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
821pages tracked
767 changed this week · last read 24 Oct 2026
Claude Code CLI
228pages tracked
146 changed this week · last read 24 Oct 2026
Claude Documentation
280pages tracked
69 changed this week · last read 24 Oct 2026
Model Context Protocol
357pages tracked
171 changed this week · last read 23 Oct 2026
100 on this page
ReadSourcePageWhatLines
25 Sep 2026
517 pages
+55087 −56673 lines
claude-docs
claude-tag/admins/healthcare
±changed
+7−8
claude-docschanged
claude-tag/admins/healthcare
Keeping PHI out of Claude's reach needs two things beyond the [general prerequisites for Claude Tag](/docs/claude-tag/admins/setup-overview):
Limiting Claude to approved channels needs an [Enterprise plan](https://claude.com/pricing), in addition to the [general prerequisites for Claude Tag](/docs/claude-tag/admins/setup
* **An Enterprise plan.** Limiting Claude to a list of approved channels uses the **Claude Tag version** setting, which you set separately for the whole workspace and for each chan
* **A Claude organization without Zero Data Retention (ZDR) or customer-managed encryption keys.** Claude Tag stores session transcripts and channel memory, so it [isn't available
If your Claude organization has Zero Data Retention (ZDR) or customer-managed encryption keys, Claude Tag [isn't available in that organization](/docs/claude-tag/concepts/security-
+7 −8 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/federated-access/troubleshooting
±changed
+10−10
claude-docschanged
claude-tag/admins/federated-access/troubleshooting
* You tested in a new thread. A thread already running isn't told about a connection added after it started; ask Claude for the service by name, or send [`@Claude !restart`](/docs/
* You tested in a new thread. A thread already running isn't told about a connection added after it started; ask Claude for the service by name, or send [`@Claude !restart`](/docs/
When a request can't be sent with a federated credential, it fails with an HTTP status and a one-line reason, which Claude usually quotes. Reasons with HTTP 403 and 502 end with th
When a request can't be sent with a federated credential, it fails with an HTTP status and a one-line reason, which Claude usually quotes. Reasons with HTTP 403 and 502 name the co
Claude's request got HTTP 502 with the reason `injection failed ("<connection name>")`.
+10 −10 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/federated-access/gcp
±changed
+1−1
claude-docschanged
claude-tag/admins/federated-access/gcp
New threads pick up the connection on their own. In a thread already running, ask Claude to use Google Cloud. If Claude still can't, send [`@Claude !restart`](/docs/claude-tag/user
New threads pick up the connection on their own. In a thread already running, ask Claude to use Google Cloud. If Claude still can't, send [`@Claude !restart`](/docs/claude-tag/user
+1 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/federated-access/connect-a-gateway
±changed
+1−1
claude-docschanged
claude-tag/admins/federated-access/connect-a-gateway
New threads pick up the connection on their own. In a thread already running, ask Claude to use the gateway and include its address. If Claude still can't see the gateway, send [`@
New threads pick up the connection on their own. In a thread already running, ask Claude to use the gateway and include its address. If Claude still can't see the gateway, send [`@
+1 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/federated-access/aws
±changed
+1−1
claude-docschanged
claude-tag/admins/federated-access/aws
New threads pick up the connection on their own. In a thread already running, ask Claude to use AWS. If Claude still can't, send [`@Claude !restart`](/docs/claude-tag/users/command
New threads pick up the connection on their own. In a thread already running, ask Claude to use AWS. If Claude still can't, send [`@Claude !restart`](/docs/claude-tag/users/command
+1 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/federated-access/authorization-server
±changed
+1−1
claude-docschanged
claude-tag/admins/federated-access/authorization-server
New threads pick up the connection on their own. In a thread already running, ask Claude to use the API and name its host. If Claude still can't, send [`@Claude !restart`](/docs/cl
New threads pick up the connection on their own. In a thread already running, ask Claude to use the API and name its host. If Claude still can't, send [`@Claude !restart`](/docs/cl
+1 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/customize
±changed
+12−11
claude-docschanged
claude-tag/admins/customize
| Setting | What it does
| :-------------------- | :---------------------------------------------------------------------------------------------------------------------------------------------------------
| Custom instructions | Standing guidance read in every session on a scope, like team conventions. Outranks channel memory.
| Respond automatically | Whether Claude replies to a channel's messages without an @-mention. **Respond automatically** exists only on channels, not on workspaces or your whole or
| Plugins | Bundles of skills that teach Claude how to use a specific tool
+12 −11 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/connections/google
±changed
+9−7
claude-docschanged
claude-tag/admins/connections/google
| Field | Value
| :----------------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------
| GCP service account key (JSON) | The JSON key file from Google Cloud Console
| Scopes (optional) | The Google API scopes to request (for example `https://www.googleapis.com/auth/drive.readonly`). The field is labeled optional, but Drive and Cal
| Subject (optional) | A user email to impersonate via domain-wide delegation. Set this for Workspace data (Drive, Calendar, Gmail, Docs).
+9 −7 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/connections/gitlab
±changed
+4−4
claude-docschanged
claude-tag/admins/connections/gitlab
| Field | Value |
| :-------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------ |
| Personal access token | The token from GitLab, starting with `glpat-`. Project and group access tokens work here too; the label is the field name, not a token-type constraint. |
| Allowed websites | `gitlab.com` (preset). For self-managed GitLab, open the **Advanced** tab and add your instance's hostname here. |
| Field | Value |
+4 −4 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/connections/custom
±changed
+2−2
claude-docschanged
claude-tag/admins/connections/custom
* The regionless hosts of IAM, STS, S3, Route 53, CloudFront, Organizations, and Global Accelerator, for example `iam.amazonaws.com`, which Agent Proxy signs for `us-east-1`
* Hosts with no region for S3 and for a fixed set of services that includes IAM, STS, Route 53, CloudFront, and Organizations, for example `iam.amazonaws.com` or `sts.amazonaws.com
Claude can call the endpoint with `curl`, an AWS SDK, or the AWS CLI. The sandbox holds no real AWS credentials, so a CLI or SDK signs the request with placeholder values; Agent Pr
Claude can call the endpoint with `curl`, an AWS SDK, or the AWS CLI. The sandbox holds no real AWS credentials, so a CLI or SDK signs the request with placeholder values; Agent Pr
+2 −2 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/configure-gitlab
±changed
+1−1
claude-docschanged
claude-tag/admins/configure-gitlab
Click **Connect** next to **GitLab** and paste the token into **Personal access token**.
Click **Connect** next to **GitLab** and paste the token into **Claude’s personal access token**.
