, or with `--transport sse` when the instructions say the endpoint uses SSE. For a `wss:
+7 −3 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
managed-settings
±changed
+5−0
claude-codechanged
managed-settings
| `allowedHttpHookUrls` | Claude Code enforces an empty [allowlist](/docs/en/settings-reference#allowedhttphookurls) until you fix the value. If only an individual entry is
| `httpHookAllowedEnvVars` | Claude Code enforces an empty [allowlist](/docs/en/settings-reference#httphookallowedenvvars) until you fix the value. If only an individual entry
| `allowedChannelPlugins` | Claude Code enforces an empty allowlist until you fix the value, so no channel plugin passed to `--channels` is admitted. If only an individual en
Before v2.1.267, Claude Code dropped `allowedHttpHookUrls`, `httpHookAllowedEnvVars`, or `allowedChannelPlugins` whole when the value or any one entry was invalid, and behaved as i
+5 −0 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
hooks-guide
±changed
+1−1
claude-codechanged
hooks-guide
| `StopFailure` | error type
| `StopFailure` | error type
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
hooks
±changed
+4−2
claude-codechanged
hooks
| `StopFailure` | error type
| `StopFailure` | error type
Matching `StopFailure` on `cloud_credential_error` requires Claude Code v2.1.267 or later, the first version that reports credential-load failures under that value rather than `ser
| `error` | Error type: `rate_limit`, `overloaded`, `authentication_failed`, `oauth_org_not_allowed`, `account_on_hold`, `billing_error`, `invalid_request`, `model
+4 −2 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
headless
±changed
+3−3
claude-codechanged
headless
| `max_retries` | integer | total retries permitted
| `max_retries` | integer | total retries permitted for this failure's cause, which can be fewer than the session-wide budget
| `error_status` | integer or null | HTTP status code, or `null` for connection errors with no HTTP response
| `error_status` | integer or null | HTTP status code of the failed attempt, or `null` when the attempt got no HTTP response from the API
| `error` | string | error category: `authentication_failed`, `oauth_org_not_allowed`, `billing_error`, `rate_limit`, `overloaded`, `invalid_request`, `model_not
+3 −3 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
errors
±changed
+177−15
claude-codechanged
errors
| `Could not load AWS credentials` / `Could not load Google Cloud credentials`
| `Could not load the default credentials` on Google Cloud's Agent Platform
| `Timed out after 60s waiting for AWS`
| `A request to AWS timed out. Check your network and proxy settings, then try again.`
| `Could not load the default credentials` on Google Cloud's Agent Platform
+177 −15 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
debug-your-config
±changed
+1−1
claude-codechanged
debug-your-config
| `Bash(rm *)` deny rule doesn't block `/bin/rm` or `find -delete` | Prefix rules match the literal command string, not the underlying executable
| `Bash(rm *)` deny rule doesn't block `/bin/rm` or `find -delete` | Bash rules match the literal command string, not the underlying executable; see [what a Bash rule doesn't m
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
cli-reference
±changed
+1−1
claude-codechanged
cli-reference
| `--disallowedTools`, `--disallowed-tools` | Deny rules. A bare tool name removes the matching tools from Claude's context: `"Edit"` removes Edit, `"*"` removes every tool,
| `--disallowedTools`, `--disallowed-tools` | Deny rules. A bare tool name removes the matching tools from Claude's context: `"Edit"` removes Edit, `"*"` removes every tool,
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
changelog
±changed
+56−0
claude-codechanged
changelog
<Update label="2.1.267" description="September 9, 2026">
* Added `maxEffortLevel` setting (top-level or per model under `modelSettings`): caps the effort level on every provider, including Bedrock, Vertex and Foundry; users can still p
* Added `--system-prompt-snapshot off` to render the system prompt fresh on every request instead of reusing the conversation's recorded prompt (for iterating on prompt text)
* Fixed Cowork scheduled tasks in the cloud failing at startup for organizations whose managed settings require sandboxing
+56 −0 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
auto-mode-config
±changed
+4−2
claude-codechanged
auto-mode-config
If you want a human checkpoint before every push or pull request, add permission rules: the [recipes below](#add-a-human-checkpoint) keep auto mode on for everything else.
