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 24 Oct 2026
100 on this page
ReadSourcePageWhatLines
1 Oct 2026
472 pages
+34123 −23853 lines
claude-code
claude-tag
±changed
+2−2
claude-codechanged
claude-tag
No line of the body moved in this read.
+2 −2 lines
Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-tag/users/good-habits
±changed
+1−1
claude-docschanged
claude-tag/users/good-habits
| A correction for this channel that you don't want to raise in the channel | [**Managed instructions**](/docs/claude-tag/admins/managed-by#correct-claude-privately-from-a-managing
| A correction for this channel that you don't want to raise in the channel | [**Managed instructions**](/docs/claude-tag/admins/managed-by#correct-claude-privately-from-a-managing
+1 −1 lines
Upstream edited 1 Oct 2026 23:06 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-tag/admins/restrict-access
±changed
+2−2
claude-docschanged
claude-tag/admins/restrict-access
A channel manager has to be a member of the channel in Slack. The channel's [Configure page](/docs/claude-tag/users/good-habits#configure-claude-for-a-channel), which opens on clau
A channel manager has to be a member of the channel in Slack. The channel's [Configure page](/docs/claude-tag/users/good-habits#configure-claude-for-a-channel), which opens on clau
Owners and Admins can already configure every channel, so you see them as **Already has full access** and can't add them.
Owners can already configure every channel, so you see them as **Already has full access** and can't add them.
+2 −2 lines
Upstream edited 1 Oct 2026 23:06 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-tag/admins/managed-by
±changed
+8−8
claude-docschanged
claude-tag/admins/managed-by
Setting up **Managed by** takes an Owner or Admin in your Claude organization, the roles that see the **Admin** tab on a channel's Configure page. A managing channel isn't a [chann
Setting up **Managed by** takes an Owner of your Claude organization or a [Claude Tag admin](/docs/claude-tag/admins/restrict-access#delegate-claude-tag-administration), the roles
A managing channel is a Slack channel that an admin selected under **Managed by** on another channel's Configure page. Any full member of the workspace who is in a managing channel
A managing channel is a Slack channel that an Owner or a [Claude Tag admin](/docs/claude-tag/admins/restrict-access#delegate-claude-tag-administration) selected under **Managed by*
A managing channel is a Slack channel that an Owner or Admin selected under **Managed by** on another channel's Configure page, and that other channel is the managed channel. When
+8 −8 lines
Upstream edited 1 Oct 2026 23:06 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-tag/admins/customize
±changed
+1−1
claude-docschanged
claude-tag/admins/customize
| Managed by | Which other Slack channels' members can write a channel's standing instructions by asking Claude, including from a private channel. An Owner or Admin in your Claude
| Managed by | Which other Slack channels' members can write a channel's standing instructions by asking Claude, including from a private channel. An Owner or a [Claude Tag admin](
+1 −1 lines
Upstream edited 1 Oct 2026 23:06 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-tag/admins/attach-to-scope
±changed
+1−1
claude-docschanged
claude-tag/admins/attach-to-scope
| Managed instructions | Full workspace members in one of the channel's [managing channels](/docs/claude-tag/admins/managed-by), which an Owner or Admin selects under **Managed by*
| Managed instructions | Full workspace members in one of the channel's [managing channels](/docs/claude-tag/admins/managed-by), which an Owner or a Claude Tag admin selects under
+1 −1 lines
Upstream edited 1 Oct 2026 23:06 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/mods/reference
±changed
+1−1
claude-codechanged
plugins/mods/reference
Elements are the building blocks of a tree a `ui.render` hook returns, and you get them from `$.ui.resolve(e)`. [Build a tree from elements](/docs/en/plugins/mods/interface#build-a
Elements are the building blocks of a tree a `ui.render` hook returns, and you get them from `$.ui.resolve(e)`. [Build a tree from elements](/docs/en/plugins/mods/interface#build-a
+1 −1 lines
Upstream edited 1 Oct 2026 23:04 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/mods/interface
±changed
+2−2
claude-codechanged
plugins/mods/interface
Note: Type a note and press Enter ⏎ add
Note: Type a note and press Enter
This table lists every element:
The [interface gallery](/docs/en/plugins/mods/gallery) has samples and screenshots of most elements. This table lists every element:
+2 −2 lines
Upstream edited 1 Oct 2026 23:04 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/mods/gallery
+new page
+389−0
claude-codenew page
plugins/mods/gallery
No line of the body moved in this read.
+389 −0 lines
Upstream edited 1 Oct 2026 23:04 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
agent-sdk/typescript
±changed
+2−2
claude-codechanged
agent-sdk/typescript
The `context-1m-2025-08-07` beta is retired as of April 30, 2026. Passing this value with Claude Sonnet 4.5 or Sonnet 4 has no effect, and requests that exceed the standard 200k-
On the Claude API, the `context-1m-2025-08-07` beta is retired for Claude Sonnet 4.5 and Claude Sonnet 4. If you still pass it with either model, requests that exceed the standar
| `resolvedModel` | `string \| undefined` | Canonical wire model ID that this entry's `value` resolves to. An alias entry such as `sonnet` resolves to an explicit model ID such as
| `resolvedModel` | `string \| undefined` | The model ID that this entry's `value` resolves to, such as `claude-sonnet-5-5` for the `sonnet` alias entry. Requires Claude Code v2.1.