+1 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/attribute-costs
±changed
+6−1
claude-docschanged
claude-tag/admins/attribute-costs
* **The null row collects spend with no attributable person.** Examples are work Claude started on its own in a thread that another app or bot posted, and a routine whose creator c
* **The null row collects spend with no attributable person.** Examples are:
* the [channel session's](/docs/claude-tag/concepts/how-it-works#lifecycle-of-a-request) own work
* work Claude started on its own in a thread started by Claude, another app, or a bot
* a routine whose creator can't be identified
+6 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
claude-tag/admins/attach-to-scope
±changed
+6−6
claude-docschanged
claude-tag/admins/attach-to-scope
| Option | Effect
| :---------- | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------
| **Inherit** | Follow the parent scope's setting
| **Allow** | Members can edit channel instructions from the Configure link
| **Block** | The Configure page is read-only for members, with a note that only admins can change channel instructions. Claude also declines to make a model the channel default
+6 −6 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
setup
±changed
+5−5
claude-codechanged
setup
| Option | Requires | [Sandboxing](/docs/en/sandboxing) | When to use |
| -------------- | ---------------------------------------------------------------------- | ---------------------------- | ----------------------------------------------- |
| Native Windows | None; [Git for Windows](https://git-scm.com/downloads/win) is optional | Not supported | Windows-native projects and tools |
| WSL 2 | WSL 2 enabled | Supported | Linux toolchains or sandboxed command execution |
| WSL 1 | WSL 1 enabled | Not supported | If WSL 2 is unavailable |
+5 −5 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
overview
±changed
+11−11
claude-codechanged
overview
| What I want to do | Best option
| ------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------
| Continue a local session from my phone or another device | [Remote Control](/docs/en/remote-control)
| Push events from Telegram, Discord, iMessage, or my own webhooks into a session | [Channels](/docs/en/channels)
| Start a task locally, continue on mobile | [`claude --cloud`](/docs/en/claude-code-on-the-web#from-terminal-to-cloud), then the [Claude mob
+11 −11 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
setup
±changed
+3−3
claude-codechanged
setup
New to the terminal? See the [terminal guide](/docs/en/terminal-guide) for step-by-step instructions.
To install Claude Code, use one of the following methods:
To install Claude Code, open a terminal and run the command for your system. If you haven't used a terminal before, the [terminal guide](/docs/en/terminal-guide) shows how to open
+3 −3 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
quickstart
±changed
+3−1
claude-codechanged
quickstart
To install Claude Code, use one of the following methods:
To install Claude Code, open a terminal and run the command for your system. If you haven't used a terminal before, the [terminal guide](/docs/en/terminal-guide) shows how to open
When the installer finishes, open a new terminal window and run `claude --version`. A working installation prints a version number. If your shell says `claude` isn't found or i
+3 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
overview
±changed
+3−1
claude-codechanged
overview
To install Claude Code, use one of the following methods:
To install Claude Code, open a terminal and run the command for your system. If you haven't used a terminal before, the [terminal guide](/docs/en/terminal-guide) shows how to o
When the installer finishes, open a new terminal window and run `claude --version`. A working installation prints a version number. If your shell says `claude` isn't found
+3 −1 lines
Upstream edited 25 Sep 2026 23:59 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
interactive-mode
±changed
+2−0
claude-codechanged
interactive-mode
When a command reaches its timeout before it finishes, Claude Code automatically [moves it to the background](/docs/en/tools-reference#background-commands) instead of stopping it,
+2 −0 lines
Upstream edited 25 Sep 2026 23:53 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
workflows
±changed
+7−1
claude-codechanged
workflows
With ultracode on, Claude decides when a task warrants a workflow. A single request can turn into several workflows in a row: one to understand the code, one to make the change, an
With ultracode on, Claude decides when a task warrants a workflow. A single request can turn into several workflows in a row: one to understand the code, one to make the change, an
Turning ultracode on already opts you in to large runs, so these checks don't apply while it's on:
+7 −1 lines
Upstream edited 25 Sep 2026 23:53 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
sub-agents
±changed
+2−2
claude-codechanged
sub-agents
Each subagent runs in its own context window with a custom system prompt, specific tool access, and independent permissions. When Claude encounters a task that matches a subagent's
Each subagent runs in its own context window with a custom system prompt, specific tool access, and independent permissions. It also sends its own requests, which count toward the
When subagents complete, their results return to your main conversation. Running many subagents that each return detailed results can consume significant context.
When subagents complete, their results return to your main conversation. Running many subagents that each return detailed results can consume significant context, and each subage
+2 −2 lines
Upstream edited 25 Sep 2026 23:53 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
costs
±changed
+3−0
claude-codechanged
costs
The subagent's own requests still draw on your usage. To spend less on them, [choose a smaller model for a subagent](/docs/en/sub-agents#choose-a-model) or [run every subagent on o
* **Subagents and workflows**: every subagent, and every agent a [dynamic workflow](/docs/en/workflows#cost) spawns, sends its own requests on top of the main conversation's. The [
+3 −0 lines
Upstream edited 25 Sep 2026 23:53 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
sessions
±changed
+3−3
claude-codechanged
sessions
A `claude -p --resume` or `claude -p --continue` run resumes in plan mode only when all four conditions hold:
* You pass [`--permission-prompt-tool`](/docs/en/cli-reference#cli-flags), so that Claude Code can present the plan for approval
A `claude -p --resume` or `claude -p --continue` run resumes in plan mode only when all of these conditions hold:
+3 −3 lines
Upstream edited 25 Sep 2026 23:51 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
hooks
±changed
+1−1
claude-codechanged
hooks
To resume a deferred session in plan mode, pass [`--permission-prompt-tool`](/docs/en/cli-reference#cli-flags) along with `--resume` so that Claude Code can present the plan for
To resume a deferred session in plan mode, pass [`--permission-prompt-tool`](/docs/en/cli-reference#cli-flags) along with `--resume` so that Claude Code can present the plan for
+1 −1 lines
Upstream edited 25 Sep 2026 23:51 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
auto-mode-config
±changed
+12−30
claude-codechanged
auto-mode-config
Auto mode is available to all users on every provider, including the Anthropic API, [Claude Platform on AWS](/docs/en/claude-platform-on-aws), Amazon Bedrock, Google Cloud's Agen
This page is the configuration reference. Turning auto mode on and off is covered on the Permission modes page:
* **Switch to auto mode mid-session, or back out of it**: see [Switch permission modes](/docs/en/permission-modes#switch-permission-modes)
* **Start a session in auto mode**: see [Start in a different permission mode](/docs/en/permission-modes#start-in-a-different-mode)
+12 −30 lines
Upstream edited 25 Sep 2026 23:49 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
troubleshoot-install
±changed
+43−43
claude-codechanged
troubleshoot-install
| What you see | Solution
| :--------------------------------------------------------------------------------------------------------- | :--------------------------------------------------------------------
| `command not found: claude` or `'claude' is not recognized` | [Fix your PATH](#command-not-found-claude-after-installation)
| `syntax error near unexpected token '<'` | [Install script returns HTML](#install-script-returns-html-instead-of
| `curl: (22) The requested URL returned error: 403` | [Install script returned 403](#install-script-returns-html-instead-of
+43 −43 lines
Upstream edited 25 Sep 2026 23:45 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
agent-sdk/mcp
±changed
+5−5
claude-codechanged
agent-sdk/mcp
| Server type | Delays the first turn? | First-turn wait timeout
| :------------------------------------------------------------------------------------- | :----------------------------------------------------- | :-------------------------------
| stdio server, or HTTP/SSE server without a cached tool list | Yes, until it connects | [`MCP_TIMEOUT`](/docs/en/env-var
| Remote server with a cached tool list, saved by Claude Code from a previous connection | No; the cached tools are available from the first turn | None; connects on its first tool
| In-process [SDK server](#sdk-mcp-servers) | Yes, until it connects and lists its tools | [`MCP_TIMEOUT`](/docs/en/env-var
+5 −5 lines
Upstream edited 25 Sep 2026 23:45 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
troubleshoot-install
±changed
+23−0
claude-codechanged
troubleshoot-install
| `'claude' is not recognized` right after an update on Windows | [Restore `claude.exe` from its backup](#claude-exe-missing-after-an-u
<h3 id="claude-exe-missing-after-an-update-on-windows">
`claude.exe` missing after an update on Windows
</h3>
+23 −0 lines
Upstream edited 25 Sep 2026 23:45 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
tools-reference
±changed
+1−1
claude-codechanged
tools-reference
| `ListMcpResourcesTool` | Lists resources exposed by connected [MCP servers](/docs/en/mcp)
| `ListMcpResourcesTool` | Lists resources exposed by connected [MCP servers](/docs/en/mcp), leaving out [MCP Apps UI resources](/docs/en/mcp#reference-mcp-resources), which are pa
+1 −1 lines
Upstream edited 25 Sep 2026 23:45 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
mcp
±changed
+10−2
claude-codechanged
mcp
Only an SDK host application, such as an [Agent SDK](/docs/en/agent-sdk/mcp) application or the [desktop app](/docs/en/desktop), can register an in-process `"type": "sdk"` server.