If you want a human checkpoint before Claude's push and pull request commands, add permission rules: the [recipes below](#add-a-human-checkpoint) keep auto mode on for everything e
These rules match commands that begin with `git push` or `gh pr create`. A push Claude writes another way, such as `git -C <dir> push` or `git -c <key>=<value> push`, [doesn't matc
| Prompt before the action | `permissions.ask` | Always prompts for content-scoped rules like the recipe above. The classifier can
+4 −2 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
amazon-bedrock
±changed
+17−0
claude-codechanged
amazon-bedrock
Except when you authenticate with an Amazon Bedrock API key, the [setup wizard](#sign-in-with-bedrock) applies the same limit to each AWS call it makes while verifying your credent
### Certificate errors behind a TLS-inspecting proxy
Claude Code applies your [CA certificate store](/docs/en/network-config#ca-certificate-store) configuration to its requests to AWS, including:
+17 −0 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
agent-sdk/typescript
±changed
+11−6
claude-codechanged
agent-sdk/typescript
| `disallowedTools` | `string[]` | `[]`
| `disallowedTools` | `string[]` | `[]`
| `permissionMode` | No | Permission mode for tool execution within this agent. See [`PermissionMode`](#permissionmode)
| `permissionMode` | No | Permission mode for tool execution within this agent. The [subagent inheritance rules](/docs/en/agent-sdk/permissions#available
`SDKAssistantMessageError` is one of: `'authentication_failed'`, `'oauth_org_not_allowed'`, `'account_on_hold'`, `'billing_error'`, `'rate_limit'`, `'overloaded'`, `'invalid_reques
+11 −6 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
agent-sdk/subagents
±changed
+1−1
claude-codechanged
agent-sdk/subagents
| `permissionMode` | `PermissionMode` | No | Permission mode for tool execution within this agent
| `permissionMode` | `PermissionMode` | No | Permission mode for tool execution within this agent. The [subagent inheritance rules
+1 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
agent-sdk/python
±changed
+3−3
claude-codechanged
agent-sdk/python
| `disallowed_tools` | `list[str]` | `[]` | Tools to deny. A bare
| `disallowed_tools` | `list[str]` | `[]` | Tools to deny. A bare
| `permissionMode` | No | Permission mode for tool execution within this agent. See [`PermissionMode`](#permissionmode)
| `permissionMode` | No | Permission mode for tool execution within this agent. The [subagent inheritance rules](/docs/en/agent-sdk/permissions#available-modes) decide when
"mode": "acceptEdits" | "auto" | "bypassPermissions" | "default" | "dontAsk" | "plan" | None, # Deprecated; ignored. Subagents inherit the parent session's permission mode; ag
+3 −3 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
agent-sdk/permissions
±changed
+7−7
claude-codechanged
agent-sdk/permissions
| Option | Effect
| :-------------------------------- | :---------------------------------------------------------------------------------------------------------------------------------------------
| `allowed_tools=["Read", "Grep"]` | `Read` and `Grep` are auto-approved. Other tools not listed here still exist and fall through to the permission mode and `canUseTool`.
| `disallowed_tools=["Bash"]` | The `Bash` tool definition is removed from the request. Claude does not see the tool and cannot attempt it.
| `disallowed_tools=["Bash(rm *)"]` | `Bash` stays available. Calls matching `rm *` are denied in every permission mode, including `bypassPermissions`. Other `Bash` calls fall thro
+7 −7 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
agent-sdk/mcp
±changed
+3−1
claude-codechanged
agent-sdk/mcp
The SDK applies the same MCP output limit as Claude Code. When a tool result is larger than 25,000 tokens, the full output is saved to a file and the tool result is replaced with a
The SDK applies the same MCP output limit as Claude Code. When a tool result with no image content is larger than 25,000 tokens, Claude Code saves the output to a file and replaces
Raise the limit with the [`MAX_MCP_OUTPUT_TOKENS`](/docs/en/env-vars) environment variable. See [MCP output limits and warnings](/docs/en/mcp#mcp-output-limits-and-warnings) for th
+3 −1 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
agent-sdk/custom-tools
±changed
+6−6
claude-codechanged
agent-sdk/custom-tools
| Option | Layer | Effect
| :------------------------ | :----------- | :--------------------------------------------------------------------------------------------------------------------------------------
| `tools: ["Read", "Grep"]` | Availability | Only the listed built-ins are in Claude's context. Unlisted built-ins are removed. MCP tools are unaffected.
| `tools: []` | Availability | All built-ins are removed. Claude can only use your MCP tools.
| allowed tools | Permission | Listed tools run without a permission prompt. Other unlisted tools remain available; calls go through the [permission flow](/docs/en/ag
+6 −6 lines
Recorded here 9 Sep 2026 20:37 UTC
claude-code
interactive-mode
±changed
+18−0
claude-codechanged
interactive-mode
## Issue reference links
When Claude mentions an issue as `owner/repo#123`, you can click the reference to open it, as long as your terminal supports hyperlinks. If Claude Code doesn't detect hyperlink sup
You get a link only for the two-part `owner/repo#123` form. These stay plain text:
+18 −0 lines
Recorded here 9 Sep 2026 20:07 UTC
claude-code
agent-sdk/streaming-output
±changed
+8−4
claude-codechanged
agent-sdk/streaming-output
By default, the Agent SDK yields complete `AssistantMessage` objects after Claude finishes generating each response. To receive incremental updates as text and tool calls are gener
By default, the Agent SDK yields a complete `AssistantMessage` for each non-empty content block, such as a text block or a tool call, after Claude finishes generating that block. T
user_message_uuid?: string;
Claude Code sets `user_message_uuid` on the turn's first non-ping stream event, and again when the message the turn is answering changes, under the conditions in [`user_message_uui
+8 −4 lines
Recorded here 9 Sep 2026 20:07 UTC
claude-code
agent-sdk/agent-loop
±changed
+19−7
claude-codechanged
agent-sdk/agent-loop
2. **Evaluate and respond.** Claude evaluates the current state and determines how to proceed. It may respond with text, request one or more tool calls, or both. The SDK yields an
2. **Evaluate and respond.** Claude evaluates the current state and determines how to proceed. It may respond with text, request one or more tool calls, or both. The SDK yields one
2. **Turn 2:** Claude calls `Read` on `auth.ts` and `auth.test.ts`. The SDK returns the file contents and yields an `AssistantMessage`.