+2 −2 lines
Upstream edited 1 Oct 2026 23:04 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
agent-sdk/python
±changed
+1−1
claude-codechanged
agent-sdk/python
The `context-1m-2025-08-07` beta is retired as of April 30, 2026. Passing this header with Claude Sonnet 4.5 or Sonnet 4 has no effect, and requests that exceed the standard 200k
On the Claude API, the `context-1m-2025-08-07` beta is retired for Claude Sonnet 4.5 and Claude Sonnet 4. If you still pass it with either model, requests that exceed the standar
+1 −1 lines
Upstream edited 1 Oct 2026 23:04 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-science/connectors-and-skills
±changed
+6−0
claude-docschanged
claude-science/connectors-and-skills
### Update skills imported from GitHub
Skills you import from GitHub don't update automatically, and Claude Science doesn't check for changes on its own. To check, go to **Settings > Skills**, scroll to **Imported**, op
+6 −0 lines
Upstream edited 1 Oct 2026 23:02 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-docs
claude-science/admin-controls
±changed
+1−1
claude-docschanged
claude-science/admin-controls
To let members manage their own skills, host them in a GitHub repository. Put one `skills/<name>/SKILL.md` folder per skill in the repository and share the link. Members import the
To let members manage their own skills, host them in a GitHub repository. Put one `skills/<name>/SKILL.md` folder per skill in the repository and share the link. Members import the
+1 −1 lines
Upstream edited 1 Oct 2026 23:02 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
settings-reference
±changed
+11−0
claude-codechanged
settings-reference
The key also stops [mods](/docs/en/plugins/mods/overview), which are plugins whose code registers hooks:
* **In managed settings**: the mods in every installed plugin stop, your organization's included
* **In any other settings file**: the mods you installed stop, and [your organization's mods](/docs/en/plugins/mods/admin#install-your-organizations-mods) keep running
+11 −0 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
sandboxing
±changed
+1−0
claude-codechanged
sandboxing
* **Mods**: a [mod](/docs/en/plugins/mods/overview) is a plugin that runs its own code inside Claude Code, and a process that a mod starts runs outside the sandbox. See [What a mod
+1 −0 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/security
±changed
+1−1
claude-codechanged
plugins/security
* **Hooks and server processes**: command hooks execute shell commands with your full user permissions. Claude Code runs hooks and MCP servers outside the sandbox.
* **Hooks and server processes**: command hooks execute shell commands with your full user permissions. Claude Code runs hooks, MCP servers, and the processes a [mod](/docs/en/plug
+1 −1 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/mods/overview
±changed
+2−0
claude-codechanged
plugins/mods/overview
Mods aren't sandboxed. If you turn on [sandboxing](/docs/en/sandboxing), the sandbox isolates the Bash commands Claude runs, and a process that a mod starts runs outside it.
+2 −0 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/mods/events
±changed
+26−1
claude-codechanged
plugins/mods/events
#### Approve or refuse a tool call before the user is asked
To decide whether a tool call may run, handle [`tool.check`](/docs/en/plugins/mods/reference#tools), the event where Claude Code makes that decision. It fires after the permission
For a fixed command or path, use a [permission rule](/docs/en/permissions#permission-rule-syntax) such as `Bash(npm test)`, which takes no code. Handle `tool.check` when the decisi
+26 −1 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/mods/admin
±changed
+47−0
claude-codechanged
plugins/mods/admin
### Allow only your organization's mods
To run your organization's mods and block the ones users bring, deploy the settings from the **Only your organization's mods** row of the [policy table](#choose-how-much-to-allow),
```json managed-settings.json theme={null}
+47 −0 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/cli-reference
±changed
+17−1
claude-codechanged
plugins/cli-reference
### plugin test
Run the tests for a [mod](/docs/en/plugins/mods/overview), a plugin whose code registers event handlers. The command needs no session, sign-in, or network. For how to write a test,
```bash theme={null}
+17 −1 lines
Upstream edited 1 Oct 2026 22:58 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
sandbox-environments
±changed
+3−1
claude-codechanged
sandbox-environments
* `/tmp`, where Claude Code writes runtime files.
* The directory where Claude Code writes runtime files. Unless you set [`CLAUDE_CODE_TMPDIR`](/docs/en/env-vars), that directory is:
* **Linux and WSL2**: `/tmp`
* **macOS**: `/private/tmp`. `/tmp` is a symlink to that directory, and Seatbelt checks the resolved path.
+3 −1 lines
Upstream edited 1 Oct 2026 22:50 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/troubleshooting
±changed
+15−0
claude-codechanged
plugins/troubleshooting
<h3 id="an-npm-plugin-source-must-name-a-registry-package">
`An npm plugin source must name a registry package`
</h3>
+15 −0 lines
Upstream edited 1 Oct 2026 22:47 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/marketplace-reference
±changed
+11−3
claude-codechanged
plugins/marketplace-reference
| `npm` | `package`, `version`, `registry` | npm package, fetched with your npm client and unpacked without running install scripts |
| `npm` | `package`, `version`, `registry` | npm registry package or tarball link, fetched with your npm client and unpacked without running install scripts |
* `package`: a package name, or a scoped name such as `@your-org/formatter`
* `version`: a version or range
* `package`: a registry package name such as `@your-org/formatter`, a name with a version appended such as `@your-org/[email protected]`, or an `https` link to the package's tarball
+11 −3 lines
Upstream edited 1 Oct 2026 22:47 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
plugins/loading
±changed
+26−20
claude-codechanged
plugins/loading
Claude Code runs the install inside the copied version directory each time it creates one:
Claude Code installs the dependencies into the copied version directory each time it creates one:
The install runs only when the plugin's root directory contains both a `package.json` and a supported lockfile. The lockfile decides which command Claude Code runs:
| Lockfile | Command |
+26 −20 lines
Upstream edited 1 Oct 2026 22:47 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
errors
±changed
+9−1
claude-codechanged
errors
| `An npm plugin source must name a registry package` | [Plugin troubleshooting](/docs/en/plugins/troubleshooting#an-npm-plugin-source-must-name-a-registry-package) |
* `The response stream was malformed`: an event arrived for a content block that had already finished, or an event arrived damaged. A damaged event is one whose data isn't valid JS
* `The response stream was malformed`: an event arrived for a content block that had already finished, or an event arrived damaged. A damaged event is one whose data isn't valid JS
If you sign in with your claude.ai account in another Claude Code window that uses the same [configuration directory](/docs/en/claude-directory), an interactive session showing thi
+9 −1 lines
Upstream edited 1 Oct 2026 22:47 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
troubleshoot-install
±changed
+3−1
claude-codechanged
troubleshoot-install
