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
81 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
15 Sep 2026
288 pages
+14462 −11107 lines
claude-code
plugin-dependencies
±changed
+1−1
claude-codechanged
plugin-dependencies
An entry can be a bare string with only the plugin name, like `"audit-logger"` in the example above, which depends on whatever version that plugin's marketplace provides. For more
An entry can be a bare string with only the plugin name, like `"audit-logger"` in the `deploy-kit` manifest, which depends on whatever version that plugin's marketplace provides. F
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
permission-modes
±changed
+1−1
claude-codechanged
permission-modes
* Committing or pushing a change that would send secrets or sensitive data outside the repository when it runs, or widen what a deploy exposes. This covers a CI workflow or deploy
* Committing or pushing a change that would send secrets or sensitive data outside the repository when it runs, or widen what a deploy exposes. This covers a CI workflow or deploy
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
overview
±changed
+7−7
claude-codechanged
overview
```bash theme={null}
```bash theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={n
```powershell theme={null}
```powershell theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} theme={null} th
```batch theme={null}
+7 −7 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
model-config
±changed
+3−3
claude-codechanged
model-config
* Cloud sessions, on [Claude Code on the web](/docs/en/claude-code-on-the-web) or in the Desktop app, run on Anthropic-managed VMs by default: settings deployed to your device do n
* Cloud sessions, on [Claude Code on the web](/docs/en/claude-code-on-the-web) or in the Desktop app, run on Anthropic-managed VMs by default: settings deployed to your device do n
* The model pickers on claude.ai and in the Desktop app hide or grey out models excluded by your organization's allowlist. The picker state is a convenience for users; enforcement
* The model pickers on claude.ai and in the Desktop app hide or grey out models excluded by your organization's allowlist. The picker state is a convenience for users; it doesn't e
The Claude Console has no model restriction control. Organizations without a Claude Enterprise plan, including those whose members authenticate through the Anthropic API, restrict
+3 −3 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
microsoft-foundry
±changed
+1−1
claude-codechanged
microsoft-foundry
1. Navigate to the [Microsoft Foundry portal](https://ai.azure.com/)
1. Go to the [Microsoft Foundry portal](https://ai.azure.com/)
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
mcp-quickstart
±changed
+1−1
claude-codechanged
mcp-quickstart
There are other ways to add a server, each covered later on this page:
There are other ways to add a server, each with its own section:
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
mcp
±changed
+23−5
claude-codechanged
mcp
`--env` accepts multiple `KEY=value` pairs. If the server name comes directly after `--env`, the CLI reads the name as another pair and rejects it, so place at least one other op
`--env` accepts multiple `KEY=value` pairs. If the server name comes directly after `--env`, the CLI reads the name as another pair and rejects it, so place at least one other op
MCP servers aren't specific to Claude Code, so a server's setup instructions may be written for Claude Desktop, Cursor, or another MCP client and give no `claude mcp add` command.
MCP servers aren't specific to Claude Code, so a server's setup instructions may be written for Claude Desktop, Cursor, or another MCP client and give no `claude mcp add` command.
**Supported syntax:**
+23 −5 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
managed-settings
±changed
+5−4
claude-codechanged
managed-settings
* **Gateway login keys**: [`forceLoginGatewayUrl`](/docs/en/settings-reference#forcelogingatewayurl) and the `"gateway"` value of [`forceLoginMethod`](/docs/en/settings-reference#f
* **Gateway login keys**: [`forceLoginGatewayUrl`](/docs/en/settings-reference#forcelogingatewayurl), [`gatewayInternalNetworks`](/docs/en/settings-reference#gatewayinternalnetwork
Both settings rank the sources the same way. Two terms recur in this section:
Both settings rank the sources the same way. These terms recur in this section:
The [gateway login keys](#choose-a-delivery-mechanism), [`forceLoginGatewayUrl`](/docs/en/settings-reference#forcelogingatewayurl) and the `"gateway"` value of [`forceLoginMethod`]
+5 −4 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
managed-mcp
±changed
+1−1
claude-codechanged
managed-mcp
* [Server-managed settings](/docs/en/server-managed-settings): deliver `allowedMcpServers` and `deniedMcpServers` from the Claude.ai admin console
* [Server-managed settings](/docs/en/server-managed-settings): deliver `allowedMcpServers` and `deniedMcpServers` from the claude.ai admin console
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
llm-gateway-rollout
±changed
+1−1
claude-codechanged
llm-gateway-rollout
Every developer machine needs the gateway address and a credential. You can distribute them centrally through [managed settings](/docs/en/managed-settings#delivery-mechanisms), so
Every developer machine needs the gateway address and a credential. You can distribute them centrally through [managed settings](/docs/en/managed-settings#delivery-mechanisms), so
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
llm-gateway-protocol
±changed
+1−1
claude-codechanged
llm-gateway-protocol
Subagent IDs are generated fresh for each spawn. Teammate agents, the named members of an [agent team](/docs/en/agent-teams), reuse a stable name-based ID across reconnections. In
Subagent IDs are generated fresh each time Claude Code spawns a subagent. Teammate agents, the named members of an [agent team](/docs/en/agent-teams), reuse a stable name-based ID
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
llm-gateway-connect
±changed
+1−0
claude-codechanged
llm-gateway-connect
| `400` errors on every request, in the gateway's own words rejecting a tool's input schema or its `pattern`, on Claude Code v2.1.265 through v2.1.267
+1 −0 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
jetbrains
±changed
+1−1
claude-codechanged
jetbrains
When Claude Code runs in a JetBrains IDE in [`acceptEdits` permission mode](/docs/en/permission-modes#auto-approve-file-edits-with-acceptedits-mode), it may be able to modify IDE c
When Claude Code runs in a JetBrains IDE in [`acceptEdits` permission mode](/docs/en/permission-modes#auto-approve-file-edits-with-acceptedits-mode), it may be able to modify IDE c
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
how-claude-code-works
±changed
+1−1
claude-codechanged
how-claude-code-works
The agentic loop, tools, and capabilities described above are the same everywhere you use Claude Code. What changes is where the code executes and how you interact with it.