3. **Turn 3:** Claude calls `Edit` to fix `auth.ts`, then calls `Bash` to re-run `npm test`. All three tests pass. The SDK yields an `AssistantMessage`.
2. **Turn 2:** Claude calls `Read` on `auth.ts` and `auth.test.ts`. The SDK yields an `AssistantMessage` for each call and returns the file contents.
+19 −7 lines
Recorded here 9 Sep 2026 20:07 UTC
claude-code
workflows
±changed
+10−2
claude-codechanged
workflows
| `Enter` or `→` | Drill into the selected phase, then into an agent to read its prompt, recent tool calls, and result |
| `Enter` or `→` | Drill into the selected phase, then into an agent's detail. In the detail, `Enter` expands or collapses it |
The agent detail lists the agent's prompt, its recent tool calls, and its result. Each call shows its state, such as still running or failed. When the agent keeps a task list of it
+10 −2 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
tools-reference
±changed
+2−1
claude-codechanged
tools-reference
* When Claude runs `cd` in the main session, the new working directory carries over to later Bash commands as long as it stays inside the project directory or an [additional workin
* When Claude runs `cd` in the main session, the new working directory carries over to later Bash commands as long as it stays inside the project directory or an [additional workin
* Subagent sessions never carry over working directory changes.
+2 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
sub-agents
±changed
+7−3
claude-codechanged
sub-agents
The second filter applies to subagents running in the background. Apart from `Agent` and `ExitPlanMode`, which follow the first filter's conditions wherever the subagent runs, a ba
The second filter applies to subagents running in the background. Apart from `Agent` and `ExitPlanMode`, which follow the first filter's conditions wherever the subagent runs, a ba
[`ListAgents`](/docs/en/cross-session-messaging) follows these filters like any built-in tool: a foreground subagent inherits it in sessions where cross-session messaging is enable
Background subagents run with a [smaller built-in tool set](#available-tools) than foreground subagents, except for conversation forks, and they surface every permission prompt in
+7 −3 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
settings-reference
±changed
+9−5
claude-codechanged
settings-reference
Set the deadline for dialogs Claude Code [forwards to a remote client](/docs/en/remote-control#limitations), such as a Remote Control or SDK host, for the approval dialog for a [he
Set the deadline for dialogs Claude Code [forwards to a remote client](/docs/en/remote-control#limitations), such as a Remote Control or SDK host, and for the approval dialog for a
Show or hide the turn duration message after each response, such as "Cooked for 1m 6s". Appears in `/config` as **Show turn duration**.
Show or hide the turn duration message after each response, such as "Cooked for 1m 6s · done 6:05 PM". The clock after "done" shows when the turn finished; [`timeFormat`](#timeform
Some terminals can show a progress indicator on the tab or in the taskbar for the program running in them. While Claude is working, Claude Code reports an in-progress state to the
+9 −5 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
sessions
±changed
+1−1
claude-codechanged
sessions
* Conversation history: the full history, including tool calls and results.
* Conversation history: the full history, including tool calls and results. A tool that was still running when the previous process ended, for example in a crash, doesn't finish or
+1 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
prompt-caching
±changed
+3−1
claude-codechanged
prompt-caching
Claude Code applies a plugin change when you run [`/reload-plugins`](/docs/en/discover-plugins#apply-plugin-changes-without-restarting) or start a new session. You pay the cost, wh
A plugin change applies when you run [`/reload-plugins`](/docs/en/discover-plugins#apply-plugin-changes-without-restarting) or start a new session, not when you run `/plugin enable
* In interactive sessions, when you add or remove a plugin in a [folder of plugins](/docs/en/plugins#test-your-plugins-locally) you passed with `--plugin-dir`, the change applies r
* **Resumed subagents**: when Claude [resumes a subagent](/docs/en/sub-agents#resume-subagents), the resumed run's first request can read the cache the original run warmed.