The three scopes match duplicates by name. Plugins and connectors match by endpoint, so one that points at the same URL or command as a server above is treated as a duplicate.
Claude Code matches duplicates across the three scopes by name. It matches plugins and connectors by endpoint, so one that points at the same URL or command as a server above count
+10 −2 lines
Upstream edited 25 Sep 2026 23:45 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
agent-sdk/mcp
±changed
+5−5
claude-codechanged
agent-sdk/mcp
| Server type | Delays the first turn? | First-turn wait timeout
| :------------------------------------------------------------------------------------- | :----------------------------------------------------- | :-------------------------------
| stdio server, or HTTP/SSE server without a cached tool list | Yes, until it connects | [`MCP_TIMEOUT`](/docs/en/env-var
| Remote server with a cached tool list, saved by Claude Code from a previous connection | No; the cached tools are available from the first turn | None; connects on its first tool
| In-process [SDK server](#sdk-mcp-servers) | Yes, until it connects and lists its tools | None; the connect and tool listi
+5 −5 lines
Upstream edited 25 Sep 2026 23:45 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
skills
±changed
+15−11
claude-codechanged
skills
| Same name in | Which one runs
| :------------------------------------------------------------------------------------------------------------- | :----------------------------------------------------------------
| Two of enterprise, personal, and project | Enterprise over personal, and personal over project. With `deploy
| Any of those locations and a [bundled skill](#bundled-skills) | Your skill replaces the bundled command, but not its aliases. A p
| A skill and a file in `.claude/commands/` | The skill
+15 −11 lines
Upstream edited 25 Sep 2026 23:44 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
agent-sdk/plugins
±changed
+8−8
claude-codechanged
agent-sdk/plugins
For complete information on plugin structure and how to create plugins, see [Plugins](/docs/en/plugins/overview.md).
For complete information on plugin structure and how to create plugins, see [Plugins](https://code.claude.com/docs/en/plugins/overview.md).
To use a plugin distributed through a [marketplace](/docs/en/plugins/overview.md) or remote repository, download it first and provide the local directory path. For the directory la
To use a plugin distributed through a [marketplace](https://code.claude.com/docs/en/plugins/overview.md) or remote repository, download it first and provide the local directory pat
If your plugin doesn't appear in the init message's `plugins` list, check its [`plugin_errors`](/docs/en/agent-sdk/typescript#sdksystemmessage) field for the reason, then work thro
+8 −8 lines
Upstream edited 25 Sep 2026 23:42 UTC Recorded here 23 Oct 2026 19:07 UTC
claude-code
agent-sdk/plugins
±changed
+7−7
claude-codechanged
agent-sdk/plugins
For complete information on plugin structure and how to create plugins, see [Plugins](/docs/en/plugins/overview).
For complete information on plugin structure and how to create plugins, see [Plugins](/docs/en/plugins/overview.md).
To use a plugin distributed through a [marketplace](/docs/en/plugins/overview) or remote repository, download it first and provide the local directory path. For the directory layou
To use a plugin distributed through a [marketplace](/docs/en/plugins/overview.md) or remote repository, download it first and provide the local directory path. For the directory la
* [Plugins](/docs/en/plugins/overview) - Complete plugin development guide
+7 −7 lines
Upstream edited 25 Sep 2026 23:42 UTC Recorded here 22 Oct 2026 18:07 UTC
claude-code
agent-sdk/plugins
±changed
+219−227
claude-codechanged
agent-sdk/plugins
<CodeGroup>
```typescript TypeScript theme={null}
import { query } from "@anthropic-ai/claude-agent-sdk";
for await (const message of query({
+219 −227 lines
Upstream edited 25 Sep 2026 23:42 UTC Recorded here 20 Oct 2026 18:37 UTC
claude-code
headless
±changed
+33−33
claude-codechanged
headless
| To load | Use |
| ----------------------- | ------------------------------------------------------- |
| To load | Use |
| - | - |
| Settings | `--settings <file-or-json>` |
+33 −33 lines
Upstream edited 25 Sep 2026 23:42 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
headless
±changed
+6−4
claude-codechanged
headless
| Field | Type | Description
| --------------- | ----- | --------------------------------------------------------------------------------------------------------------------------------------------------------
| `plugins` | array | plugins that loaded successfully, each with `name` and `path`
| `plugin_errors` | array | plugin load-time errors, each with `plugin`, `type`, and `message`. Includes unsatisfied dependency versions and `--plugin-dir` load failures such as a
| Field | Type | Description
+6 −4 lines
Upstream edited 25 Sep 2026 23:42 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
agent-sdk/plugins
±changed
+1−1
claude-codechanged
agent-sdk/plugins
If your plugin doesn't appear in the init message:
If your plugin doesn't appear in the init message's `plugins` list, check its [`plugin_errors`](/docs/en/agent-sdk/typescript#sdksystemmessage) field for the reason, then work thro
+1 −1 lines
Upstream edited 25 Sep 2026 23:42 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
permission-modes
±changed
+8−2
claude-codechanged
permission-modes
The [actions no mode auto-approves](#actions-no-mode-auto-approves) still prompt in this mode.