The [agentic loop](#the-agentic-loop), [tools](#tools), and capabilities are the same everywhere you use Claude Code. What changes is where the code executes and how you interact w
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
hooks
±changed
+32−2
claude-codechanged
hooks
When you run `/clear` in an interactive session, the matching SessionStart hooks run in the background and the prompt accepts input again right away. Claude's first response still
When you start an interactive session, resume a conversation at launch with `--continue` or `--resume`, or run `/clear`, SessionStart hooks run in the background. You can type righ
When you switch conversations with `/resume` inside a session, the switch waits for the hooks to finish instead. If you run `/clear` or switch to another conversation while backgro
<a id="bash" />
+32 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
google-vertex-ai
±changed
+1−1
claude-codechanged
google-vertex-ai
1. Navigate to the [Google Cloud's Agent Platform Model Garden](https://console.cloud.google.com/vertex-ai/model-garden)
1. Go to the [Google Cloud's Agent Platform Model Garden](https://console.cloud.google.com/vertex-ai/model-garden)
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
goal
±changed
+1−1
claude-codechanged
goal
The [evaluator](#how-evaluation-works) judges your condition against what Claude has surfaced in the conversation. It doesn't run commands or read files independently, so write the
The [evaluator](#how-evaluation-works) judges your condition against what Claude has surfaced in the conversation. It doesn't run commands or read files independently, so write the
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
github-actions-cloud-providers
±changed
+2−2
claude-codechanged
github-actions-cloud-providers
Beyond the prerequisites, you create four things: a GitHub identity for the Claude Code GitHub Action, the cloud-side trust configuration, the repository secrets, and the workflow
Beyond the prerequisites, you create a GitHub identity for the Claude Code GitHub Action, the cloud-side trust configuration, the repository secrets, and the workflow file. The ste
* **Custom GitHub App**: create your own app, described below, when you want only the three permissions the Claude Code GitHub Action uses rather than the [official app's full
* **Custom GitHub App**: create your own app when you want only the three permissions the Claude Code GitHub Action uses rather than the [official app's full set](/docs/en/gith
+2 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
features-overview
±changed
+1−1
claude-codechanged
features-overview
Skills are extra capabilities in Claude's toolkit. They can be reference material (like an API style guide) or invocable workflows you trigger with `/<name>` (like `/deploy`).
Skills are extra capabilities in Claude's toolkit. They can be reference material (like an API style guide) or invocable workflows you trigger with `/<name>` (like `/deploy`).
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
errors
±changed
+131−11
claude-codechanged
errors
| `Cloud sessions are disabled by your organization's policy`
| `Couldn't verify your organization's policy for cloud sessions`
| `No available ports for OAuth redirect`
| `GitHub isn't connected to your Claude account, so this repository can't be cloned in the cloud`
| `Its agent definition was not restored: the folder its definition file came from is not trusted`
+131 −11 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
env-vars
±changed
+2−1
claude-codechanged
env-vars
| `CLAUDE_CODE_BASH_EDIT_DIFF` | Set to `0` to turn off the [diff of the files a Bash command changed](/docs/en/hooks#bash), or `1` to record it in every
| `CLAUDE_CODE_SESSIONEND_HOOKS_TIMEOUT_MS` | Override the time budget in milliseconds for [SessionEnd](/docs/en/hooks#sessionend) hooks. Applies to session exit, `/c
| `CLAUDE_CODE_SESSIONEND_HOOKS_TIMEOUT_MS` | Override the time budget in milliseconds for [SessionEnd](/docs/en/hooks#sessionend) hooks. The value is also the timeou
+2 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
discover-plugins
±changed
+1−1
claude-codechanged
discover-plugins
If you see `Executable not found in $PATH` in the `/plugin` Errors tab after installing a plugin, install the required binary from the table above.
If you see `Executable not found in $PATH` in the `/plugin` Errors tab after installing a plugin, install the binary the [code intelligence](#code-intelligence) table lists for t
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
devcontainer
±changed
+2−2
claude-codechanged
devcontainer
The Dev Container Feature always installs the latest Claude Code release. To pin a specific Claude Code version for reproducible builds, install it from your Dockerfile with `npm i
The Dev Container Feature always installs the latest Claude Code release. To pin a specific Claude Code version for reproducible builds, install it from your Dockerfile with `npm i
* [Server-managed settings](/docs/en/server-managed-settings): deliver managed policy from the Claude.ai admin console so engineers cannot bypass it by editing repository files
* [Server-managed settings](/docs/en/server-managed-settings): deliver managed policy from the claude.ai admin console so engineers cannot bypass it by editing repository files
+2 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
desktop-ios-simulator
±changed
+1−1
claude-codechanged
desktop-ios-simulator
You and Claude drive the same device, so your taps change the app state Claude sees. To have Claude check a specific screen, navigate to it by tapping, then ask. While Claude is dr
You and Claude drive the same device, so your taps change the app state Claude sees. To have Claude check a specific screen, tap through to it, then ask. While Claude is driving th
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
desktop
±changed
+3−3
claude-codechanged
desktop
The Browser follows the same [site allowlist and blocklist controls](https://support.claude.com/en/articles/13065128-claude-in-chrome-admin-controls) as the Claude in Chrome extens
The Browser follows the same [site allowlist and blocklist controls](https://support.claude.com/en/articles/13065128-claude-in-chrome-admin-controls) as the Claude in Chrome extens
| `autoPort` | boolean | How to handle port conflicts. See below
| `autoPort` | boolean | How to handle port conflicts. See [Port conflicts](#port-conflicts)
Localhost addresses open directly, exactly like the default port address. This includes `localhost`, any `*.localhost` subdomain, `127.0.0.1`, and `::1`. For security, a localhost
+3 −3 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
deep-links
±changed
+1−1
claude-codechanged
deep-links
1. The browser or app hands the URL to your operating system.
1. The browser or app passes the URL to your operating system.
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
code-review
±changed
+3−1
claude-codechanged
code-review
To dismiss a finding without a code change, resolve its thread; replying doesn't dismiss it.
Claude reports the findings as text in the reply in both of these runs, even when a host application requests the findings list described below:
Claude reports the findings as text in the reply in both of these runs, even when a host application requests a findings list:
+3 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
cloud-environments
±changed
+2−2
claude-codechanged
cloud-environments
To get the versions of most of the tools in this table, ask Claude to run `check-tools` in a cloud session. It's a shell command installed on the session VM, not a slash command; y
To get the versions of most of the tools in this table, ask Claude to run `check-tools` in a cloud session. It's a shell command installed on the session VM, not a command you type
* **No cloud-only scoping**: hooks run in both local and cloud sessions. To skip local execution, check the `CLAUDE_CODE_REMOTE` environment variable as shown above.