+3 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
plugins
±changed
+6−1
claude-codechanged
plugins
The plugin root is the individual plugin's own directory: the one you pass to `--plugin-dir` or that contains `.claude-plugin/plugin.json`. It is never `~/.claude/`. For example,
The plugin root is the individual plugin's own directory, such as `my-first-plugin/` from the [quickstart](#quickstart). It is never `~/.claude/`. For example, Claude Code doesn'
To load several plugins from one place, pass a folder that holds them, such as `--plugin-dir ./plugins`. Loading a folder of plugins requires Claude Code v2.1.265 or later. Claude
* **What loads**: if the folder has no manifest or plugin components at its top level, Claude Code treats it as a folder of plugins. Each immediate subfolder that has a `.claude-pl
+6 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
plugin-dependencies
±changed
+2−0
claude-codechanged
plugin-dependencies
When both plugins sit in one parent folder, you can pass that folder to `--plugin-dir` once. If the folder isn't itself a plugin, Claude Code loads each child folder that has a `.c
+2 −0 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
permissions
±changed
+6−7
claude-codechanged
permissions
The label shown for a tool in the transcript and permission dialog can differ from its canonical name. For example, the tool labeled `Stop Task` in the transcript has the canonical
The label shown for a tool in the transcript and permission dialog can differ from its canonical name. For example, the tool labeled `Stop Task` in the transcript has the canonical
* Extra spaces: `curl http://github.com`
| Rule | Matches | Does not match |
| --------------------- | ----------------------------------------- | ---------------------------- |
+6 −7 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
permission-modes
±changed
+1−1
claude-codechanged
permission-modes
Each interface has its own control for switching permission modes during a session and its own way of choosing the permission mode new sessions start in. Asking Claude in chat to c
Each interface has its own control for switching permission modes during a session and its own way of choosing the permission mode new sessions start in. Select your interface to s
+1 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
large-codebases
±changed
+9−9
claude-codechanged
large-codebases
* **Yourself only**: use `.claude/settings.local.json` at the repository root, which loads in every CLI session inside the repository regardless of starting directory, except in th
* **Yourself only**: use `.claude/settings.local.json` at the repository root, which loads in every CLI session inside the repository regardless of starting directory, except in th
The example below blocks build artifacts and a vendored SDK:
The example below blocks build artifacts and a vendored SDK. Its directory patterns end with `/**/*` rather than `/**` so that each rule covers everything inside the directory but
"Read(./**/dist/**)",
+9 −9 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
keybindings
±changed
+19−17
claude-codechanged
keybindings
| Action | Default | Description
| :-------------------- | :-------------------------------- | :---------------------------------------------------------------------------------------------------------------------
| `chat:cancel` | Escape | Cancel current input
| `chat:clearInput` | Ctrl+L | Force a full screen redraw, preserving input and conversation. In [fullscreen rendering](/docs/en/fullscreen#clear-the
| `chat:clearScreen` | Cmd+K | Same as `chat:clearInput`. See [Clear the conversation](/docs/en/fullscreen#clear-the-conversation) for how Cmd+K beha
+19 −17 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
interactive-mode
±changed
+13−0
claude-codechanged
interactive-mode
### Complete a command mid-prompt
Command completion also works partway through a prompt: type `/` after a space, then the first letters of a name, as in `run the tests, then /com`. Only commands whose names start
* **In [fullscreen rendering](/docs/en/fullscreen)**: the matches open as a list while you type, with no row highlighted, so `Enter` still sends your prompt as typed. Press `Tab` t
+13 −0 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
hooks
±changed
+8−2
claude-codechanged
hooks
When several hooks return `additionalContext` for the same event, Claude receives all of the values. If a value exceeds 10,000 characters, Claude Code writes the full text to a fil
When several hooks return `additionalContext` for the same event, Claude receives all of the values.
If a value exceeds 10,000 characters, Claude Code writes the text to a file in the session directory and passes Claude the file path with a short preview instead.
+8 −2 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
headless
±changed
+6−1
claude-codechanged
headless
By default, Claude Code emits only subagent `tool_use` and `tool_result` blocks. Pass [`--forward-subagent-text`](/docs/en/cli-reference#cli-flags) or set [`CLAUDE_CODE_FORWARD_SUB
The first message from a subagent running in the [foreground](/docs/en/sub-agents#run-subagents-in-foreground-or-background) is a `user` message carrying the prompt that drives it.
* **By default**: the subagent's `tool_use` and `tool_result` blocks.