The [actions no mode auto-approves](#actions-no-mode-auto-approves) still prompt in this mode. The [Remove-Item in PowerShell](#remove-item-in-powershell) denies also apply in this
When you enable the [PowerShell tool](/docs/en/tools-reference#powershell-tool), Claude Code gives `Remove-Item` its own check, separate from the `rm` critical-path list. The outco
When you enable the [PowerShell tool](/docs/en/tools-reference#powershell-tool), Claude Code gives `Remove-Item` and the `cmd` built-ins `rd`, `rmdir`, `del`, and `erase` their own
+8 −2 lines
Upstream edited 25 Sep 2026 23:41 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
env-vars
±changed
+1−0
claude-codechanged
env-vars
| `CLAUDE_CODE_DISABLE_POWERSHELL_CMD_RM_DENY` | Set to `1` to turn off the [PowerShell tool](/docs/en/tools-reference#powershell-tool) check that denies the `cmd` built
+1 −0 lines
Upstream edited 25 Sep 2026 23:41 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
keybindings
±changed
+22−10
claude-codechanged
keybindings
Actions available in the `MessageSelector` context:
| Action | Default | Description |
| :----------------------- | :---------------------------------------- | :---------------- |
| `messageSelector:up` | Up, K, Ctrl+P | Move up in list |
+22 −10 lines
Upstream edited 25 Sep 2026 23:41 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
connectors/building/after-publishing
±changed
+8−8
claude-docschanged
connectors/building/after-publishing
> If your submission isn't live yet, see [Track your directory submission](/docs/directory/submission-status.md), which explains what each status in the portal means and who acts n
> If your submission isn't live yet, see [Track your directory submission](https://claude.com/docs/directory/submission-status.md), which explains what each status in the portal me
To add, change, or remove tools, deploy the change to your server. The tool names shown on your directory listing are part of the listing details, so update them with a [listing ed
To add, change, or remove tools, deploy the change to your server. The tool names shown on your directory listing are part of the listing details, so update them with a [listing ed
A version that doesn't pass, or that is held for a reviewer, doesn't take your listing down: the directory keeps serving the last published version until a newer one is published.
+8 −8 lines
Upstream edited 25 Sep 2026 23:38 UTC Recorded here 23 Oct 2026 18:07 UTC
claude-docs
connectors/building/after-publishing
±changed
+8−8
claude-docschanged
connectors/building/after-publishing
> If your submission isn't live yet, see [Track your directory submission](/docs/directory/submission-status), which explains what each status in the portal means and who acts next
> If your submission isn't live yet, see [Track your directory submission](/docs/directory/submission-status.md), which explains what each status in the portal means and who acts n
To add, change, or remove tools, deploy the change to your server. The tool names shown on your directory listing are part of the listing details, so update them with a [listing ed
To add, change, or remove tools, deploy the change to your server. The tool names shown on your directory listing are part of the listing details, so update them with a [listing ed
A version that doesn't pass, or that is held for a reviewer, doesn't take your listing down: the directory keeps serving the last published version until a newer one is published.
+8 −8 lines
Upstream edited 25 Sep 2026 23:38 UTC Recorded here 22 Oct 2026 18:07 UTC
claude-docs
connectors/building/after-publishing
±changed
+3−3
claude-docschanged
connectors/building/after-publishing
<Note>
If your submission isn't live yet, see [Track your directory submission](/docs/directory/submission-status), which explains what each status in the portal means and who acts next
</Note>
> [!NOTE]
>
+3 −3 lines
Upstream edited 25 Sep 2026 23:38 UTC Recorded here 20 Oct 2026 18:07 UTC
claude-docs
plugins/pre-submission-checklist
±changed
+1−1
claude-docschanged
plugins/pre-submission-checklist
A first submission that fails the security scan is rejected, and a later version that fails can't go live. A new version that the scan flags is held for a reviewer. The **Versions*
A first submission that fails the security scan is rejected, and a later version that fails can't go live. A new version that the scan flags can be held for a reviewer. The **Versi
+1 −1 lines
Upstream edited 25 Sep 2026 23:38 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-docs
connectors/building/after-publishing
±changed
+1−1
claude-docschanged
connectors/building/after-publishing
A plugin listing's name and short description come from `plugin.json` and the README of the version that's live. To change them, edit those files and publish a new version, as [Plu
A plugin listing's name and short description come from `plugin.json` and the README of the version that's live. To change them, edit those files and publish a new version, as [Plu
+1 −1 lines
Upstream edited 25 Sep 2026 23:38 UTC Recorded here 26 Sep 2026 00:07 UTC
claude-code
claude-directory
±changed
+89−89
claude-codechanged
claude-directory
| File | Location | Purpose
| ----------------------- | ------------------------------------------ | -----------------------------------------------------------------------------------------------------------
| `managed-settings.json` | System-level, varies by OS | Enterprise-enforced settings that you can't override, apart from [narrow exceptions](/docs/en/settings#secu
| `CLAUDE.local.md` | Project root | Your private preferences for this project, loaded alongside CLAUDE.md. Create it manually and add it to `.g
| `AGENTS.md` | Project root, `.claude/`, or any directory | Project instructions you write for AI coding agents. Claude Code can [load it](/docs/en/memory#agents-md) o
+89 −89 lines
Upstream edited 25 Sep 2026 23:24 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
hooks
±changed
+1−1
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` | Path to the session's [scratchpad directory](/docs/en/claude-directory#session-scratchpad-directory), where Claude keeps temporary working files. Absent when
+1 −1 lines
Upstream edited 25 Sep 2026 23:24 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
claude-directory
±changed
+21−1
claude-codechanged
claude-directory
### Session scratchpad directory
The scratchpad is a per-session directory that Claude Code gives Claude for temporary files: intermediate results, helper scripts, and drafts that don't belong in your project. Whe
The scratchpad lives under Claude Code's temp directory rather than `~/.claude`. Find the current session's path for your platform:
+21 −1 lines
Upstream edited 25 Sep 2026 23:24 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
platforms
±changed
+23−23
claude-codechanged
platforms
| Platform | Best for | What you get
| :-------------------------------- | :------------------------------------------------------------------------------------------------- | :----------------------------------------
| [CLI](/docs/en/quickstart) | Terminal workflows, scripting, remote servers | Full feature set, [Agent SDK](/docs/
| [Desktop](/docs/en/desktop) | Visual review, parallel sessions, managed setup | Diff viewer, app preview, [computer
| [VS Code](/docs/en/vs-code) | Working inside VS Code without switching to a terminal | Inline diffs, integrated terminal, f
+23 −23 lines
Upstream edited 25 Sep 2026 23:19 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
remote-control
±changed
+22−5
claude-codechanged
remote-control
* **Feature-flag evaluation**: [`DISABLE_TELEMETRY`, `DO_NOT_TRACK`, `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC`, and `DISABLE_GROWTHBOOK`](/docs/en/env-vars) each disable the featu
* **Feature-flag evaluation**: if you set an [environment variable that turns off feature-flag evaluation](/docs/en/env-vars#features-that-need-feature-flag-fetching), whether Remo
* If you set `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC` or `DISABLE_GROWTHBOOK`, Remote Control is unavailable. Unset the variable wherever it's set, in your shell environment or
* If you set only `DISABLE_TELEMETRY` or `DO_NOT_TRACK`, Remote Control stays available unless your organization requires [Trusted Devices](#trusted-devices). If it does, unset t
You can start a Remote Control session from the CLI or the VS Code extension. The CLI offers three invocation modes; VS Code uses the `/remote-control` command.