Parallel sessions on one machine share a saved login and coordinate its renewal so that only one process refreshes the token at a time. Before v2.1.211, waking the machine from sle
Parallel sessions on one machine share a saved login and coordinate its renewal so that only one process refreshes the token at a time. For what the other sessions do after you sig
Before v2.1.211, waking the machine from sleep could cause two sessions to renew with the same token, which revoked the saved login and prompted every open session to log in again
+3 −1 lines
Upstream edited 1 Oct 2026 22:44 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
agent-sdk/streaming-vs-single-mode
±changed
+2−2
claude-codechanged
agent-sdk/streaming-vs-single-mode
When you run the example, the TypeScript version prints each response as it completes. The Python version's `receive_response()` loop ends at the first result message, so it prints
When you run the example, the TypeScript version prints each response as it completes. The Python version's `receive_response()` loop ends at the first result message, so it prints
If a query ends with an error result, such as `error_max_turns`, a single message `query()` call raises an error that includes the failure text after yielding the final result mess
If a query ends with an error result, such as `error_max_turns`, a single message `query()` call raises an error that includes the failure text after yielding the final result mess
+2 −2 lines
Upstream edited 1 Oct 2026 22:40 UTC Recorded here 23 Oct 2026 19:07 UTC
claude-code
agent-sdk/streaming-vs-single-mode
±changed
+2−2
claude-codechanged
agent-sdk/streaming-vs-single-mode
When you run the example, the TypeScript version prints each response as it completes. The Python version's `receive_response()` loop ends at the first result message, so it prints
When you run the example, the TypeScript version prints each response as it completes. The Python version's `receive_response()` loop ends at the first result message, so it prints
If a query ends with an error result, such as `error_max_turns`, a single message `query()` call raises an error that includes the failure text after yielding the final result mess
If a query ends with an error result, such as `error_max_turns`, a single message `query()` call raises an error that includes the failure text after yielding the final result mess
+2 −2 lines
Upstream edited 1 Oct 2026 22:40 UTC Recorded here 22 Oct 2026 18:07 UTC
claude-code
agent-sdk/streaming-vs-single-mode
±changed
+219−223
claude-codechanged
agent-sdk/streaming-vs-single-mode
<CodeGroup>
```typescript TypeScript theme={null}
import { query, type SDKUserMessage } from "@anthropic-ai/claude-agent-sdk";
import { readFile } from "fs/promises";
+219 −223 lines
Upstream edited 1 Oct 2026 22:40 UTC Recorded here 20 Oct 2026 18:37 UTC
claude-code
agent-sdk/streaming-vs-single-mode
±changed
+2−0
claude-codechanged
agent-sdk/streaming-vs-single-mode
If an image block's `source` is missing or isn't an object, the SDK doesn't report an error. Claude Code sends Claude a text note in place of the image, such as `[Image could not b
+2 −0 lines
Upstream edited 1 Oct 2026 22:40 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
llm-gateway-protocol
±changed
+1−1
claude-codechanged
llm-gateway-protocol
Deliver each response's full event sequence without dropping, duplicating, or reordering events. When an event references a content block whose `content_block_start` never arrived,
Deliver each response's full event sequence without dropping, duplicating, or reordering events. When an Amazon Bedrock guardrail blocks a reply, forward the events it sends unchan
+1 −1 lines
Upstream edited 1 Oct 2026 22:40 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
amazon-bedrock
±changed
+2−0
claude-codechanged
amazon-bedrock
When the guardrail blocks a response partway through, the text streamed so far stays and the reply ends with the message configured on the guardrail for blocked responses.
+2 −0 lines
Upstream edited 1 Oct 2026 22:40 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
accessibility
±changed
+9−5
claude-codechanged
accessibility
| [`CLAUDE_AX_PREPARK_MS`](/docs/en/env-vars#variables) | Environment variable | How long Claude Code waits, with the cursor at the start of the line, before it writes a new or cha
| [`CLAUDE_AX_PREPARK_MS`](/docs/en/env-vars#variables) | Environment variable | When you set it, how many milliseconds Claude Code holds the terminal cursor at the start of the cu
* Diffs read as plain text, line by line with `+` and `-` marking added and removed lines, so you can hear the proposed change in a file edit approval prompt before you answer it
Claude Code also waits at two points so your screen reader can keep up:
+9 −5 lines
Upstream edited 1 Oct 2026 22:37 UTC Recorded here 1 Oct 2026 23:07 UTC
claude-code
claude-tag
±changed
+3−3
claude-codechanged
claude-tag
No line of the body moved in this read.
+3 −3 lines
Recorded here 1 Oct 2026 22:37 UTC
claude-code
worktrees
±changed
+2−0
claude-codechanged
worktrees
A subagent in its own worktree takes the instruction files it [starts with](/docs/en/sub-agents#what-loads-at-startup) from your main conversation, not from its worktree. When that
+2 −0 lines
Upstream edited 1 Oct 2026 22:26 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
memory
±changed
+2−2
claude-codechanged
memory
Claude also discovers `CLAUDE.md` and `CLAUDE.local.md` files in subdirectories under your current working directory. Instead of loading them at launch, they are included when Clau
Claude also discovers `CLAUDE.md` and `CLAUDE.local.md` files in subdirectories under your current working directory. Instead of loading them at launch, they are included when Clau
Rules without a `paths` field are loaded unconditionally and apply to all files. Path-scoped rules trigger when Claude reads files matching the pattern, not on every tool use. As o
Rules without a `paths` field are loaded unconditionally and apply to all files. Path-scoped rules trigger when Claude reads files matching the pattern, not on every tool use. Matc
+2 −2 lines
Upstream edited 1 Oct 2026 22:26 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
remote-control
±changed
+1−1
claude-codechanged
remote-control
To bring back a session you started with `claude --remote-control` or `/remote-control`, resume the conversation with `claude --continue` or `claude --resume`. If Remote Control do
To bring back a session you started with `claude --remote-control` or `/remote-control`, resume the conversation with `claude --continue` or `claude --resume`. For the permission m
+1 −1 lines
Upstream edited 1 Oct 2026 22:25 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
hooks
±changed
+6−12
claude-codechanged
hooks
Type `/hooks` in Claude Code to open a read-only browser for your configured hooks. The menu shows every hook event with a count of configured hooks, lets you drill into matchers,
The menu displays all five hook types: `command`, `prompt`, `agent`, `http`, and `mcp_tool`. Each hook is labeled with a `[type]` prefix and a source indicating where it was define
* `User Settings`: from `~/.claude/settings.json`
+6 −12 lines
Upstream edited 1 Oct 2026 22:23 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
hooks-guide
±changed
+12−9
claude-codechanged
hooks-guide
Type `/hooks` to open the hooks browser. You'll see a list of all available hook events, with a count next to each event that has hooks configured. Select `Notification` to con
Type `/hooks` at the Claude Code prompt to open the hooks browser. Your new hook appears in the list under `Notification`.