* **With [`--forward-subagent-text`](/docs/en/cli-reference#cli-flags) or [`CLAUDE_CODE_FORWARD_SUBAGENT_TEXT`](/docs/en/env-vars)**: the subagent's text and thinking blocks too, s
+6 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
env-vars
±changed
+8−6
claude-codechanged
env-vars
| `CLAUDE_CODE_ARTIFACT_AUTO_OPEN` | Set to `0` to stop Claude Code from opening the browser automatically when a new [artifact](/docs/en/artifacts) is publi
| `CLAUDE_CODE_ARTIFACT_AUTO_OPEN` | Set to `0` to stop Claude Code from opening the browser automatically when a new [artifact](/docs/en/artifacts#create-an
| `CLAUDE_CODE_DISABLE_THINKING` | Set to `1` to omit the `thinking` parameter from API requests entirely. This is a compatibility option for proxies and g
| `CLAUDE_CODE_DISABLE_THINKING` | Set to `1` to omit the `thinking` parameter from API requests entirely. This is a compatibility option for proxies and g
| `CLAUDE_CODE_NONBLOCKING_STDOUT` | Set to `1` to write terminal output through a second non-blocking file descriptor, so a terminal that stops reading, suc
+8 −6 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
commands
±changed
+2−2
claude-codechanged
commands
| `/import [codex\|gemini] [--dry-run] [--yes]` | Bring configuration from other coding agents on your machine, curre
| `/init` | Initialize project with a `CLAUDE.md` guide. Set `CLAUDE_CODE_NEW_I
| `/import [codex\|gemini\|cursor] [--dry-run] [--yes]` | Bring configuration from OpenAI Codex, Google Gemini CLI, or Cursor
| `/init` | Initialize project with a `CLAUDE.md` guide. Set `CLAUDE_CODE_NEW_I
+2 −2 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
cli-reference
±changed
+5−5
claude-codechanged
cli-reference
| `claude import [codex\|gemini]` | Start an interactive session that runs [`/import`](/docs/en/commands#all-commands) to bring configuration from other coding agents into Claude C
| `claude import [source]` | Start an interactive session that runs [`/import`](/docs/en/commands#all-commands) to bring configuration from other coding agents into Claude C
| `claude rm <id>` | Remove a [background session](/docs/en/agent-view#manage-sessions-from-the-shell) from the list. The conversation transcript stays on your local
| `claude rm <id>` | Remove a [background session](/docs/en/agent-view#manage-sessions-from-the-shell) from the list. Accepts `--discard-unpushed <commit>@<worktree-
| `--bg`, `--background` | Start the session as a [background agent](/docs/en/agent-view) and return immediately. Prints the session ID and management comm
+5 −5 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
artifacts
±changed
+25−1
claude-codechanged
artifacts
* Claude declares a runtime capability for the page, such as [connector calls](#pull-live-data-with-mcp-connectors)
* Claude declares a runtime capability for the page, such as [connector calls](#pull-live-data-with-mcp-connectors) or [file downloads](#offer-a-file-download)
## Read an artifact shared with you
+25 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
agent-view
±changed
+1−1
claude-codechanged
agent-view
| `~/.claude/jobs/<id>/tmp/` | Per-session scratch directory. Writes here don't prompt for permission. Removed when the session is deleted
| `~/.claude/jobs/<id>/tmp/` | Per-session scratch directory. Claude's `Write` and `Edit` calls here don't prompt for permission. Removed when the session is deleted
+1 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
agent-teams
±changed
+1−1
claude-codechanged
agent-teams
[`CLAUDE_CODE_SUBAGENT_MODEL_FORCE`](/docs/en/sub-agents#run-every-subagent-on-one-model) applies to teammates as well as to subagents.
If you set [`CLAUDE_CODE_SUBAGENT_MODEL_FORCE=1`](/docs/en/sub-agents#run-every-subagent-on-one-model), the first two sources don't apply. Claude Code picks every teammate's model
+1 −1 lines
Recorded here 9 Sep 2026 19:37 UTC
claude-code
settings-reference
±changed
+20−13
claude-codechanged
settings-reference
See [Ask before switching](/docs/en/model-config#ask-before-switching). Requires Claude Code v2.1.170 or later.
See [Ask before switching](/docs/en/model-config#ask-before-switching).
Under `"merge"`, Claude Code combines each key by its kind. This table gives the rule for each kind; the restriction allowlist and highest-source-only rows name every key they cove
| Kind of key | How Claude Code combines it
+20 −13 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
server-managed-settings
±changed
+20−5
claude-codechanged
server-managed-settings
Server-managed settings let organization Owners centrally configure Claude Code from [**Admin Settings > Claude Code > Managed settings**](https://claude.ai/admin-settings/claude-c
Server-managed settings let organization Owners centrally configure Claude Code from [**Admin Settings > Claude Code > Managed settings**](https://claude.ai/admin-settings/claude-c
If a later fetch finds the server-managed settings removed, Claude Code runs that helper right away instead of at the next launch. The [`policyHelper`](/docs/en/settings-reference#
Every other key in the cached `env` block, such as telemetry and OpenTelemetry configuration, applies at startup as before. Once the server confirms the payload, and you approve it
+20 −5 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
permissions
±changed
+1−6
claude-codechanged
permissions
Use both for defense-in-depth:
* Permission deny rules block Claude from even attempting to access restricted resources
* Sandbox restrictions prevent Bash commands from reaching resources outside defined boundaries, even if a prompt injection bypasses Claude's decision-making
* Filesystem restrictions in the sandbox combine the [`sandbox.filesystem`](/docs/en/sandboxing) settings with Read and Edit deny rules; both are merged into the final sandbox boun
+1 −6 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
managed-settings
±changed
+5−4
claude-codechanged
managed-settings
* **Running sessions**: a session picks up most changes on the schedule in the table without a restart. Claude Code reads [`forceRemoteSettingsRefresh`](/docs/en/settings-reference
* **Running sessions**: most changes reach a running session on the schedule in the [delivery mechanism table](#choose-a-delivery-mechanism), without a restart.