* **No cloud-only scoping**: hooks run in both local and cloud sessions. To skip local execution, exit early unless the `CLAUDE_CODE_REMOTE` environment variable is `true`, the way
+2 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
claude-security
±changed
+3−3
claude-codechanged
claude-security
While the scan runs, it reports each stage as it starts, with the detail available under [`/workflows`](/docs/en/workflows). Results land in a timestamped directory in your rep
While the scan runs, it reports each stage as it starts, with the detail available under [`/workflows`](/docs/en/workflows). Results are written to a timestamped directory in y
Run `/claude-security` again and pick **Suggest patches**, then choose which findings to address. Reviewed patches land in the report's `patches/` folder; [Fix findings](#fix-f
Run `/claude-security` again and pick **Suggest patches**, then choose which findings to address. Reviewed patches are written to the report's `patches/` folder; [Fix findings]
Applying a patch is always your decision. Patches land in the report's `patches/` folder, one `F<n>.patch` per finding with a note beside it explaining the change. Apply one from y
+3 −3 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
claude-platform-on-aws
±changed
+1−1
claude-codechanged
claude-platform-on-aws
The `/login` and `/logout` commands don't sign you into a Claude.ai subscription for Claude Platform on AWS. Authentication runs through your AWS credentials or workspace API key
The `/login` and `/logout` commands don't sign you into a claude.ai subscription for Claude Platform on AWS. Authentication runs through your AWS credentials or workspace API key
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
claude-directory
±changed
+1−1
claude-codechanged
claude-directory
Beyond the config you author, `~/.claude` holds data Claude Code writes during sessions. These files are plaintext. Anything that passes through a tool lands in a transcript on dis
Beyond the config you author, `~/.claude` holds data Claude Code writes during sessions. These files are plaintext. Anything that passes through a tool is written to a transcript o
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
claude-apps-gateway-deploy
±changed
+12−4
claude-codechanged
claude-apps-gateway-deploy
**Deploy on your private network.** Claude Code only connects to a gateway whose address is private. This is a security guard, because a trusted gateway can push settings that ru
**Deploy on your private network.** Claude Code only connects to a gateway whose address is private. This is a security guard, because a trusted gateway can push settings that ru
* **Audit events**: single-line JSON per security-relevant event. Pipe stderr to your log aggregator. The events emitted include `config.load`, `session.mint`, `session.refresh`, `
* **Audit events**: single-line JSON per security-relevant event. Pipe stderr to your log aggregator.
+12 −4 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
claude-apps-gateway-config
±changed
+14−3
claude-codechanged
claude-apps-gateway-config
The `managed` block defines role-based access policies keyed on IdP groups or email domain. Policies are evaluated in order; the first match is selected, then merged onto the `matc
The `managed` block defines role-based access policies keyed on IdP groups or email domain. Policies are evaluated in order; the first match is selected, then merged onto the `matc
* **Logs and traces**: can carry full bash commands, tool inputs, and file paths, covering anything Claude Code does on a developer's machine
* **Logs and traces**: can carry full Bash commands, tool inputs, and file paths, covering anything Claude Code does on a developer's machine
+14 −3 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
claude-apps-gateway
±changed
+56−11
claude-codechanged
claude-apps-gateway
**Deploy on your private network.** Claude Code only connects to a gateway whose address is private. This is a security guard, because a trusted gateway can push settings that ru
**Deploy on your private network.** Claude Code only connects to a gateway whose address is private. This is a security guard, because a trusted gateway can push settings that ru
| You need | Details
| --------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------
| Claude Code v2.1.195 or later | The `claude gateway` subcommand and the gateway sign-in flow ship in v2.1.195. Earlier public builds don't include them. Both the machin
+56 −11 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
checkpointing
±changed
+1−1
claude-codechanged
checkpointing
Checkpointing does not track files modified by bash commands. For example, if Claude Code runs:
Checkpointing does not track files modified by Bash commands. For example, if Claude Code runs:
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
channels-reference
±changed
+3−3
claude-codechanged
channels-reference
The permission relay examples later on this page import `zod` directly, so it installs alongside the MCP SDK. Create a new directory and install both:
The [permission relay](#relay-permission-prompts) examples import `zod` directly, so it installs alongside the MCP SDK. Create a new directory and install both:
The file does three things in order:
The file configures the server, connects over stdio, and starts an HTTP listener, in that order:
The local dialog closes, the `reply` tool runs, and Claude's reply lands in the stream.
+3 −3 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
channels
±changed
+6−2
claude-codechanged
channels
When the install asks for an installation scope, choose the user scope option so the plugin is available across all your projects. If the install summary reports `Run /relo
When the install asks for an installation scope, choose the user scope option so the plugin is available across all your projects.
If the install summary reports `Run /reload-plugins to activate.`, you don't need to act on it here, because restarting in the next step picks up the plugin.
When the install asks for an installation scope, choose the user scope option so the plugin is available across all your projects. If the install summary reports `Run /reload-p
+6 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
best-practices
±changed
+1−1
claude-codechanged
best-practices
This guide covers patterns that have proven effective across Anthropic's internal teams and for engineers using Claude Code across various codebases, languages, and environments. F
This guide covers patterns that have proven effective across Anthropic's internal teams and for engineers using Claude Code across various codebases, languages, and environments. F
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
authentication
±changed
+5−5
claude-codechanged
authentication
Claude Code supports multiple authentication methods depending on your setup. Individual users can log in with a Claude.ai account, while teams can use Claude for Teams or Enterpri
Claude Code supports multiple authentication methods depending on your setup. Individual users can log in with a claude.ai account, while teams can use Claude for Teams or Enterpri
* **Claude Pro or Max subscription**: log in with your Claude.ai account. Subscribe at [claude.com/pricing](https://claude.com/pricing?utm_source=claude_code\&utm_medium=docs\&utm_
* **Claude for Teams or Enterprise**: log in with the Claude.ai account your team admin invited you to.