<Tip>
The `/hooks` menu is read-only. To add, modify, or remove hooks, edit your settings JSON directly or ask Claude to make the change.
</Tip>
+12 −9 lines
Upstream edited 1 Oct 2026 22:23 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
plugins/mods/troubleshoot
±changed
+2−0
claude-codechanged
plugins/mods/troubleshoot
Some settings stop a mod and leave the rest of its plugin working. [Turn mods on or off](/docs/en/plugins/mods/overview#turn-mods-on-or-off) names them.
+2 −0 lines
Upstream edited 1 Oct 2026 22:17 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
plugins/mods/overview
±changed
+2−0
claude-codechanged
plugins/mods/overview
`disableAllHooks` and your organization's `allowManagedModsOnly` stop a mod and leave the rest of its plugin in place: the plugin stays installed, and its skills, commands, agents,
+2 −0 lines
Upstream edited 1 Oct 2026 22:17 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/network-requirements
±changed
+8−8
claude-docschanged
claude-tag/admins/network-requirements
The range is shared across Anthropic services, and dedicated per-organization egress addresses aren't available, so the allowlist entry admits Anthropic's infrastructure as a whole
The range is shared across Anthropic services, and dedicated per-organization egress addresses aren't available, so the allowlist entry admits Anthropic's infrastructure as a whole
> If your Google Cloud project uses VPC Service Controls, an IP-based access level doesn't admit Claude's queries through the [BigQuery connection](/docs/claude-tag/admins/connecti
> If your Google Cloud project uses VPC Service Controls, an IP-based access level doesn't admit Claude's queries through the [BigQuery connection](https://claude.com/docs/claude-t
In an Anthropic-hosted environment, every request from a channel's sandbox to your service passes through [Agent Proxy](/docs/claude-tag/concepts/agent-identity.md#agent-proxy), wh
+8 −8 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 23 Oct 2026 18:07 UTC
claude-docs
claude-tag/admins/network-requirements
±changed
+7−7
claude-docschanged
claude-tag/admins/network-requirements
The range is shared across Anthropic services, and dedicated per-organization egress addresses aren't available, so the allowlist entry admits Anthropic's infrastructure as a whole
The range is shared across Anthropic services, and dedicated per-organization egress addresses aren't available, so the allowlist entry admits Anthropic's infrastructure as a whole
> If your Google Cloud project uses VPC Service Controls, an IP-based access level doesn't admit Claude's queries through the [BigQuery connection](/docs/claude-tag/admins/connecti
> If your Google Cloud project uses VPC Service Controls, an IP-based access level doesn't admit Claude's queries through the [BigQuery connection](/docs/claude-tag/admins/connecti
In an Anthropic-hosted environment, every request from a channel's sandbox to your service passes through [Agent Proxy](/docs/claude-tag/concepts/agent-identity#agent-proxy), which
+7 −7 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 22 Oct 2026 18:07 UTC
claude-docs
claude-tag/admins/network-requirements
±changed
+9−5
claude-docschanged
claude-tag/admins/network-requirements
export const BetaNote = () => <Info>Claude Tag is in public beta. Features and behavior described here may change before general availability.</Info>;
<BetaNote />
> [!NOTE]
>
+9 −5 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 20 Oct 2026 18:07 UTC
claude-docs
claude-tag/overview
±changed
+1−1
claude-docschanged
claude-tag/overview
The list below covers common ways teams use Claude Tag. Each link opens a guide with the prompts to paste and the connections the task needs.
Each link opens a guide with the prompts to paste and the connections the task needs.
+1 −1 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/workspaces
±changed
+2−2
claude-docschanged
claude-tag/admins/workspaces
You can connect multiple Slack workspaces to one Claude organization. After the first pairing, the page no longer opens on setup, and the Slack row appears under **Where Claude Tag
You can connect multiple Slack workspaces to one Claude organization.
The reverse doesn't hold. A Slack workspace or Enterprise Grid pairs with one Claude organization at a time.
A Slack workspace or Enterprise Grid pairs with one Claude organization at a time.
+2 −2 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/troubleshooting
±changed
+5−10
claude-docschanged
claude-tag/admins/troubleshooting
This page covers errors you might hit setting up and administering Claude Tag: Slack app permissions, guest and shared channels, console errors, channels and threads where nothing
This page covers errors you might hit setting up and administering Claude Tag: Slack app permissions, guest and shared channels, console errors, channels and threads where nothing
Two phrases in the message are links:
* **reinstalls the Claude app** opens the reinstall flow. The fix below starts from this link.
+5 −10 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/setup-overview
±changed
+1−1
claude-docschanged
claude-tag/admins/setup-overview
To finish this step later, select **Skip** and confirm past the warning that Claude won't be able to act in the unconnected tools. Claude still works from what's in Slack: it can c
To finish this step later, select **Skip** and confirm past the warning that Claude won't be able to act in the unconnected tools. Claude still works from what's in Slack: it can c
+1 −1 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/network-requirements
±changed
+0−2
claude-docschanged
claude-tag/admins/network-requirements
Allowlist changes on enterprise systems can take days to take effect, which is why the [prerequisites for setup](/docs/claude-tag/admins/setup-overview) send you here before you st
+0 −2 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/audit
±changed
+0−2
claude-docschanged
claude-tag/admins/audit
[Verify your setup](/docs/claude-tag/admins/setup-overview#verify-your-setup) uses this check to validate a new connection.
+0 −2 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/add-connections
±changed
+5−5
claude-docschanged
claude-tag/admins/add-connections
<Tip>Claude starts delivering work before you connect anything. On Slack content alone, it can [catch a team up on a channel or thread](/docs/claude-tag/users/use-cases/catch-up),
<Tip>Claude starts delivering work before you connect anything. On Slack content alone, it can [catch a team up on a channel or thread](/docs/claude-tag/users/use-cases/catch-up),
Read-only connections are most useful in combination: an answer that joins the ticket, the deploy, and the error rate needs all three systems connected. Connecting many systems rea
Read-only connections are most useful in combination: an answer that joins the ticket, the deploy, and the error rate needs all three systems connected.