* Changes to [`forceRemoteSettingsRefresh`](/docs/en/settings-reference#forceremotesettingsrefresh), [`requiredMinimumVersion`](/docs/en/settings-reference#requiredminimumversion
* A new or changed [`policyHelper`](/docs/en/settings-reference#policyhelper) entry takes effect at the next launch, except that a helper shadowed at launch by server-managed set
This table shows how Claude Code combines each kind of key under `"merge"`; the [`managedSourcesBehavior` entry](/docs/en/settings-reference#managedsourcesbehavior) names every key
+5 −4 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
hooks
±changed
+1−8
claude-codechanged
hooks
Set `type` to `"agent"` and provide a `prompt` string. The configuration fields are the same as [prompt hooks](#prompt-hook-configuration), except that agent hooks have a longer de
| Field | Required | Description |
| :-------- | :------- | :------------------------------------------------------------------------------------------ |
| `type` | yes | Must be `"agent"` |
+1 −8 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
errors
±changed
+39−0
claude-codechanged
errors
| `Cloud gateway session expired — run /login to reconnect.`
| `Cloud gateway <url> no longer accepts this session`
| `Managed settings were not approved; exiting without applying them.`
### Cloud gateway session expired
+39 −0 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
claude-apps-gateway-deploy
±changed
+5−2
claude-codechanged
claude-apps-gateway-deploy
Don't redirect requests to the gateway's device-authorization and token endpoints. Claude Code doesn't follow redirects on those requests, so an ingress rule that redirects them, s
The mismatch message includes the gateway hostname and the first 16 characters of each fingerprint, the pinned one and the presented one.
| CLI `/login` stops with `The gateway's TLS certificate changed during sign-in: it no longer matches the one you trusted` | A si
+5 −2 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
claude-apps-gateway
±changed
+8−3
claude-codechanged
claude-apps-gateway
The CLI fingerprints the gateway's TLS leaf certificate on first connect and pins it per hostname. Publish the expected SHA-256 fingerprint alongside the gateway URL so developers
The CLI fingerprints the gateway's TLS leaf certificate on first connect and pins it per hostname. It checks that pin again during sign-in, on silent session refreshes, and on mana
Publish the expected SHA-256 fingerprint alongside the gateway URL so developers have something to compare against. The `/login` prompt shows the first 16 characters of the fingerp
Once signed in, the [model picker](/docs/en/model-config) shows the models in the developer's `availableModels` allowlist, managed settings apply at startup and refresh hourly, and
+8 −3 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
authentication
±changed
+7−1
claude-codechanged
authentication
If your organization uses [server-managed settings](/docs/en/server-managed-settings), they apply to this sign-in on Claude Code v2.1.257 or later.
To confirm which source Claude Code chose, run `/status`: a `Profile` row names the source in place of the `Login method` row. If you start Claude Code with `--debug`, it also writ
To confirm which source Claude Code chose, run `/status`. A `Profile` row names the source in place of the `Login method` row, and when the profile is the credential in use, `Organ
+7 −1 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
agent-sdk/typescript
±changed
+23−17
claude-codechanged
agent-sdk/typescript
| `forwardSubagentText` | `boolean` | `false`
| `forwardSubagentText` | `boolean` | `false`
* `user_message_uuid`: the `uuid` of the message you sent that started this turn. See [`user_message_uuid`](#user_message_uuid) for which results carry it.
* `user_message_uuid`: the `uuid` of the message you sent that this turn answered. See [`user_message_uuid`](#user_message_uuid) for which results carry it.
* `request_sent_wall_ms`: epoch milliseconds at which Claude Code dispatched the API request, for joins against server-side timestamps. Present on the success arm only, together wi
+23 −17 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
agent-sdk/python
±changed
+1−1
claude-codechanged
agent-sdk/python
| `forward_subagent_text` | `bool` | `False` | Forward subagent text
| `forward_subagent_text` | `bool` | `False` | Forward subagent text
+1 −1 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
agent-sdk/claude-code-features
±changed
+7−7
claude-codechanged
agent-sdk/claude-code-features
| Input | Behavior
| :------------------------------------------------------------------------------------------------------------------------------ | :-----------------------------------------------
| Managed policy settings | Endpoint-managed policy, such as an MDM plist, r
| `~/.claude.json` global config | Always read
| Auto memory at `~/.claude/projects/<project>/memory/` | Loaded into the system prompt at session start.
+7 −7 lines
Recorded here 9 Sep 2026 19:10 UTC
claude-code
workflows
±changed
+9−1
claude-codechanged
workflows
If you pass a `schema` on an `agent()` call, that subagent returns JSON matching the shape instead of prose. Claude Code checks the schema before starting the subagent: when it can
If the subagent's output still fails validation after five attempts, the call fails with an error that includes the last validation failure. To change the attempt count, set [`MAX_
When you stop a run, it stays in the task panel while any of its agents' processes are still running. If you stop it again, Claude Code re-signals those processes.