+22 −5 lines
Upstream edited 25 Sep 2026 23:19 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
platforms
±changed
+1−1
claude-codechanged
platforms
| [Remote Control](/docs/en/remote-control) | Drive a running session from [claude.ai/code](https://claude.ai/code) or the Claude mobile app | Your machine (CLI
| [Remote Control](/docs/en/remote-control) | Drive a running session from [claude.ai/code](https://claude.ai/code) or the Claude mobile app | Your machine (CLI
+1 −1 lines
Upstream edited 25 Sep 2026 23:19 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
agent-sdk/typescript
±changed
+16−3
claude-codechanged
agent-sdk/typescript
Claude Code sets the list together with `user_message_uuid` on each reply frame that carries that field and on the result. For the full set of frames that carry `user_message_uuid`
Claude Code sets the list together with `user_message_uuid` on each reply frame that carries that field and on the result. For the full set of turn frames that echo the answered me
Claude Code also emits this message when an MCP server's [prompts](/docs/en/mcp#use-mcp-prompts-as-commands) join or leave the list, for example when a server finishes connecting a
};
+16 −3 lines
Upstream edited 25 Sep 2026 23:19 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
headless
±changed
+5−1
claude-codechanged
headless
When you [resume the session](#continue-conversations), Claude Code continues the turn that SIGTERM left unfinished.
When you [resume the session](#continue-conversations), Claude Code leaves the interrupted turn as it is, and your next prompt drives the conversation. To have Claude Code continue
### If the working directory is deleted
+5 −1 lines
Upstream edited 25 Sep 2026 23:09 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
server-managed-settings
±changed
+14−14
claude-codechanged
server-managed-settings
| Approach | Best for | Security model
| :------------------------------------------------------------------------ | :------------------------------------------------------- | :------------------------------------------
| **Server-managed settings** | Organizations without MDM, or users on unmanaged devices | Settings that Claude Code fetches from Anth
| **[Endpoint-managed settings](/docs/en/managed-settings#delivery-mechanisms)** | Organizations with MDM or endpoint management | Settings deployed to devices via MDM c
| Approach | Best for | Security model |
+14 −14 lines
Upstream edited 25 Sep 2026 23:09 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
server-managed-settings
±changed
+11−2
claude-codechanged
server-managed-settings
* The cache at `~/.claude/remote-settings.json` stores the salvaged payload with invalid entries removed, apart from invalid `cleanupPeriodDays` and `desktopSessionCleanupPeriodDay
* When no field in the payload can be salvaged and the payload isn't only those retention keys, Claude Code rejects the payload, keeps the last-accepted cached settings, and writes
* A startup that runs on the cache at `~/.claude/remote-settings.json` treats invalid entries the way the fetch that wrote the cache did:
* Entries that failed validation stay dropped.
* [Keys that fail closed](/docs/en/managed-settings#keys-that-fail-closed) keep their stricter values.
+11 −2 lines
Upstream edited 25 Sep 2026 23:09 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
errors
±changed
+6−1
claude-codechanged
errors
Your session is eligible for [server-managed settings](/docs/en/server-managed-settings), but Claude Code couldn't fetch them, so it shows this warning in interactive sessions. The
Your session is eligible for [server-managed settings](/docs/en/server-managed-settings), but Claude Code couldn't fetch them or couldn't apply what the server returned, so it show
The parenthesized cause names what failed, such as `network error`, `request timed out`, or `authentication rejected (401)`. The cause `no setting in the server response could be a
+6 −1 lines
Upstream edited 25 Sep 2026 23:09 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
settings-reference
±changed
+1−1
claude-codechanged
settings-reference
* **Default**: unset, so auto-connect follows your organization's admin default when one is set, and otherwise Claude Code's current default
* **Default**: unset, so the [auto-connect default](/docs/en/remote-control#enable-remote-control-for-all-sessions) applies
+1 −1 lines
Upstream edited 25 Sep 2026 23:07 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
managed-settings
±changed
+1−1
claude-codechanged
managed-settings
If you turn telemetry off, Claude Code stops sending the usage data that feeds your organization's [analytics dashboard](/docs/en/analytics) for the developers the policy reaches.
If you turn telemetry off, Claude Code stops sending the usage data that feeds your organization's [analytics dashboard](/docs/en/analytics) for the developers the policy reaches.