* **Claude Pro or Max subscription**: log in with your claude.ai account. Subscribe at [claude.com/pricing](https://claude.com/pricing?utm_source=claude_code\&utm_medium=docs\&utm_
+5 −5 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
artifacts
±changed
+2−2
claude-codechanged
artifacts
If Claude tells you it can't read comments, check three things:
If Claude tells you it can't read comments, confirm your version, your session, and your feature-flag setting:
An artifact can act as a lightweight editor for a decision you then hand back to Claude. Ask for an export control that produces text you can paste into the terminal, so the result
An artifact can act as a lightweight editor for a decision you then send back to Claude. Ask for an export control that produces text you can paste into the terminal, so the result
+2 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
amazon-bedrock
±changed
+2−2
claude-codechanged
amazon-bedrock
2. Navigate to the [Amazon Bedrock console](https://console.aws.amazon.com/bedrock/)
2. Go to the [Amazon Bedrock console](https://console.aws.amazon.com/bedrock/)
Mantle is an Amazon Bedrock endpoint that serves Claude models through the native Anthropic API shape rather than the Amazon Bedrock Invoke API. It uses the same AWS credentials, I
Mantle is an Amazon Bedrock endpoint that serves Claude models through the native Anthropic API shape rather than the Amazon Bedrock Invoke API. It uses the same [AWS credentials](
+2 −2 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-view
±changed
+1−0
claude-codechanged
agent-view
| v2.1.268 | After the first `←` shows `Press ← again to open agents`, or `Press ← again to go back to agents` in an attached session, [the first press that comes at least a second
+1 −0 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/user-input
±changed
+6−6
claude-codechanged
agent-sdk/user-input
| Field | Description |
| ------------- | -------------------------------------------------------------------------------------------------------------------------------------- |
| `question` | The full question text to display |
| `header` | Short label for the question (max 12 characters) |
| `options` | Array of 2-4 choices, each with `label` and `description`. TypeScript: optionally `preview` (see [below](#option-previews-typescript)) |
+6 −6 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/typescript-v2-preview
±changed
+1−1
claude-codechanged
agent-sdk/typescript-v2-preview
V2 was an experimental session API that removed the need for async generators and yield coordination. Instead of managing generator state across turns, each turn was a separate `se
V2 was an experimental session API that removed the need for async generators and yield coordination. Instead of managing generator state across turns, each turn was a separate `se
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/typescript
±changed
+1−1
claude-codechanged
agent-sdk/typescript
| `autoAllowBashIfSandboxed` | `boolean` | `true` | Auto-approve bash commands when sandbox is enabled
| `autoAllowBashIfSandboxed` | `boolean` | `true` | Auto-approve Bash commands when sandbox is enabled
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/skills
±changed
+1−1
claude-codechanged
agent-sdk/skills
By default, both you and Claude can invoke any skill. You can restrict either path through the skill's [frontmatter](/docs/en/skills#control-who-invokes-a-skill). For a definition
By default, both you and Claude can invoke any skill. You can restrict either path through the skill's [frontmatter](/docs/en/skills#control-who-invokes-a-skill). For definitions o
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/python
±changed
+7−7
claude-codechanged
agent-sdk/python
| Parameter | Type | Description |
| :------------- | :---------------------------------------------- | :------------------------------------------------------------------ |
| `name` | `str` | Unique identifier for the tool |
| `description` | `str` | Human-readable description of what the tool does |
| `input_schema` | `type \| dict[str, Any]` | Schema defining the tool's input parameters (see below) |
+7 −7 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/plugins
±changed
+1−1
claude-codechanged
agent-sdk/plugins
Plugins allow you to extend Claude Code with custom functionality that can be shared across projects. Through the Agent SDK, you can programmatically load plugins from local direct
Plugins let you extend Claude Code with custom functionality that can be shared across projects. Through the Agent SDK, you can programmatically load plugins from local directories
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/file-checkpointing
±changed
+1−1
claude-codechanged
agent-sdk/file-checkpointing
With the `replay-user-messages` option set (shown above), each user message in the response stream has a UUID that serves as a checkpoint.
With the `replay-user-messages` option set, each user message in the response stream has a UUID that serves as a checkpoint.
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/custom-tools
±changed
+1−1
claude-codechanged
agent-sdk/custom-tools
These snippets reuse the `weatherServer` from the [example above](#weather-tool-example) to ask Claude what the weather is in a specific location.
These snippets reuse the `weatherServer` from the [weather tool example](#weather-tool-example) to ask Claude what the weather is in a specific location.
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/claude-code-features
±changed
+1−1
claude-codechanged
agent-sdk/claude-code-features
`CLAUDE.md` files and `.claude/rules/*.md` files give your agent persistent context about your project: coding conventions, build commands, architecture decisions, and instructions
`CLAUDE.md` files and `.claude/rules/*.md` files give your agent persistent context about your project: coding conventions, build commands, architecture decisions, and instructions
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
agent-sdk/agent-loop
±changed
+1−1
claude-codechanged
agent-sdk/agent-loop
All result subtypes carry `total_cost_usd`, `usage`, `num_turns`, and `session_id` so you can track cost and resume even after errors. Two things to guard for:
All result subtypes carry `total_cost_usd`, `usage`, `num_turns`, and `session_id` so you can track cost and resume even after errors. Guard for these cases:
+1 −1 lines
Recorded here 15 Sep 2026 21:37 UTC
claude-code
devcontainer
±changed
+53−59
claude-codechanged
devcontainer
<Warning>
While the dev container provides substantial protections, no system is completely immune to all attacks.
When executed with `--dangerously-skip-permissions`, dev containers do not prevent a malicious project from exfiltrating anything accessible inside the container, including the C
Only use dev containers when developing with trusted repositories, and monitor Claude's activities.
Avoid mounting host secrets such as `~/.ssh` or cloud credential files into the container; prefer repository-scoped or short-lived tokens.