+9 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
ultrareview
±changed
+1−1
claude-codechanged
ultrareview
| `--timeout <minutes>` | Maximum minutes to wait for the review to finish. Defaults to 30
| `--timeout <minutes>` | Maximum minutes to wait for the review to finish. Defaults to 45
+1 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
tools-reference
±changed
+4−0
claude-codechanged
tools-reference
Claude Code decides permission for a Glob call before it checks whether the search directory exists. It still runs the read-permission check for a missing `path` outside the [worki
Claude Code decides permission for a Grep call before it checks whether the search `path` exists. It still runs the read-permission check for a missing `path` outside the [working
+4 −0 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
statusline
±changed
+1−1
claude-codechanged
statusline
| `workspace.repo.host`, `workspace.repo.owner`, `workspace.repo.name` | Repository identity parsed from the `origin` remote, for example `"github.com"`, `"anthropics"`
| `workspace.repo.host`, `workspace.repo.owner`, `workspace.repo.name` | Repository identity parsed from the `origin` remote, for example, `"github.com"`, `"anthropics"
+1 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
skills
±changed
+30−22
claude-codechanged
skills
| Field | Required | Description
| :------------------------- | :---------- | :--------------------------------------------------------------------------------------------------------------------------------------
| `name` | No | Display name shown in skill listings. Defaults to the directory name. See [How a skill gets its command name](#how-a-skill-gets-its-com
| `description` | Recommended | What the skill does and when to use it. Claude uses this to decide when to apply the skill. If omitted, uses the first paragraph of mar
| `when_to_use` | No | Additional context for when Claude should invoke the skill, such as trigger phrases or example requests. Appended to `description` in t
+30 −22 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
permissions
±changed
+1−1
claude-codechanged
permissions
`bypassPermissions` mode skips permission prompts, including for writes to [protected paths](/docs/en/permission-modes#protected-paths) such as `.git` and `.claude`. The [cross-s
In `bypassPermissions` mode, Claude Code skips permission prompts, including for writes to [protected paths](/docs/en/permission-modes#protected-paths) such as `.git` and `.claud
+1 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
hooks
±changed
+12−8
claude-codechanged
hooks
| `scratchpad_dir` | Path to the session's scratchpad directory, where Claude keeps temporary working files. Absent when the session has no scratchpad or the temp directory is una
"scratchpad_dir": "/tmp/claude-1000/-home-user-my-project/abc123/scratchpad",
A permission request or elicitation that arrives while another dialog is on screen keeps the same six-second gate, timed from when the request arrives. Its notification can reach
| Matcher | When it fires |
+12 −8 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
google-vertex-ai
±changed
+5−1
claude-codechanged
google-vertex-ai
Claude Code uses `ANTHROPIC_VERTEX_PROJECT_ID` as the project ID for Google Cloud's Agent Platform requests. The `GCLOUD_PROJECT` and `GOOGLE_CLOUD_PROJECT` environment variables
Claude Code addresses Google Cloud's Agent Platform requests to the project in `ANTHROPIC_VERTEX_PROJECT_ID`, even when `GCLOUD_PROJECT`, `GOOGLE_CLOUD_PROJECT`, or the credentia
Before running the command, Claude Code requests an access token with your current credentials to confirm they're actually expired, and skips the command when they still work.
If the check doesn't finish within five seconds, Claude Code also skips the command and runs it only after a request fails with a credential error. Before v2.1.261, a check that ti
+5 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
github-actions
±changed
+2−0
claude-codechanged
github-actions
`/install-github-app` works only with github.com repositories. If your repository's git remote is on gitlab.com or bitbucket.org, the command prints a notice and exits instead of s
+2 −0 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
fullscreen
±changed
+1−0
claude-codechanged
fullscreen
* Clicking also expands the output of a `!` shell command, whether an older truncated result or the live progress row while the command runs. Requires Claude Code v2.1.257 or lat
+1 −0 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
env-vars
±changed
+2−2
claude-codechanged
env-vars
| `ANTHROPIC_VERTEX_PROJECT_ID` | GCP project ID for Google Cloud's Agent Platform requests. Overridden by `GCLOUD_PROJECT`, `GOOGLE_CLOUD_PROJECT`, or th
| `ANTHROPIC_VERTEX_PROJECT_ID` | GCP project ID that Google Cloud's Agent Platform requests are addressed to. See [Configure GCP credentials](/docs/en/go
| `MAX_STRUCTURED_OUTPUT_RETRIES` | Number of times Claude Code retries when the model's response fails validation against the [`--json-schema`](/docs/en/cl
| `MAX_STRUCTURED_OUTPUT_RETRIES` | Number of attempts Claude Code allows when the model's response fails validation against the [`--json-schema`](/docs/en/
+2 −2 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
debug-your-config
±changed
+3−1
claude-codechanged
debug-your-config
Edits to `settings.json` take effect in the running session after a brief file-stability delay. You don't need to restart. If `/hooks` still shows the old definition a few seconds
When you edit `settings.json`, the change takes effect in the running session after a brief file-stability delay, even if you create the file or the project's `.claude/` folder its
If `/hooks` still shows the old definition a few seconds after you save, run `/hooks` again to refresh the view.