+1 −1 lines
Upstream edited 25 Sep 2026 23:07 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
env-vars
±changed
+4−4
claude-codechanged
env-vars
| `DISABLE_TELEMETRY` | Set to any non-empty value, such as `1`, to opt out of telemetry. **Setting it to `0` or `false` still opts out**, unlik
| `DISABLE_TELEMETRY` | Set to any non-empty value, such as `1`, to opt out of telemetry. **Setting it to `0` or `false` still opts out**, unlik
| `DO_NOT_TRACK` | Set to `1` to opt out of telemetry, with the same effect as `DISABLE_TELEMETRY`, including making [Remote Control](/docs
| `DO_NOT_TRACK` | Set to `1` to opt out of telemetry, with the same effect as `DISABLE_TELEMETRY`, including on [feature-flag fetching](#f
* Use [Remote Control](/docs/en/remote-control#requirements)
+4 −4 lines
Upstream edited 25 Sep 2026 23:07 UTC Recorded here 25 Sep 2026 23:37 UTC
claude-code
troubleshoot-install
±changed
+10−0
claude-codechanged
troubleshoot-install
| `Claude Code access has not been granted for this account` | [Get a role that includes Claude Code](#claude-code-access-has-not-be
### Claude Code access has not been granted for this account
If the sign-in page shows `Authorization failed` with the message `Claude Code access has not been granted for this account. Contact your administrator.` after you log in from Clau
+10 −0 lines
Upstream edited 25 Sep 2026 23:04 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
mcp
±changed
+8−0
claude-codechanged
mcp
### Images in tool results
When an MCP tool returns a PNG, JPEG, GIF, or WebP image, Claude sees the image inline in the conversation. The inline copy may be scaled down or compressed to fit the model's imag
If you disable session persistence with [`--no-session-persistence`](/docs/en/cli-reference#cli-flags) or [`CLAUDE_CODE_SKIP_PROMPT_HISTORY`](/docs/en/env-vars), Claude Code writes
+8 −0 lines
Upstream edited 25 Sep 2026 23:02 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
claude-directory
±changed
+1−1
claude-codechanged
claude-directory
| `projects/<project>/<session>/tool-results/` | Large tool outputs spilled to separate files
| `projects/<project>/<session>/tool-results/` | Large tool outputs spilled to separate files, an
+1 −1 lines
Upstream edited 25 Sep 2026 23:02 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
plugins/manifest-reference
±changed
+6−4
claude-codechanged
plugins/manifest-reference
Every component path must resolve inside the plugin root and must exist. `claude plugin validate` doesn't check the `outputStyles`, `lspServers`, `monitors`, or `themes` paths, so
* **Containment**: a path that resolves outside the plugin root doesn't load, and the `/plugin` **Errors** tab shows `<component> path escapes plugin directory: <path>`. A path con
* **Existence**: a path that doesn't exist doesn't load, and the `/plugin` **Errors** tab shows `<component> path not found: <path>`. `claude plugin validate` reports it as `Path n
Every component path must resolve inside the plugin root and must exist. `claude plugin validate` checks the paths under every component key:
+6 −4 lines
Upstream edited 25 Sep 2026 23:01 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
claude-apps-gateway-config
±changed
+2−0
claude-codechanged
claude-apps-gateway-config
No model request is sent to the provider, so a replica's CPU per request is an estimate and reads lower than production, which also encrypts its traffic to the provider. Confirm a
+2 −0 lines
Upstream edited 25 Sep 2026 23:01 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
plugin-evals
±changed
+91−91
claude-codechanged
plugin-evals
| Target | What runs
| :-------------------------------------------------------- | :---------------------------------------------------------------------------------------------------------------------
| A plugin's root directory, such as `.` | Every case under its eval directory, with that plugin loaded
| A single `prompt.md` or `case.yaml` file | That case, with its enclosing plugin loaded
| Target | What runs |
+91 −91 lines
Upstream edited 25 Sep 2026 23:00 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
plugin-evals
±changed
+15−0
claude-codechanged
plugin-evals
* Git 2.31 or later, if git is installed. Run `git --version` to check. With an older git, `claude plugin eval` [stops before running any case](#git-is-too-old-for-claude-plugin-ev
<h3 id="git-is-too-old-for-claude-plugin-eval">
"is too old for claude plugin eval"
</h3>
+15 −0 lines
Upstream edited 25 Sep 2026 23:00 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
llm-gateway-protocol
±changed
+2−1
claude-codechanged
llm-gateway-protocol
The headers carry only what the rows below list: fixed vocabularies, tool names, and durations, never prompt text or file contents. Every value is printable ASCII.
The headers carry only what the rows below list: fixed vocabularies, tool names, durations, and a random prompt identifier, never prompt text or file contents. Every value is print
| `x-claude-code-prompt-id` | Random UUID that identifies the user prompt a request serves. Requests serving one prompt share the value, including the turns of subagents
+2 −1 lines
Upstream edited 25 Sep 2026 23:00 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
agent-sdk/observability
±changed
+10−10
claude-codechanged
agent-sdk/observability
| Signal | What it contains | Enable with |
| ---------- | --------------------------------------------------------------------------- | ------------------------------------------------------------------- |
| Metrics | Counters for tokens, cost, sessions, lines of code, and tool decisions | `OTEL_METRICS_EXPORTER` |
| Log events | Structured records for each prompt, API request, API error, and tool result | `OTEL_LOGS_EXPORTER` |
| Traces | Spans for each interaction, model request, tool call, and hook (beta) | `OTEL_TRACES_EXPORTER` plus `CLAUDE_CODE_ENHANCED_TELEMETRY_BETA=1` |
+10 −10 lines
Upstream edited 25 Sep 2026 22:58 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
monitoring-usage
±changed
+12−9
claude-codechanged
monitoring-usage
MCP tools, WebFetch, and WebSearch record this event too, on Claude Code v2.1.283 or later.
* A call to any tool other than Read, Edit, Write, and Bash, including MCP tools and WebFetch
* A call to any tool other than Read, Edit, Write, Bash, WebFetch, WebSearch, and MCP tools
* A WebFetch or WebSearch call that Claude Code moved to the background because you interrupted the turn to [send your queued messages right away](/docs/en/interactive-mode#when-cl
+12 −9 lines
Upstream edited 25 Sep 2026 22:58 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
agent-sdk/observability
±changed
+1−1
claude-codechanged
agent-sdk/observability
| `OTEL_LOG_TOOL_CONTENT=1` | A [`tool.output` span event](/docs/en/monitoring-usage#tool-output-span-event) on `claude_code.tool` with file contents and Bash output, truncated at
| `OTEL_LOG_TOOL_CONTENT=1` | A [`tool.output` span event](/docs/en/monitoring-usage#tool-output-span-event) on `claude_code.tool` with file contents, Bash output, and what MCP too
+1 −1 lines
Upstream edited 25 Sep 2026 22:58 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
self-hosted-environments-deploy
±changed
+13−13
claude-codechanged
self-hosted-environments-deploy
| Host | Port | Used for
| :----------------------------------------------------------------- | :----------------------------------------- | :---------------------------------------------------------------
| `api.anthropic.com` | 443, HTTPS; WSS for the SCM connector only | Runner control plane and session streaming, model inference, fea
| Your git host, such as `github.com` or your GitHub Enterprise host | 443 or 22 | Cloning and pushing repositories. Not needed if the runner uses
| Host | Port | Used for |
+13 −13 lines
Upstream edited 25 Sep 2026 22:49 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
self-hosted-environments-deploy
±changed
+7−3
claude-codechanged
self-hosted-environments-deploy
Swap `linux-x64` for `linux-arm64` if your nodes are ARM, or for `linux-x64-musl` or `linux-arm64-musl` on a musl-based image such as Alpine; see [Alpine Linux setup](/docs/en/setu
Swap `linux-x64` for `linux-arm64` if your nodes are ARM, or for `linux-x64-musl` or `linux-arm64-musl` on a musl-based image such as Alpine; see [Alpine Linux setup](/docs/en/setu
docker build --build-arg CLAUDE_CODE_VERSION=2.1.267 -t <your-registry>/claude-runner:latest .
```
docker build \
+7 −3 lines
Upstream edited 25 Sep 2026 22:49 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
desktop-quickstart
±changed
+8−7
claude-codechanged
desktop-quickstart
> Install Claude Code on desktop and start your first coding session
> Install the Claude desktop app, open the Code tab, and start your first Claude Code session on a project folder on your computer.
The desktop app gives you Claude Code with a graphical interface built for running multiple sessions side by side: a sidebar for managing parallel work, a drag-and-drop layout with
The desktop app gives you Claude Code with a graphical interface, so you can ask Claude to work on the code in a folder on your computer and review its changes without using a term
Claude Code requires a [Pro, Max, Team, or Enterprise subscription](https://claude.com/pricing?utm_source=claude_code\&utm_medium=docs\&utm_content=desktop_quickstart_pricing).