A connection grants access; a plugin teaches Claude how to use it well. A plugin is a packaged set of skills: reusable instructions for working with a specific tool or following a
+5 −5 lines
Upstream edited 1 Oct 2026 22:12 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-docs
claude-tag/admins/connections/overview
±changed
+3−1
claude-docschanged
claude-tag/admins/connections/overview
The presets and guides cover common services, not the full set Claude can connect to. Any app with an API can be added as a custom connection or a custom MCP server. See [Connect a
## Connect a service that has no guide
You can add any app that has an API as a custom connection or a custom MCP server, including services this page has no guide for. See [Connect a custom service](/docs/claude-tag/ad
+3 −1 lines
Upstream edited 1 Oct 2026 22:10 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
troubleshoot-install
±changed
+13−1
claude-codechanged
troubleshoot-install
| `EACCES: permission denied` during install | [Fix the install directory's permissions](#permission-errors-during-installation) |
The installer needs write access to `~/.local/bin/` and `~/.claude/` on macOS and Linux. On Windows the install location is under `%USERPROFILE%`, which is writable by your user by
An install that fails on permissions names the path it couldn't create or write. On Windows the install writes under `%USERPROFILE%`, which is writable by your user by default, so
On macOS and Linux the install writes to these locations:
+13 −1 lines
Upstream edited 1 Oct 2026 22:08 UTC Recorded here 1 Oct 2026 22:37 UTC
claude-code
claude-tag
±changed
+1−1
claude-codechanged
claude-tag
No line of the body moved in this read.
+1 −1 lines
Recorded here 1 Oct 2026 22:07 UTC
claude-docs
claude-tag/concepts/settings-map
±changed
+1−1
claude-docschanged
claude-tag/concepts/settings-map
On the Enterprise plan, an Owner or a [Claude Tag admin](/docs/claude-tag/admins/restrict-access#delegate-claude-tag-administration) can name [channel managers](/docs/claude-tag/ad
On the Enterprise plan, an Owner can name [channel managers](/docs/claude-tag/admins/restrict-access#delegate-channel-setup-to-channel-managers) for a channel, and so can a [Claude
+1 −1 lines
Upstream edited 1 Oct 2026 22:06 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-docs
claude-tag/concepts/glossary
±changed
+2−2
claude-docschanged
claude-tag/concepts/glossary
A member of your Claude organization whom an Owner or a [Claude Tag admin](#claude-tag-admin) has named to set up specific channels. For each channel assigned to them, a channel ma
A member of your Claude organization named to set up specific channels. For each channel assigned to them, a channel manager sets the default model, adds repositories their own Git
A member of your Claude organization whose custom role includes the **Claude Tag Admin** permission, available on the Enterprise plan. A Claude Tag admin manages Access bundles and
A member of your Claude organization whose custom role includes the **Claude Tag Admin** permission, available on the Enterprise plan. A Claude Tag admin manages Access bundles, at
+2 −2 lines
Upstream edited 1 Oct 2026 22:06 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-docs
claude-tag/admins/restrict-access
±changed
+8−6
claude-docschanged
claude-tag/admins/restrict-access
Each row of the **Auto-join channels** table is one pattern, added with **Add pattern**. A row can also carry [access bundles](/docs/claude-tag/admins/attach-to-scope#attach-a-bund
Each row of the **Auto-join channels** table is one pattern, added with **Add pattern**. A row can also carry [access bundles](/docs/claude-tag/admins/attach-to-scope#attach-a-bund
By default, workspace search covers public channels across the workspace, including ones Claude hasn't been added to. The **Channels Claude can search** setting narrows workspace s
By default, workspace search covers public channels across the workspace, including ones Claude hasn't been added to. The **Channels Claude can search** setting narrows workspace s
* Add and remove [channel managers](#delegate-channel-setup-to-channel-managers)
+8 −6 lines
Upstream edited 1 Oct 2026 22:06 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-docs
claude-tag/admins/healthcare
±changed
+1−1
claude-docschanged
claude-tag/admins/healthcare
* Search every public channel by keyword, including public channels it hasn't been added to. [No admin setting turns this search off](/docs/claude-tag/admins/restrict-access#contro
* Search every public channel by keyword, including public channels it hasn't been added to. [No admin setting turns this search off](/docs/claude-tag/admins/restrict-access#contro
+1 −1 lines
Upstream edited 1 Oct 2026 22:06 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-docs
claude-tag/admins/customize
±changed
+1−1
claude-docschanged
claude-tag/admins/customize
On the Enterprise plan, an Owner or a [Claude Tag admin](/docs/claude-tag/admins/restrict-access#delegate-claude-tag-administration) can name [channel managers](/docs/claude-tag/ad
On the Enterprise plan, an Owner can name [channel managers](/docs/claude-tag/admins/restrict-access#delegate-channel-setup-to-channel-managers) for a channel, and so can a [Claude
+1 −1 lines
Upstream edited 1 Oct 2026 22:06 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-docs
claude-tag/admins/attach-to-scope
±changed
+1−1
claude-docschanged
claude-tag/admins/attach-to-scope
A bundle attach rule binds a bundle to every channel whose name matches a pattern, instead of channel by channel. Rules live in the **Auto-join channels** table, the same table tha
A bundle attach rule binds a bundle to every channel whose name matches a pattern, instead of channel by channel. Rules are listed in the **Auto-join channels** table, the same tab
+1 −1 lines
Upstream edited 1 Oct 2026 22:06 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
troubleshooting
±changed
+5−0
claude-codechanged
troubleshooting
If the error comes back after a `/clear`, run [`/context`](/docs/en/debug-your-config) and compare the `Messages` row with the rows above it:
* **`Messages` is the largest row**: a file or tool output in the new conversation is refilling the window, so work through steps 1 to 3 again
* **The other rows together are larger**: what loads at session start leaves too little room to work in, so [trim what loads at startup](/docs/en/errors#prompt-is-too-long)
+5 −0 lines
Upstream edited 1 Oct 2026 22:04 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
self-hosted-environments-configuration
±changed
+15−0
claude-codechanged
self-hosted-environments-configuration
### Pass the system prompt flags through
The system prompt and appended system prompt that Anthropic's control plane sends for a session reach your wrapper as file paths, not as inline text. The runner writes each prompt
+15 −0 lines
Upstream edited 1 Oct 2026 22:01 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
memory
±changed
+3−1
claude-codechanged
memory
Auto memory is on by default. To toggle it, open `/memory` in a session and use the auto memory toggle, which saves `autoMemoryEnabled` to your user settings at `~/.claude/settings
Auto memory is on by default in local sessions. Outside [Claude Tag](https://claude.com/docs/claude-tag/overview) sessions, a session in a [self-hosted environment](/docs/en/self-h
To toggle it, open `/memory` in a session and use the auto memory toggle, which saves `autoMemoryEnabled` to your user settings at `~/.claude/settings.json`. To turn it off for a s
+3 −1 lines
Upstream edited 1 Oct 2026 22:01 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
errors
±changed
+17−0
claude-codechanged
errors
| `Error: Cannot use both --append-subagent-system-prompt and --append-subagent-system-prompt-file. Please use only one.` | [Command-line errors](#conflict-between-a-system-prompt-
<h3 id="conflict-between-a-system-prompt-flag-and-its-file-form">
Conflict between a system prompt flag and its file form
</h3>
+17 −0 lines
Upstream edited 1 Oct 2026 21:58 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
settings-reference
±changed
+3−2
claude-codechanged
settings-reference
For example, `cd build && docker compose up` stays sandboxed under a `docker *` entry, and adding a `cd` entry doesn't change that.