+53 −59 lines
Upstream edited 15 Sep 2026 21:25 UTC Recorded here 20 Oct 2026 18:37 UTC
claude-code
devcontainer
±changed
+4−4
claude-codechanged
devcontainer
| File | Purpose
| ---------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------
| File | Purpose |
| - | - |
| [`Dockerfile`](https://github.com/anthropics/claude-code/blob/main/.devcontainer/Dockerfile) | Base image, development tools, and the Claude Code install
+4 −4 lines
Upstream edited 15 Sep 2026 21:25 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
self-hosted-environments-configuration
±changed
+48−48
claude-codechanged
self-hosted-environments-configuration
| Variable | Description
| :---------------------------------- | :-------------------------------------------------------------------------------------------------------------------------------------------
| `CLAUDE_CODE_SESSION_ACCESS_TOKEN` | The session JWT, prefixed `sk-ant-cc-`. Its `act` claim identifies the session creator, with the creator's email and upstream identity-provi
| `CCR_SESSION_ACCOUNT_EMAIL` | The session creator's email, pre-extracted by the runner from the token's `act.email` claim without signature verification. Suitable for lab
| `CLAUDE_RUNNER_CLIENT_PLATFORM` | The client surface that created the session, such as `web_claude_ai`, `desktop_app`, `ios`, `claude_code_cli`, or `scheduled_trigger`. Anthr
+48 −48 lines
Upstream edited 15 Sep 2026 21:22 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
deep-links
±changed
+10−10
claude-codechanged
deep-links
| Parameter | Description
| --------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------
| `q` | Text to pre-fill in the prompt box. [URL-encode](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/encodeURIComponent) the value. Use `
| `cwd` | Absolute path to use as the working directory. Network and UNC paths are rejected, and so are paths that contain `..` segments or invisible or bidirectional control c
| `repo` | A GitHub `owner/name` slug. Claude Code resolves it to a local clone it has seen before and starts there. If you have no matching clone, the session opens in your hom
+10 −10 lines
Upstream edited 15 Sep 2026 21:22 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
github-actions-cloud-providers
±changed
+17−17
claude-codechanged
github-actions-cloud-providers
[Claude Code GitHub Actions](/docs/en/github-actions) calls the Claude API by default. To route inference through your own cloud account instead, set the Claude Code GitHub Action'
[Claude Code GitHub Actions](/docs/en/github-actions.md) calls the Claude API by default. To route inference through your own cloud account instead, set the Claude Code GitHub Acti
> This page builds on the [GitHub Actions setup](/docs/en/github-actions#setup). It assumes you already know the workflow file and the `anthropics/claude-code-action` step, and cov
> This page builds on the [GitHub Actions setup](/docs/en/github-actions.md#setup). It assumes you already know the workflow file and the `anthropics/claude-code-action` step, and
* **Amazon Bedrock**: access granted to Claude models. Cross-region inference profiles, such as the `us.` model IDs in this page's examples, need access granted in every region o
+17 −17 lines
Upstream edited 15 Sep 2026 21:21 UTC Recorded here 22 Oct 2026 18:07 UTC
claude-code
agent-sdk/typescript-v2-preview
±changed
+5−5
claude-codechanged
agent-sdk/typescript-v2-preview
> To migrate, use the [`query()` API](/docs/en/agent-sdk/typescript) and the [session options](/docs/en/agent-sdk/sessions) it accepts. Pass an `AsyncIterable<SDKUserMessage>` for
> To migrate, use the [`query()` API](/docs/en/agent-sdk/typescript) and the [session options](/docs/en/agent-sdk/sessions.md) it accepts. Pass an `AsyncIterable<SDKUserMessage>` f
> The SDK bundles a native Claude Code binary for your platform as an optional dependency, so most installs need no separate Claude Code install. See the [quickstart's install note
> The SDK bundles a native Claude Code binary for your platform as an optional dependency, so most installs need no separate Claude Code install. See the [quickstart's install note
The V2 session API does not support every V1 feature. The following require the [V1 SDK](/docs/en/agent-sdk/typescript):
+5 −5 lines
Upstream edited 15 Sep 2026 21:21 UTC Recorded here 22 Oct 2026 18:07 UTC
claude-code
github-actions-cloud-providers
±changed
+258−264
claude-codechanged
github-actions-cloud-providers
<Info>
This page builds on the [GitHub Actions setup](/docs/en/github-actions#setup). It assumes you already know the workflow file and the `anthropics/claude-code-action` step, and cov
</Info>
> [!NOTE]
>
+258 −264 lines
Upstream edited 15 Sep 2026 21:21 UTC Recorded here 20 Oct 2026 18:37 UTC
claude-code
agent-sdk/typescript-v2-preview
±changed
+8−8
claude-codechanged
agent-sdk/typescript-v2-preview
<Warning>
The V2 session API is no longer supported. TypeScript Agent SDK 0.3.142 removes `unstable_v2_createSession`, `unstable_v2_resumeSession`, `unstable_v2_prompt`, and the `SDKSessio
To migrate, use the [`query()` API](/docs/en/agent-sdk/typescript) and the [session options](/docs/en/agent-sdk/sessions) it accepts. Pass an `AsyncIterable<SDKUserMessage>` for
</Warning>
+8 −8 lines
Upstream edited 15 Sep 2026 21:21 UTC Recorded here 20 Oct 2026 18:37 UTC
claude-code
github-actions-cloud-providers
±changed
+10−10
claude-codechanged
github-actions-cloud-providers
| Secret | Needed for | Value |
| -------------------------------- | ----------------------------- | ------------------------------------------- |
| `AWS_ROLE_TO_ASSUME` | Amazon Bedrock | The ARN of the IAM role |
| `GCP_WORKLOAD_IDENTITY_PROVIDER` | Google Cloud's Agent Platform | The provider's full resource name |
| `GCP_SERVICE_ACCOUNT` | Google Cloud's Agent Platform | The service account's email address |
+10 −10 lines
Upstream edited 15 Sep 2026 21:21 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
agent-sdk/agent-loop
±changed
+48−48
claude-codechanged
agent-sdk/agent-loop
| Category | Tools | What they do |
| :------------------ | :-------------------------------------------------------------- | :-------------------------------------------------------------------------- |
| **File operations** | `Read`, `Edit`, `Write` | Read, modify, and create files |
| **Search** | `Glob`, `Grep` | Find files by pattern, search content with regex |
| **Execution** | `Bash` | Run shell commands, scripts, git operations |
+48 −48 lines
Upstream edited 15 Sep 2026 21:21 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
skills
±changed
+7−3
claude-codechanged
skills
Injected commands never prompt for permission. When a command's permission check returns anything other than allow, Claude Code aborts the invocation. This includes a rule that wou
To keep an unmatched command from aborting here, pre-approve it with [`allowed-tools`](#pre-approve-tools-for-a-skill). A matching ask or deny rule still aborts the invocation rega
#### Permission checks on injected commands
+7 −3 lines
Recorded here 15 Sep 2026 21:07 UTC
claude-code
errors
±changed
+9−5
claude-codechanged
errors
The quoted command varies between runs: the review starts several `git` commands against `origin/HEAD` at once and reports whichever fails first, so you may see `git log` or a diff
Claude Code shows the same error for any skill that [injects dynamic context](/docs/en/skills#when-an-injected-command-fails). A failed injected command aborts that skill's invocat