+8 −7 lines
Upstream edited 25 Sep 2026 22:48 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
sandboxing
±changed
+40−40
claude-codechanged
sandboxing
| Prefix | Meaning | Example
| :---------------- | :------------------------------------------------------------------------------------- | :--------------------------------------------------------------------
| `/` | Absolute path from filesystem root | `/tmp/build` stays `/tmp/build`
| `~/` | Relative to home directory | `~/.kube` becomes `$HOME/.kube`
| Prefix | Meaning | Example |
+40 −40 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
desktop-scheduled-tasks
±changed
+16−16
claude-codechanged
desktop-scheduled-tasks
| | [Cloud](/docs/en/routines) | [Desktop](/docs/en/desktop-scheduled-tasks) | [`/loop`](/docs/en/scheduled-tasks)
| :------------------------- | :---------------------------------- | :------------------------------------- | :---------------------------------------------------------------------
| Runs on | Cloud, Anthropic-managed by default | Your machine | Your machine
| Requires machine on | No | Yes | Yes
| Requires open session | No | No | Yes
+16 −16 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
agent-sdk/tool-search
±changed
+7−7
claude-codechanged
agent-sdk/tool-search
| Value | Behavior
| :------- | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------
| (unset) | Tool search is on. Tool definitions are deferred and discovered on demand. Falls back to loading upfront on Google Cloud's Agent Platform models earlier than the Claud
| `true` | Tool search is always on, except on a Microsoft Foundry deployment hosted on Azure, where the server-side rejection still forces upfront loading, and on Google Cloud's
| `auto` | Counts the tokens in the tool definitions that tool search can defer and compares the total against the model's context window. When the total reaches 10% of the windo
+7 −7 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
skills
±changed
+14−14
claude-codechanged
skills
Every skill needs a `SKILL.md` file with two parts: YAML frontmatter between `---` markers that tells Claude when to use the skill, and markdown content with the instructions C
Every skill needs a `SKILL.md` file with two parts: YAML frontmatter between `---` markers that tells Claude when to use the skill, and markdown content with the instructions C
When a nested skill shares a name with another skill, both stay available. With a `deploy` skill at the repository root and another in `apps/web/.claude/skills/`:
When a nested skill's directory name matches another skill's name, both stay available. With a `deploy` skill at the repository root and another in `apps/web/.claude/skills/`:
When two skills share a name, where each one came from decides which one `/name` runs. The table covers the enterprise, personal, project, nested, plugin, and claude.ai locations,
+14 −14 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
settings-reference
±changed
+3−3
claude-codechanged
settings-reference
Control which paths sandboxed commands can read and write. By default they can write to the working directory, the session temp directory, and directories you add with `--add-dir`,
Control which paths sandboxed commands can read and write. By default they can write to the working directory, the per-user temp directory, and directories you add with `--add-dir`
Add paths where sandboxed commands can write, beyond the working directory, the session temp directory, and the directories you've added with `--add-dir`, `/add-dir`, or `permissio
Add paths where sandboxed commands can write, beyond the working directory, the per-user temp directory, and the directories you've added with `--add-dir`, `/add-dir`, or `permissi
* **Default**: unset, so sandboxed commands can write to the working directory, the session temp directory, directories you've added with `--add-dir` or `/add-dir`, and directories
+3 −3 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
sandboxing
±changed
+6−6
claude-codechanged
sandboxing
Ask Claude to run a command, such as a build or a test suite. By default, commands inside the sandbox can write to the working directory, the session temp directory, and any [d
Ask Claude to run a command, such as a build or a test suite. By default, commands inside the sandbox can write to the working directory, a [per-user temp directory](/docs/en/e
The session temp directory is writable inside the sandbox by default, alongside the working directory. Unless you [disable filesystem isolation](#disable-filesystem-isolation), Cla
A per-user temp directory is writable inside the sandbox by default, alongside the working directory. Unless you [disable filesystem isolation](#disable-filesystem-isolation), Clau
By default, sandboxed commands can write to the current working directory, the session temp directory, and any [directories you've added](/docs/en/permissions#additional-directorie
+6 −6 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
routines
±changed
+2−2
claude-codechanged
routines
Pick a preset frequency for a recurring run, or schedule a single one-off run at a specific timestamp. See [Add a schedule trigger](#add-a-schedule-trigger) for timezone ha
Pick a preset frequency for a recurring run, or schedule a single one-off run at a specific timestamp. See [Add a schedule trigger](#add-a-schedule-trigger) for timezone ha
Runs may start a few minutes after the scheduled time due to stagger. The offset is consistent for each routine.
If you schedule a run exactly on the hour, such as 9:00, it can start several minutes late. To start close to the scheduled time, pick a few minutes past the hour, for example 9:07
+2 −2 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
permission-modes
±changed
+4−1
claude-codechanged
permission-modes
[`permissions.allow`](/docs/en/permissions#manage-permissions) rules in settings files do not pre-approve protected-path writes. The safety check runs before Claude Code evaluates
[`permissions.allow`](/docs/en/permissions#manage-permissions) rules in settings files do not pre-approve protected-path writes. The safety check runs before Claude Code evaluates
* For the project's `.claude/` folder: **Yes, and allow Claude to edit files in this project's .claude folder for this session**
* For `~/.claude/`: **Yes, and allow Claude to edit files in its \~/.claude folder for this session**
+4 −1 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
desktop-scheduled-tasks
±changed
+1−1
claude-codechanged
desktop-scheduled-tasks
* **Daily**: shows a time picker, defaults to 9:00 AM local time
* **Daily**: runs every day at the local time you pick
+1 −1 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
claude-apps-gateway
±changed
+1−1
claude-codechanged
claude-apps-gateway
| HTTPS | The gateway must be reachable over `https://` from developer laptops and from any browser used for sign-in; the gateway serves the devic
| HTTPS | The gateway must be reachable over `https://` from developer laptops and from any browser used for sign-in; the gateway serves the devic
+1 −1 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
agent-sdk/tool-search
±changed
+1−1
claude-codechanged
agent-sdk/tool-search
You can also add a system prompt section listing available tool categories. This gives the agent context about what kinds of tools are available to search for. Pass the text throug
You can also add a system prompt section listing available tool categories. This gives the agent context about what kinds of tools are available to search for. Pass the text in the
+1 −1 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
advisor
±changed
+1−0
claude-codechanged
advisor
* **Unavailable**: the advisor call failed, and the line reads `Advisor unavailable (<error_code>)`, where `<error_code>` is the code the call returned.