* A `git clone`, `git init`, `git worktree add`, `git worktree move`, or `git bundle create` with a path argument that is absolute, starts with `~`, or contains a `..` segment
For example, `cd build && docker compose up` stays sandboxed under a `docker *` entry, and adding a `cd` entry doesn't change that. Under a `git *` entry, `git clone <url> vendor/l
+3 −2 lines
Upstream edited 1 Oct 2026 21:55 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
agent-sdk/typescript
±changed
+1−1
claude-codechanged
agent-sdk/typescript
For the `Agent` tool, `tool_use_result` is [`AgentOutput`](#agent-2). On a `completed` result, `content` holds the subagent's report without the agent ID and usage trailer that Cla
For the `Agent` tool, `tool_use_result` is [`AgentOutput`](#agent-2). Render from it rather than parsing the `tool_result` text. A `completed` result's `content` holds the subagent
+1 −1 lines
Upstream edited 1 Oct 2026 21:55 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
agent-sdk/subagents
±changed
+1−1
claude-codechanged
agent-sdk/subagents
The parent receives the subagent's final message as the Agent tool result, but may summarize it in its own response. To preserve subagent output verbatim in the user-facing respo
The parent receives the subagent's final report, but may summarize it in its own response. To preserve subagent output verbatim in the user-facing response, include an instructio
+1 −1 lines
Upstream edited 1 Oct 2026 21:55 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
agent-sdk/agent-loop
±changed
+1−1
claude-codechanged
agent-sdk/agent-loop
* **Use subagents for subtasks.** Each subagent starts with a fresh conversation (no prior message history, though it does load its own system prompt and project-level context like
* **Use subagents for subtasks.** Each subagent starts with a fresh conversation (no prior message history, though it does load its own system prompt and project-level context like
+1 −1 lines
Upstream edited 1 Oct 2026 21:55 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
debug-your-config
±changed
+4−2
claude-codechanged
debug-your-config
Run `/hooks` to list every hook registered for the current session, grouped by event. If a hook you defined doesn't appear, it isn't being read: hooks go under the `"hooks"` key in
Run `/hooks` to list every hook registered for the current session, grouped by event. If a hook you defined doesn't appear, Claude Code didn't load it. Check for these causes:
* The hook is defined in a standalone file. Hooks go under the `"hooks"` key in a [settings file](/docs/en/settings#settings-files).
* The `matcher` value is an array instead of a single string. Claude Code lists the entry as an invalid setting when you start an interactive session and in `claude doctor`. If the
+4 −2 lines
Upstream edited 1 Oct 2026 21:54 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
plugins/mods/troubleshoot
±changed
+9−0
claude-codechanged
plugins/mods/troubleshoot
### `no command.run hook answered it`
You run a command your mod added, and the reply names the mod and the command, as in `first-mod registered /tally but no command.run hook answered it`, then tells you to add a hook
* **No hook answered the command**: the module has no `command.run` hook, the hook's [filter](/docs/en/plugins/mods/events#filter-which-events-a-hook-handles) names a different com
+9 −0 lines
Upstream edited 1 Oct 2026 21:47 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
plugins/mods/interface
±changed
+8−1
claude-codechanged
plugins/mods/interface
| `closeOnEscape` | Makes Esc close the pane. Pass `true` or leave the field out, because Claude Code refuses `false`. |
| `closeOnEscape` | Makes Esc close the pane |
`focus`, `closeOnEscape`, and `holdToasts` are optional and accept only `true`. To leave one off, omit it. Passing `false` throws an error such as `ui.open: focus is true or left o
+8 −1 lines
Upstream edited 1 Oct 2026 21:47 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
costs
±changed
+1−1
claude-codechanged
costs
* **Add custom compaction instructions**: `/compact Focus on code samples and API usage` tells Claude what to preserve during summarization. In a fresh session, `/compact` prints `
* **Add custom compaction instructions**: `/compact Focus on code samples and API usage` tells Claude what to preserve during summarization.
+1 −1 lines
Upstream edited 1 Oct 2026 21:40 UTC Recorded here 1 Oct 2026 22:07 UTC
claude-code
agent-sdk/skills
±changed
+1−1
claude-codechanged
agent-sdk/skills
A `compact_boundary` message only arrives when compaction ran. With nothing to summarize, `/compact` reports the reason instead of raising. The run still ends with a `success` re
A `compact_boundary` message only arrives when compaction ran. When a continued session has messages but nothing `/compact` can summarize, the run still ends with a `success` res
+1 −1 lines
Upstream edited 1 Oct 2026 21:40 UTC Recorded here 1 Oct 2026 22:07 UTC
api
test-and-evaluate/strengthen-guardrails/reduce-hallucinations
±changed
+1−1
apichanged
test-and-evaluate/strengthen-guardrails/reduce-hallucinations
* **Chain-of-thought verification**: Ask Claude to explain its reasoning step-by-step before giving a final answer. This can reveal faulty logic or assumptions.