* `Shell command permission check failed for pattern "..."`: the command's permission check returned something other than allow. Injected commands never prompt, so the invocation a
+9 −5 lines
Recorded here 15 Sep 2026 21:07 UTC
claude-code
changelog
±changed
+67−0
claude-codechanged
changelog
<Update label="2.1.273" description="September 15, 2026">
* Added `x-claude-code-request-class`, `x-claude-code-agent-type`, `x-claude-code-prev-tool-durations`, `x-claude-code-compaction` and `x-claude-code-context-compacted` request h
* Added a notification when an MCP server disconnects mid-session and automatic reconnection gives up, pointing at `/mcp`
* Added forking a session started with `claude --remote-control` or `/remote-control` from the Claude app; the fork runs as a background session on your computer
+67 −0 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/typescript
±changed
+8−8
claude-codechanged
agent-sdk/typescript
| `fallbackModel` | `string`
| `fallbackModel` | `string`
| `maxBudgetUsd` | `number`
| `maxBudgetUsd` | `number`
| `settings` | `string \| Settings`
+8 −8 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/subagents
±changed
+5−5
claude-codechanged
agent-sdk/subagents
| Limit | Set it with | Default |
| :---------- | :------------------------------------------------------- | :----------------------------------------------------------------------------------------------------- |
| Depth | [`CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH`](/docs/en/env-vars) | `3` layers of subagents below your main agent. `1` stops your subagents from spawning any of their o
| Concurrency | [`CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS`](/docs/en/env-vars) | `20` subagents running at once, counting every subagent Claude spawns with the Agent tool
| Spend | `maxBudgetUsd` in TypeScript, `max_budget_usd` in Python | No limit. Compared against `total_cost_usd`, so subagent requests count |
+5 −5 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/quickstart
±changed
+4−1
claude-codechanged
agent-sdk/quickstart
[uv](https://docs.astral.sh/uv/) is a fast Python package manager that handles virtual environments automatically:
[Install uv](https://docs.astral.sh/uv/), a fast Python package manager that handles virtual environments automatically. Then initialize a project and add the SDK:
Each of these snippets sets fields on the same options object. For more information, see [Configure your agent](/docs/en/agent-sdk/configuration).
* **[Configure your agent](/docs/en/agent-sdk/configuration)**: compose the options object and find the page that covers each setting
+4 −1 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/python
±changed
+53−53
claude-codechanged
agent-sdk/python
| `set_model(model)` | Change the model for the current session. Pass `None` to reset to default
| `set_model(model)` | Change the model for the current session. Pass `None` to reset to [Claude Code's default model](/docs/en/model-config)
| Property | Type | Default | Description
| :---------------------------- | :------------------------------------------------------------------------------------ | :--------------------------------- | :--------------------
| `tools` | `list[str] \| ToolsPreset \| None` | `None` | Tools configuration.
+53 −53 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/cost-tracking
±changed
+1−1
claude-codechanged
agent-sdk/cost-tracking
`maxBudgetUsd`, or `max_budget_usd` in Python, is compared against the same running total, so a `/clear` also starts the budget over.
`maxBudgetUsd` (TypeScript) or `max_budget_usd` (Python) is compared against the same running total, so a `/clear` also starts the budget over.
+1 −1 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/configuration
+new page
+293−0
claude-codenew page
agent-sdk/configuration
No line of the body moved in this read.
+293 −0 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/claude-code-features
±changed
+1−1
claude-codechanged
agent-sdk/claude-code-features
When you omit `settingSources`, `query()` reads the same filesystem settings as the Claude Code CLI: user, project, and local settings, CLAUDE.md files, and `.claude/` skills, agen
When you omit `settingSources`, `query()` reads the same filesystem settings as the Claude Code CLI: user, project, and local settings, CLAUDE.md files, and `.claude/` skills, agen
+1 −1 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/agent-loop
±changed
+2−2
claude-codechanged
agent-sdk/agent-loop
With [streaming input](/docs/en/agent-sdk/streaming-vs-single-mode), a message that is still queued when a turn ends at the max-turns limit stays queued. Claude Code doesn't add it
With [streaming input](/docs/en/agent-sdk/streaming-vs-single-mode), a message that is still queued when a turn ends at the max-turns limit stays queued. Claude Code doesn't add it
If you don't set `model`, the SDK uses Claude Code's default, which depends on your authentication method and subscription. Set it explicitly (for example, `model="claude-sonnet-5"
Set the `model` option to choose which model runs the session. For more information, see [Choose a model](/docs/en/agent-sdk/configuration#choose-a-model).
+2 −2 lines
Recorded here 15 Sep 2026 20:37 UTC
claude-code
agent-sdk/session-storage
±changed
+16−16
claude-codechanged
agent-sdk/session-storage
> Mirror session transcripts to S3, Redis, or your own backend so other hosts can resume your sessions.
By default, the SDK writes session transcripts to JSONL files under `~/.claude/projects/` on the local filesystem. A `SessionStore` adapter lets you mirror those transcripts to you
> Mirror Agent SDK session transcripts to your own object store, key-value store, or database so other hosts can resume your sessions.
+16 −16 lines
Recorded here 15 Sep 2026 20:07 UTC
claude-code
agent-sdk/hosting
±changed
+3−3
claude-codechanged
agent-sdk/hosting
declare const sessionStore: SessionStore; // S3, Redis, Postgres, or your own adapter
declare const sessionStore: SessionStore; // an object store, key-value store, database, or your own adapter
session_store: SessionStore = ... # S3, Redis, Postgres, or your own adapter
session_store: SessionStore = ... # an object store, key-value store, database, or your own adapter
Default local disk is lost on restart, scale-down, or a move to a different node. For any session a user expects to resume, mirror the transcript to durable storage with a [`Sessio
+3 −3 lines
Recorded here 15 Sep 2026 20:07 UTC
claude-code
agent-sdk/session-storage
±changed
+13−13
claude-codechanged
agent-sdk/session-storage
| Method | Required | Called when
| :--------------------- | :------- | :---------------------------------------------------------------------------------------------------------------------------------------------
| `append` | Yes | After each batch of transcript entries is written locally. Entries are JSON-safe objects, one per line in the local JSONL.
| `load` | Yes | Before the subprocess spawns when `resume` is set or `continue: true` resolves the newest store session, and once per session when listing fal
| `listSessions` | No | By `listSessions({ sessionStore })` and by `query()`/`startup()` with `continue: true`. If undefined, `continue: true` throws, and `listSessio
+13 −13 lines
Upstream edited 15 Sep 2026 19:41 UTC Recorded here 28 Sep 2026 23:37 UTC
claude-code
code-review
±changed
+1−1
claude-codechanged
code-review
The agents read the file's text as-is, so `REVIEW.md` is plain instructions: [`@` import syntax](/docs/en/memory#import-additional-files) is not expanded, and referenced files are
Put the rules you want enforced directly in `REVIEW.md`.