+3 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
cross-session-messaging
±changed
+5−1
claude-codechanged
cross-session-messaging
You can message a session shown as `offline` in [the listing](#see-which-sessions-claude-can-reach), one whose Remote Control connection has dropped. The send goes through, but the
* When the dialog stays unanswered past the [`dialogExpiry`](/docs/en/settings-reference#dialogexpiry) deadline, Claude Code closes it and drops the message. The deadline defaults
* When the dialog stays unanswered past the [`dialogExpiry`](/docs/en/settings-reference#dialogexpiry) deadline, Claude Code closes it and drops the message. The deadline defaults
* While no terminal is attached to a [background session](/docs/en/agent-view), Claude Code leaves the dialog open past the deadline. After you attach, if the dialog stays unanswer
+5 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
cli-reference
±changed
+1−1
claude-codechanged
cli-reference
| `claude ultrareview [target]` | Run [ultrareview](/docs/en/ultrareview#run-ultrareview-non-interactively) non-interactively. Prints findings to stdout and exits 0 on success or
| `claude ultrareview [target]` | Run [ultrareview](/docs/en/ultrareview#run-ultrareview-non-interactively) non-interactively. Prints findings to stdout and exits 0 on success or
+1 −1 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
agent-view
±changed
+3−3
claude-codechanged
agent-view
* **Working, waiting on you, or attached**: the process keeps running. A running subagent, workflow, or monitor counts as working.
* **Finished and unattached for about an hour**: the supervisor stops the process to free resources. The conversation stays on disk, and the next time you attach or reply, the sess
* **Working, paused on a permission prompt or other dialog, or attached**: the process keeps running. A running subagent, workflow, or monitor counts as working.
* **Finished or waiting for your next message, and unattached for about an hour**: the supervisor stops the process to free resources. A session that ended its turn by asking you a
Once a session has finished and sat unattached for about an hour, the supervisor stops its process to free resources. Attaching starts a fresh process from where it left off and sw
+3 −3 lines
Recorded here 9 Sep 2026 18:07 UTC
claude-code
agent-sdk/permissions
±changed
+11−2
claude-codechanged
agent-sdk/permissions
Apply the active [permission mode](#permission-modes). `bypassPermissions` approves everything that reaches this step except `rm` and `rmdir` removals targeting a [critical pat
Apply the active [permission mode](#permission-modes):
* In `bypassPermissions` mode, Claude Code approves everything that reaches this step except `rm` and `rmdir` removals targeting a [critical path](/docs/en/permission-modes#cri
* In `acceptEdits` mode, Claude Code approves the file operations listed under [Accept edits mode](#accept-edits-mode-acceptedits).
+11 −2 lines
Recorded here 9 Sep 2026 18:07 UTC
api
release-notes/overview
±changed
+1−0
apichanged
release-notes/overview
* [Per-message effort](https://platform.claude.com/docs/en/build-with-claude/effort#change-effort-mid-conversation-beta) changes, in beta, are also available on [Google Cloud](http
+1 −0 lines
Recorded here 9 Sep 2026 17:37 UTC
api
models/fable-5-1/whats-new-fable-5-1
±changed
+2−2
apichanged
models/fable-5-1/whats-new-fable-5-1
On Claude Fable 5.1 you can change the [effort](https://platform.claude.com/docs/en/build-with-claude/effort) level mid-conversation without invalidating the prompt cache. Raise it
On Claude Fable 5.1 you can change the [effort](https://platform.claude.com/docs/en/build-with-claude/effort) level mid-conversation without invalidating the prompt cache. Raise it
5. Re-run your evals. Refusal handling, fallback, fallback credit, and token counts carry over unchanged. Cache reads cost less (see [Pricing](https://platform.claude.com/docs/en/m
5. Re-run your evals. Refusal handling, fallback, fallback credit, and the tokenizer carry over unchanged. Cache reads cost less (see [Pricing](https://platform.claude.com/docs/en/
+2 −2 lines
Recorded here 9 Sep 2026 17:37 UTC
api
models/fable-5-1/migration-guide
±changed
+2−2
apichanged
models/fable-5-1/migration-guide
* Re-evaluate `effort` with a fresh sweep, starting at `high`, and re-baseline cost and latency on your own workloads. Token counts are roughly unchanged. Prompt cache reads cost a
* Re-evaluate `effort` with a fresh sweep, starting at `high`, and re-baseline cost and latency on your own workloads. The tokenizer is unchanged. Prompt cache reads cost a quarter
* Re-baseline cost on your own workloads. Token counts are roughly unchanged. Per-token pricing differs.
* Re-baseline cost on your own workloads. The tokenizer is unchanged. Per-token pricing differs.
+2 −2 lines
Recorded here 9 Sep 2026 17:37 UTC
Nothing on this page matches that. Press Enter to search every page of the ledger.