* **Chain-of-thought verification**: Use [thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) with `display: "summarized"`, and review the summarized reasonin
+1 −1 lines
Recorded here 1 Oct 2026 21:37 UTC
api
test-and-evaluate/develop-tests
±changed
+8−8
apichanged
test-and-evaluate/develop-tests
* **Encourage reasoning:** Ask the LLM to reason first before producing an evaluation score, and then discard the reasoning. This increases evaluation performance, particularly for
* **Encourage reasoning:** Use a grader model with [thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) on, so that it reasons before it produces an evaluatio
Think through your reasoning in <thinking> tags, then output 'correct' or 'incorrect' in <result> tags."""
Output 'correct' or 'incorrect' in <result> tags."""
Think through your reasoning in <thinking> tags, then output 'correct' or 'incorrect' in <result> tags.`;
+8 −8 lines
Recorded here 1 Oct 2026 21:37 UTC
api
release-notes/overview
±changed
+1−2
apichanged
release-notes/overview
- The [Admin API](https://platform.claude.com/docs/en/manage-claude/admin-api) is now available in the `ant` CLI and the Python, TypeScript, C#, Go, Java, PHP, and Ruby SDKs under
* The [Admin API](https://platform.claude.com/docs/en/manage-claude/admin-api) is now available in the `ant` CLI and the Python, TypeScript, C#, Go, Java, PHP, and Ruby SDKs under
+1 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
models/sonnet-5-5/whats-new-sonnet-5-5
±changed
+2−2
apichanged
models/sonnet-5-5/whats-new-sonnet-5-5
Every thinking block records which model produced it. Each model reads its own blocks and only some other models' blocks. Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet
Every thinking block records which model produced it. Each model reads its own blocks and only some other models' blocks. Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet
So a conversation that moves from Claude Sonnet 5 onto Claude Sonnet 5.5, or from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API, keeps its reasoning, and any other move
So a conversation that moves from Claude Sonnet 5 onto Claude Sonnet 5.5, or from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API and Google Cloud, keeps its reasoning, a
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
models/opus-5-5/whats-new-opus-5-5
±changed
+2−2
apichanged
models/opus-5-5/whats-new-opus-5-5
Every thinking block records which model produced it, and each model reads its own blocks and only some other models'. Claude Opus 5.5 reads thinking blocks from Claude Opus 5, fro
Every thinking block records which model produced it, and each model reads its own blocks and only some other models'. Claude Opus 5.5 reads thinking blocks from Claude Opus 5, fro
* **More safeguard categories.** The model runs a biology safety classifier in addition to the cybersecurity one, and requests that push it to reproduce its internal reasoning in t
* **More safeguard categories.** The model runs a biology safety classifier in addition to the cybersecurity one. As on Claude Opus 5, requests that push it to reproduce its intern
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
models/opus-5-5/migration-guide
±changed
+3−3
apichanged
models/opus-5-5/migration-guide
* If a router or fallback can move a conversation from Claude Opus 5.5 to another model, expect that model to run without Claude Opus 5.5's thinking blocks (Claude Fable 5.1 and Cl
* If a router or fallback can move a conversation from Claude Opus 5.5 to another model, expect that model to run without Claude Opus 5.5's thinking blocks (Claude Fable 5.1 and Cl
On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read Claude Opus 5.5 thinking blocks; no other model does. A router or fallback that moves a conversation from Claude Opus
On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read Claude Opus 5.5 thinking blocks; no other model does. A router or fallback that moves a conversation from Claude Opus
Claude Opus 5.5 can return `stop_reason: "refusal"` with a `stop_details` category. Its classifiers cover a broader set of categories than Claude Opus 5's, so expect `stop_details.
+3 −3 lines
Recorded here 1 Oct 2026 21:37 UTC
api
models/fable-5-1/migration-guide
±changed
+1−1
apichanged
models/fable-5-1/migration-guide
5. **Safety classifiers and fallback routing:** Claude Fable 5.1 runs safety classifiers covering the same `stop_details` categories as Claude Fable 5, a broader set than Claude Op
5. **Safety classifiers and fallback routing:** Claude Fable 5.1 runs safety classifiers covering the same `stop_details` categories as Claude Fable 5, a broader set than Claude Op
+1 −1 lines
Recorded here 1 Oct 2026 21:37 UTC
api
managed-agents/self-hosted-sandboxes
±changed
+2−2
apichanged
managed-agents/self-hosted-sandboxes
VERSION=1.36.0
VERSION=1.37.0
ARG ANT_VERSION=1.36.0
ARG ANT_VERSION=1.37.0
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
managed-agents/quickstart
±changed
+2−2
apichanged
managed-agents/quickstart
VERSION=1.36.0
VERSION=1.37.0
implementation("com.anthropic:anthropic-java:2.66.0")
implementation("com.anthropic:anthropic-java:2.67.0")
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
managed-agents/events-and-streaming
±changed
+3−3
apichanged
managed-agents/events-and-streaming
.addType("agent.tool_use")
.addType("agent.tool_result")
.addType(BetaManagedAgentsSessionEventType.AGENT_TOOL_USE)
.addType(BetaManagedAgentsSessionEventType.AGENT_TOOL_RESULT)
* **Use system prompts:** Add logging instructions to the system prompt to make the agent explain its reasoning
+3 −3 lines
Recorded here 1 Oct 2026 21:37 UTC
api
managed-agents/environments
±changed
+76−25
apichanged
managed-agents/environments
"networking": {"type": "unrestricted"}
"networking": {"type": "limited", "allow_package_managers": true}
type: unrestricted
type: limited
allow_package_managers: true
+76 −25 lines
Recorded here 1 Oct 2026 21:37 UTC
api
manage-claude/user-management
±changed
+2−2
apichanged
manage-claude/user-management
For complete parameter details and response schemas, see [Add group member](https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/create) in the API referen
For complete parameter details and response schemas, see [Add group member](https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/add) in the API reference.
For complete parameter details and response schemas, see [Remove group member](https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/delete) in the API refe
For complete parameter details and response schemas, see [Remove group member](https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/remove) in the API refe
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
manage-claude/spend-limits-api
±changed
+2−2
apichanged
manage-claude/spend-limits-api
For complete parameter details and response schemas, see [List effective spend limits](https://platform.claude.com/docs/en/api/beta/organization/spend_limits/list_effective) in the
For complete parameter details and response schemas, see [List effective spend limits](https://platform.claude.com/docs/en/api/beta/organization/spend_limits/effective/list) in the
For complete parameter details and response schemas, see [Create a spend limit](https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create) in the API reference.