+1 −1 lines
Recorded here 15 Sep 2026 19:37 UTC
claude-docs
third-party/claude-desktop/admin-console
±changed
+28−1
claude-docschanged
third-party/claude-desktop/admin-console
The console stores no API keys, tokens, or secrets, and refuses them anywhere in the configuration, including in request headers and MCP server settings. Users authenticate to your
The console stores no API keys, tokens, or secrets, and refuses them anywhere in the configuration, including in request headers and MCP server settings. Users authenticate to your
### Choose how users authenticate to your provider
+28 −1 lines
Recorded here 15 Sep 2026 18:46 UTC
claude-docs
government/security/security-and-data-handling
±changed
+3−7
claude-docschanged
government/security/security-and-data-handling
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
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
Not in Chat or Cowork. Administrators can turn the built-in **Web search**, **Web fetch**, and **Shell commands** tools on or off, or require approval on every use, but cannot
Not in Chat or Cowork. Administrators can turn the built-in **Web search**, **Web fetch**, and **Shell commands** tools on or off, or require approval on every use, but cannot
For most tools, users who see an approval prompt can choose **Always allow**, which suppresses that prompt for them going forward. Administrators can remove that option on the
+3 −7 lines
Recorded here 15 Sep 2026 18:46 UTC
claude-docs
government/config/settings
±changed
+1−1
claude-docschanged
government/config/settings
The **Shell commands** card controls whether Claude can run shell commands during tasks in Claude Desktop. It is on by default, and turning it off also turns off Advanced file anal
The **Shell commands** card controls whether Claude can run shell commands during tasks in Claude Desktop. It is on by default, and turning it off also turns off Advanced file anal
+1 −1 lines
Recorded here 15 Sep 2026 18:46 UTC
claude-docs
government/config/settings
±changed
+1−1
claude-docschanged
government/config/settings
How much of Claude Desktop's own event log goes to your collector, in addition to the usage telemetry from Chat, Cowork, and Code. These records arrive under the `claude-desktop` s
How much of Claude Desktop's own event log goes to your collector, in addition to the usage telemetry from Chat, Cowork, and Code. These records arrive under the `claude-desktop` s
+1 −1 lines
Recorded here 15 Sep 2026 18:37 UTC
claude-docs
third-party/claude-desktop/telemetry
±changed
+4−4
claude-docschanged
third-party/claude-desktop/telemetry
Without this host reachable, Chat conversations, Cowork tasks, and Code sessions cannot start on a device that has not yet downloaded these components. App updates often change one
Without this host reachable, Chat conversations, Cowork tasks, and Code sessions cannot start on a device that has not yet downloaded these components. App updates often change one
With `disableEssentialTelemetry`, `disableNonessentialTelemetry`, `disableNonessentialServices`, and `disableAutoUpdates` all set to `true`, the desktop application makes **no outb
An app that receives its configuration from the [Enterprise Admin Console](/docs/third-party/claude-desktop/admin-console) still connects to Anthropic with all four of these keys s
+4 −4 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
third-party/claude-desktop/overview
±changed
+2−2
claude-docschanged
third-party/claude-desktop/overview
Because inference runs in your cloud tenant, Claude Desktop on 3P operates inside whatever compliance boundary your provider and region give you. The desktop application itself con
Because inference runs in your cloud tenant, Claude Desktop on 3P operates inside whatever compliance boundary your provider and region give you. The desktop application itself con
With Anthropic-bound telemetry, non-essential services, and updates all disabled, the only remaining Anthropic-operated egress is `downloads.claude.ai` for the VM workspace bundle
With Anthropic-bound telemetry, non-essential services, updates, and the [model catalog fetch](/docs/third-party/claude-desktop/configuration#modelcatalogenabled) all disabled, the
+2 −2 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
third-party/claude-desktop/mdm
±changed
+1−1
claude-docschanged
third-party/claude-desktop/mdm
`downloads.claude.ai` is required to run the app regardless of your configuration: it serves the VM workspace bundle and the latest Claude Code binary, fetched at session start.
`downloads.claude.ai` is required to run the app regardless of your configuration: it serves the VM workspace bundle and the latest Claude Code binary, fetched at session start.
+1 −1 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
third-party/claude-desktop/installation
±changed
+1−1
claude-docschanged
third-party/claude-desktop/installation
Pair the offline installer with [`disableAutoUpdates`](/docs/third-party/claude-desktop/configuration#disableautoupdates): the app cannot reach the update feed from an air-gapped n
Pair the offline installer with [`disableAutoUpdates`](/docs/third-party/claude-desktop/configuration#disableautoupdates). The app cannot reach the update feed from an air-gapped n
+1 −1 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
third-party/claude-desktop/configuration
±changed
+6−5
claude-docschanged
third-party/claude-desktop/configuration
```json theme={null}
```json theme={null} theme={null} theme={null} theme={null} theme={null}
```json theme={null}
```json theme={null} theme={null} theme={null} theme={null} theme={null}
```json theme={null}
+6 −5 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
government/security/security-and-data-handling
±changed
+1−1
claude-docschanged
government/security/security-and-data-handling
Connectors are called from the desktop application, outside the sandbox. The built-in Microsoft 365 connector and administrator-added connectors call their endpoints directly f
Connectors are called from the desktop application, outside the sandbox. The built-in Microsoft 365 connector and administrator-added connectors call their endpoints directly f
+1 −1 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
government/desktop/plugins
±changed
+7−6
claude-docschanged
government/desktop/plugins
In Claude for Government, plugins reach you in four ways:
In Claude for Government, plugins reach you in three ways:
* You upload a plugin file you have.