+1 −0 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
agent-sdk/typescript
±changed
+7−1
claude-codechanged
agent-sdk/typescript
| `verbatimPrompts` | `boolean`
client_composed?: true;
Set `shouldQuery` to `false` to append the message to the transcript without triggering an assistant turn. The message is held and merged into the next user message that does trigg
Set `shouldQuery` or `client_composed` to change how Claude Code handles a message you send:
+7 −1 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-code
agent-sdk/python
±changed
+52−50
claude-codechanged
agent-sdk/python
verbatim_prompts: bool = False
| Property | Type | Default | Description
| :---------------------------- | :------------------------------------------------------------------------------------ | :--------------------------------- | :--------------------
| `tools` | `list[str] \| ToolsPreset \| None` | `None` | Tools configuration.
| `allowed_tools` | `list[str]` | `[]` | Tools to auto-approve
+52 −50 lines
Upstream edited 25 Sep 2026 22:47 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-docs
claude-tag/admins/managed-by
±changed
+29−29
claude-docschanged
claude-tag/admins/managed-by
| You want | Use
| :---------------------------------------------------------------------------------------------------------------------------- | :-------------------------------------------------
| The same rules in every channel of a workspace or your whole organization | [Custom instructions](/docs/claude-tag/admins/atta
| A central team to write the rules for a few channels it runs, such as a help desk or an on-call channel, without being admins | **Managed by**
| The people who work in a channel to set its conventions themselves | The **Channel instructions** field on the channel'
+29 −29 lines
Upstream edited 25 Sep 2026 22:37 UTC Recorded here 28 Sep 2026 22:07 UTC
claude-docs
claude-tag/admins/managed-by
±changed
+1−0
claude-docschanged
claude-tag/admins/managed-by
| One of those channels is archived | One of the managing channels is archived in Slack
+1 −0 lines
Upstream edited 25 Sep 2026 22:37 UTC Recorded here 25 Sep 2026 23:07 UTC
claude-docs
claude-tag/admins/restrict-access
±changed
+3−3
claude-docschanged
claude-tag/admins/restrict-access
* **Their thread replies aren't read.** In a thread an allowed member started, a restricted member's replies don't reach Claude as content. Claude sees that a message arrived, but
<Warning>On a Slack Enterprise Grid whose workspaces are paired to different Claude organizations, one organization's access settings govern the entire grid, so your restrictions m
* **Their thread replies aren't treated as requests.** In a thread an allowed member started, a restricted member's replies reach Claude marked as coming from someone who can't use
+3 −3 lines
Upstream edited 25 Sep 2026 22:37 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-docs
claude-tag/users/models
±changed
+1−1
claude-docschanged
claude-tag/users/models
Claude sets the channel's default model, which applies to new threads in that channel. Threads already underway keep the model they started with until someone in them asks Claude t
Claude sets the channel's default model. New threads in the channel start on it. A thread already underway switches to it at the next message anyone posts there, unless someone in
+1 −1 lines
Upstream edited 25 Sep 2026 22:36 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-docs
claude-tag/admins/customize
±changed
+2−2
claude-docschanged
claude-tag/admins/customize
The setting applies to new sessions; threads already underway keep the model they started with. The footer of each Claude reply in Slack names the model that handled it, so you can
When you change the setting, new sessions start on the new model. A thread already underway switches to it at the next message anyone posts there, unless someone in that thread has
Channel members can also change the model from Slack. Asking Claude in a thread switches that thread, and asking it to make a model the channel default changes what new threads in
Channel members can also change the model from Slack. Asking Claude in a thread switches that thread, and asking it to make a model the channel default changes this setting for the
+2 −2 lines
Upstream edited 25 Sep 2026 22:36 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-code
claude-apps-gateway-config
±changed
+31−0
claude-codechanged
claude-apps-gateway-config
When you [add your own labels](#add-your-own-labels), the gateway also pushes `OTEL_RESOURCE_ATTRIBUTES`.
#### Add your own labels
+31 −0 lines
Upstream edited 25 Sep 2026 22:34 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-code
permissions
±changed
+2−0
claude-codechanged
permissions
Before it starts or restarts a [background session](/docs/en/agent-view), Claude Code also checks workspace trust for the directory the session runs in. If you run `claude --bg` fr
+2 −0 lines
Upstream edited 25 Sep 2026 22:31 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-code
errors
±changed
+23−1
claude-codechanged
errors
| `Workspace not trusted.` when starting or restarting a background session
You tried to start a [background session](/docs/en/agent-view) in a directory that doesn't exist anymore. This happens when you dispatch from agent view or run `/background` after
You tried to start a [background session](/docs/en/agent-view) in a directory that doesn't exist anymore. Claude Code doesn't start the session, and the message names the missing d
### Workspace not trusted when dispatching a background session
+23 −1 lines
Upstream edited 25 Sep 2026 22:31 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-code
cli-reference
±changed
+2−2
claude-codechanged
cli-reference
| `--bg`, `--background` | Start the session as a [background agent](/docs/en/agent-view) and return immediately. Prints the session ID and management comm
| `--bg`, `--background` | Start the session as a [background agent](/docs/en/agent-view) and return immediately. Prints the session ID and management comm
| `--setting-sources` | Comma-separated list of setting sources to load (`user`, `project`, `local`)
| `--setting-sources` | Comma-separated list of setting sources to load (`user`, `project`, `local`). See [agent view](/docs/en/agent-view#what-carries-
+2 −2 lines
Upstream edited 25 Sep 2026 22:31 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-code
agent-view
±changed
+21−5
claude-codechanged
agent-view
* `--setting-sources`
If you run `claude --bg` from a terminal in a directory you haven't [trusted](/docs/en/permissions#project-allow-rules-and-workspace-trust), the workspace trust dialog appears firs
* Delete the session again to remove the directory anyway, by pressing `Ctrl+X` twice on its row in agent view or running the `claude rm <id> --force-remove-worktree <worktree-id
* Fix what stands in the way, such as committing or stashing the uncommitted changes, closing whatever is using the directory, or fixing the hook, then delete the session again.
+21 −5 lines
Upstream edited 25 Sep 2026 22:31 UTC Recorded here 25 Sep 2026 22:37 UTC
claude-code
agent-teams
±changed
+12−12
claude-codechanged
agent-teams
| | Subagents | Agent te
| :---------------- | :-------------------------------------------------------------------------------------------------------------------------------------------------- | :-------
| **Context** | Own context window; results return to the caller | Own cont
| **Communication** | Return a result to the caller. Subagents that Claude named when it spawned them can also [message each other](/docs/en/sub-agents#what-loads-at-startup) | Tea
| **Coordination** | Main agent manages all work | Self-coo
+12 −12 lines
Upstream edited 25 Sep 2026 22:12 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
agent-teams
±changed
+3−1
claude-codechanged
agent-teams
Each teammate has its own context window. When spawned, a teammate loads the same project context as a regular session: CLAUDE.md, MCP servers, and skills. It also receives the spa
Each teammate has its own context window. When spawned, a teammate loads the same project context as a regular session: CLAUDE.md, MCP servers, and skills. If you start the lead wi
A teammate also receives the spawn prompt from the lead. The lead's conversation history does not carry over.
+3 −1 lines
Upstream edited 25 Sep 2026 22:12 UTC Recorded here 25 Sep 2026 22:37 UTC
Nothing on this page matches that. Press Enter to search every page of the ledger.