For complete parameter details and response schemas, see [Create a spend limit](https://platform.claude.com/docs/en/api/beta/organization/spend_limits/set) in the API reference.
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
get-started
±changed
+2−2
apichanged
get-started
implementation("com.anthropic:anthropic-java:2.66.0")
implementation("com.anthropic:anthropic-java:2.67.0")
<version>2.66.0</version>
<version>2.67.0</version>
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
cli-sdks-libraries/sdks/java
±changed
+2−2
apichanged
cli-sdks-libraries/sdks/java
implementation("com.anthropic:anthropic-java:2.66.0")
implementation("com.anthropic:anthropic-java:2.67.0")
<version>2.66.0</version>
<version>2.67.0</version>
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
cli-sdks-libraries/libraries/openai-sdk
±changed
+1−1
apichanged
cli-sdks-libraries/libraries/openai-sdk
You can enable [thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) by adding the `thinking` parameter. On current models thinking is adaptive, with Claude de
You can enable [thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) by adding the `thinking` parameter. On current models thinking is adaptive, with Claude de
+1 −1 lines
Recorded here 1 Oct 2026 21:37 UTC
api
cli-sdks-libraries/cli/quickstart
±changed
+1−1
apichanged
cli-sdks-libraries/cli/quickstart
VERSION=1.36.0
VERSION=1.37.0
+1 −1 lines
Recorded here 1 Oct 2026 21:37 UTC
api
claude_api_primer
±changed
+13−21
apichanged
claude_api_primer
### Chain of thought
When using tools, Claude often shows its "chain of thought," that is, the step-by-step reasoning it uses to break down the problem and determine which tools to use.
### Parallel tool use
+13 −21 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/thinking
±changed
+2−2
apichanged
build-with-claude/thinking
To see the model's reasoning, read the `thinking` blocks rather than prompting for reasoning in the response text. On Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, and Clau
To see the model's reasoning, read the `thinking` blocks rather than prompting for reasoning in the response text. On Claude Fable 5.1, Claude Opus 5.5, Claude Opus 5, Claude Sonne
**Switching models mid-conversation.** Keep passing thinking blocks back unchanged when you switch models, for example after a [classifier refusal fallback](https://platform.claude
**Switching models mid-conversation.** Keep passing thinking blocks back unchanged when you switch models, for example after a [classifier refusal fallback](https://platform.claude
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/structured-outputs
±changed
+2−0
apichanged
build-with-claude/structured-outputs
A property that asks for the model's thinking or step-by-step reasoning may lead to a `reasoning_extraction` refusal. Ask for a short explanation instead. See [Keep reasoning in th
+2 −0 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/refusals-and-fallback
±changed
+22−9
apichanged
build-with-claude/refusals-and-fallback
| `category` | What it means
| ------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------
| `"cyber"` | The request could enable cyber harm, such as malware or exploit development. Benign cybersecurity work can also trigger this category.
| `"bio"` | The request could enable biological harm, such as dangerous lab methods. Beneficial life sciences work can also trigger this category.
| `"frontier_llm"` | The request could assist the development of competing AI models, which is restricted under [Anthropic's commercial terms](https://www.anthropic.com/leg
+22 −9 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/prompt-engineering/prompting-claude-sonnet-5-5
±changed
+1−1
apichanged
build-with-claude/prompt-engineering/prompting-claude-sonnet-5-5
If your prompts ask the model to include its reasoning in the response, remove those instructions, because they invite `reasoning_extraction` declines. With adaptive thinking, read
If your prompts ask the model to include its reasoning in the response, remove those instructions, because they invite `reasoning_extraction` declines. With adaptive thinking, read
+1 −1 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/prompt-engineering/prompting-claude-opus-5-5
±changed
+2−2
apichanged
build-with-claude/prompt-engineering/prompting-claude-opus-5-5
* **Remove instructions that stood in for thinking.** If your prompt asked the model to write out its reasoning in the response as a substitute for thinking, remove that instructio
* **Remove instructions that stood in for thinking.** If your prompt asked the model to write out its reasoning in the response as a substitute for thinking, remove that instructio
* **Reasoning extraction:** Requests that push the model to reproduce its internal reasoning in the response text can be declined with the `reasoning_extraction` category, which is
* **Reasoning extraction:** Requests that push the model to reproduce its internal reasoning in the response text may be declined with the `reasoning_extraction` category. If your
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/prompt-engineering/prompting-claude-opus-5
±changed
+4−0
apichanged
build-with-claude/prompt-engineering/prompting-claude-opus-5
## Reasoning in the response
Prompts, skills, and tool descriptions that ask Claude Opus 5 to write out its thinking or reasoning, verbatim or in a fixed format, may be declined with the `reasoning_extraction`
+4 −0 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/prompt-engineering/prompting-claude-fable-5-1
±changed
+2−0
apichanged
build-with-claude/prompt-engineering/prompting-claude-fable-5-1
Prompts, skills, and tool descriptions that ask the model to write out its thinking or reasoning may be declined with the `reasoning_extraction` category. Ask for a short explana
+2 −0 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/prompt-engineering/prompting-claude-fable-5
±changed
+2−2
apichanged
build-with-claude/prompt-engineering/prompting-claude-fable-5
Terse shorthand is fine between tool calls (that's you thinking out loud, and brevity there is good). Your final summary is different: it's for a reader who didn't see any of that.
Terse shorthand is fine between tool calls. Your final summary is different: it's for a reader who didn't see any of that.
* **Don't instruct Claude to reproduce its reasoning in the response.** Prompts, skills, or harness instructions that tell the model to echo, transcribe, or explain its internal re
* **Don't instruct Claude to reproduce its reasoning in the response.** Prompts, skills, or harness instructions that tell the model to echo, transcribe, or explain its internal re
+2 −2 lines
Recorded here 1 Oct 2026 21:37 UTC
api
build-with-claude/prompt-engineering/overview
±changed
+1−1
apichanged
build-with-claude/prompt-engineering/overview
An example-filled tutorial that covers the prompt engineering concepts found in the docs.
An example-filled tutorial on prompt engineering fundamentals. It was written for Claude 3 models, so follow Prompting best practices where the two differ.
+1 −1 lines
Recorded here 1 Oct 2026 21:37 UTC
Nothing on this page matches that. Press Enter to search every page of the ledger.