* You ask Claude to create a plugin with you.
* You add a plugin marketplace and install plugins from it.
+7 −6 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
government/deploy-desktop/windows-checklist
±changed
+4−0
claude-docschanged
government/deploy-desktop/windows-checklist
### If some devices are not ready for Cowork
You can deploy Claude Desktop to devices that do not yet meet the [Cowork virtualization](#cowork-virtualization) requirements. Before you deploy to those devices, have a tenant ad
+4 −0 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
government/deploy-desktop/configure
±changed
+12−12
claude-docschanged
government/deploy-desktop/configure
* **The app is current.** The configuration mechanism on this page requires Claude Desktop 1.10628.0 or later.
* **The app is current.** Deploy Claude Desktop 1.46388.1 or later.
| What the app receives | Where it is set
| ---------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------
| Which of Chat, Cowork, and Code the user can open, and whether Advanced file analysis is on in Chat | [Product availability](/docs/government/config/settings#product-availabilit
+12 −12 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
government/config/settings
±changed
+11−1
claude-docschanged
government/config/settings
> Reference for the product settings on the Config page in Claude for Government, including session timeout, maximum session length, organization instructions, telemetry, automatic
> Reference for the product settings on the Config page in Claude for Government, including session timeout, maximum session length, organization instructions, telemetry, automatic
### Restart deadline for configuration changes
How long a member can put off restarting Claude Desktop after the app detects a configuration change, as a whole number of hours from 0 to 336 (14 days). When the deadline passes,
+11 −1 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
government/config/plugins-and-connectors
±changed
+1−1
claude-docschanged
government/config/plugins-and-connectors
In Claude Desktop, members can also add plugins of their own, by uploading a plugin file or having Claude create one, as described in [Plugins in Claude Desktop](/docs/government/d
Two switches, **Let members add plugin marketplaces** and **Let members add their own plugins**, control whether members can also add plugins of their own in Claude Desktop, as des
+1 −1 lines
Recorded here 15 Sep 2026 18:07 UTC
claude-docs
third-party/claude-desktop/overview
±changed
+8−8
claude-docschanged
third-party/claude-desktop/overview
| Component | Standard Claude Desktop | Claude Desktop on 3P
| ---------------------- | -------------------------- | ----------------------------------------------------------------------------------------------------------------------------
| Model inference | Anthropic API | Your configured provider endpoint (Google Cloud's Agent Platform, Amazon Bedrock, Microsoft Foundry, or gateway), or the Ant
| Web application | Loaded from claude.ai | Bundled inside the desktop app
| User identity | Anthropic account | Local device identity only (Anthropic account when managed from the claude.ai admin console)
+8 −8 lines
Upstream edited 15 Sep 2026 17:58 UTC Recorded here 28 Sep 2026 22:07 UTC
api
test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks
±changed
+5−1
apichanged
test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks
, Anthropic runs additional classifiers that detect potential prompt injections in screenshots and steer Claude to ask for user confirmation before acting. See that page for deta
or the
[browser use tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/browser-use-tool)
+5 −1 lines
Recorded here 15 Sep 2026 17:37 UTC
api
release-notes/overview
±changed
+4−0
apichanged
release-notes/overview
### September 14, 2026
* The Messages API can now [compact a conversation on demand](https://platform.claude.com/docs/en/build-with-claude/compaction#compact-on-demand-with-the-compaction-parameter) on t
+4 −0 lines
Recorded here 15 Sep 2026 17:37 UTC
api
models/fable-5-1/migration-guide
±changed
+2−2
apichanged
models/fable-5-1/migration-guide
1. Capture the exact request bodies it sends over a few normal turns, including a compaction or a tool change if your product has them. For each pair of consecutive requests, co
1. Capture the exact request bodies it sends over a few normal turns, including a compaction or a tool change if your product has them. For each pair of consecutive requests, co
5. **Trim context on the server, or compact in a shape that carries no stale thinking:** If your code truncates or summarizes older turns on the client, the simplest fix is to move
5. **Trim context on the server, or compact in a shape that carries no stale thinking:** If your code truncates or summarizes older turns on the client, the simplest fix is to move
+2 −2 lines
Recorded here 15 Sep 2026 17:37 UTC
api
build-with-claude/preserved-thinking
±changed
+59−36
apichanged
build-with-claude/preserved-thinking
| Change between requests | Later
| -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -----
| Append messages at the end | Valid
| Add a tool with `defer_loading: true` that nothing has referenced yet | Valid
| Remove `thinking` blocks from the start of the history, from the end, or all of them | Valid
+59 −36 lines
Recorded here 15 Sep 2026 17:37 UTC
api
build-with-claude/mid-conversation-system-messages
±changed
+3−3
apichanged
build-with-claude/mid-conversation-system-messages
`tool_addition` and `tool_removal` are content blocks in the `content` array of a `role: "system"` message, and they can be mixed with `text` blocks in the same message. The messag
`tool_addition` and `tool_removal` are content blocks in the `content` array of a `role: "system"` message, and they can be mixed with `text` blocks in the same message. The messag
The following request declares `get_weather` in `tools`, then withdraws it after the first user turn with a `tool_removal` block. Mid-conversation tool changes are in beta, so the
+3 −3 lines
Recorded here 15 Sep 2026 17:37 UTC
api
build-with-claude/compaction
±changed
+184−0
apichanged
build-with-claude/compaction
The previous steps describe threshold compaction, which most of this page covers. With the `compact-2026-09-04` beta header, you can instead request a summary on demand. That reque
## Compact on demand with the `compaction` parameter
The `compact-2026-09-04` beta adds a second way to compact. Threshold compaction summarizes partway through a request once the threshold you set is reached. With this beta, you ins
+184 −0 lines
Recorded here 15 Sep 2026 17:37 UTC
api
agents-and-tools/tool-use/computer-use-tool
±changed
+3−3
apichanged
agents-and-tools/tool-use/computer-use-tool
Anthropic has trained the model to resist these prompt injections and has added an extra layer of defense. If you use the computer use tools, classifiers will automatically run on
These precautions remain important even with the classifier defense layer in place.
Anthropic has trained the model to resist these prompt injections and has added an extra layer of defense. If you use the computer use tools, classifiers will automatically scan wh
+3 −3 lines
Recorded here 15 Sep 2026 17:37 UTC
Nothing on this page matches that. Press Enter to search every page of the ledger.