docs: Claude Code v2.1.306 - project cloud threads stop loading repo skills, settings, hooks and .mcp.json

content/claude/cowork/guide/dispatch.md

@@ -64,7 +64,7 @@ Each child task shows its current state in the sidebar. Select any task to open
 
 When a child task needs permission to take an action (such as running a command or writing a file outside its workspace), the prompt is forwarded to you. If you don't respond within ten minutes, the request is automatically denied and the task continues without that action.
 
-Permission prompts behave the same as in a normal Cowork session.
+Permission prompts behave the same as in a normal Cowork session. A tool that is marked as needing a person's approval can ask on every call, even if you chose to always allow it. If nobody can be asked, the call can be refused.
 
 ## Continue from a finished task
 

content/en/docs/claude-code/agent-sdk/mcp.md

@@ -161,8 +161,12 @@ Servers from [settings files](#from-a-config-file) such as `.mcp.json`, or from
 * **Otherwise, with [tool search](/docs/en/agent-sdk/tool-search) on, which is the default**: the first turn waits up to 2 seconds, and only for pending servers that set [`alwaysLoad: true`](/docs/en/mcp#exempt-a-server-from-deferral) or that you [named](#named-servers). The rest keep connecting in the background, and Claude [reaches their tools](/docs/en/mcp#tool-availability) once they connect.
 * **Otherwise, with tool search off**: the first turn waits up to 2 seconds for every pending settings-file and plugin server. Tool search is off when your [`ENABLE_TOOL_SEARCH` value or model](/docs/en/agent-sdk/tool-search#configure-tool-search) turns it off, or when you exclude the `ToolSearch` tool, for example through `disallowedTools`.
 
+### Named servers
+
 You name a server by listing any of its tools, such as `mcp__github__list_issues` or `mcp__github__*`, in [`allowedTools`](#allow-mcp-tools) or in a `permissions.allow` rule in the `settings` option. An [`mcp_tool` hook](/docs/en/hooks#mcp-tool-hook-fields) also names the server it calls when you configure it on a prompt-submission, tool-use, permission, subagent, task, elicitation, or worktree event. Before v2.1.284, naming a server had no effect on the first-turn wait.
 
+### Set the first-turn wait yourself
+
 To choose the first-turn wait yourself, set `CLAUDE_CODE_MCP_STARTUP_WAIT_MS` in the [`env` option](/docs/en/agent-sdk/configuration#set-environment-variables), for example `CLAUDE_CODE_MCP_STARTUP_WAIT_MS: "5000"`. The first turn then waits up to that many milliseconds for pending settings-file and plugin servers, whether or not tool search is on. The same limit replaces the `MCP_TIMEOUT` wait for the stdio, HTTP, and SSE servers in `options.mcpServers`.
 
 Three waits apply whatever value you set, including `0`:
@@ -173,6 +177,8 @@ Three waits apply whatever value you set, including `0`:
 
 Set the variable to `0` to skip the wait for every other server. Servers still pending when the wait ends keep connecting in the background. `CLAUDE_CODE_MCP_STARTUP_WAIT_MS` requires Claude Code v2.1.274 or later.
 
+### Make startup wait for servers
+
 Two settings make startup itself wait, before Claude Code sends the init message:
 
 * **[`MCP_CONNECTION_NONBLOCKING`](/docs/en/env-vars) set to `0`**: startup waits for the whole connection batch, up to 5 seconds by default. Set [`MCP_CONNECT_TIMEOUT_MS`](/docs/en/env-vars), in milliseconds, to change that cap. Servers still pending at the cap keep connecting in the background.

content/en/docs/claude-code/agent-sdk/tool-search.md

@@ -46,7 +46,7 @@ You control whether and when tool definitions are deferred with the `ENABLE_TOOL
 | :- | :- |
 | (unset) | Tool search is on, with tool definitions deferred and discovered on demand. Falls back to loading upfront on models earlier than Opus 4 and Sonnet 4, on the [earlier Agent Platform models](#configure-tool-search), and on Microsoft Foundry deployments hosted on Azure. |
 | `true` | Tool search is on, with the same fallbacks as (unset). |
-| `auto` | Counts the tokens in the tool definitions that tool search can defer and compares the total against the model's context window. When the total reaches 10% of the window, tool search activates. Below that, the SDK loads every tool definition into context upfront. |
+| `auto` | Tool search is off, and tool definitions load upfront, until the definitions it can defer reach 10% of the model's context window. Tool search then activates, and can [stay on in the same conversation](/docs/en/mcp#tools-that-stay-deferred-below-the-threshold) even if those definitions later drop below 10%. |
 | `auto:N` | Same as `auto` with a custom percentage. `auto:5` activates when those definitions reach 5% of the context window. Lower values activate sooner. |
 | `false` | Tool search is off. All tool definitions are loaded into context on every turn. |
 

content/en/docs/claude-code/claude-code-on-the-web.md

@@ -264,7 +264,7 @@ Cloud sessions support [built-in commands](/docs/en/commands) that produce text
 
 * **`/model`, `/effort`, `/color`, and `/rename`**: pass the value as an argument, for example `/model sonnet`, instead of opening the terminal picker or slider. The argument forms follow each command's [availability notes](/docs/en/commands#all-commands).
 * **`/fast`**: toggles [fast mode](/docs/en/fast-mode#use-fast-mode-in-cloud-sessions) for the session when fast mode is [available on your account](/docs/en/fast-mode#requirements). Requires Claude Code v2.1.271 or later in the session's environment.
-* **`/config`**: in your browser at claude.ai/code, opens the Claude Code section of your settings instead of setting a value, and text after the command, including `key=value`, is ignored. To change a setting for a cloud session, set an [environment variable](/docs/en/cloud-environments#set-environment-variables) on the environment, or in a session with one repository, commit the key to that repository's `.claude/settings.json`. [Settings in cloud sessions](/docs/en/settings#settings-in-cloud-sessions) lists what each session reads.
+* **`/config`**: in your browser at claude.ai/code, opens the Claude Code section of your settings instead of setting a value, and text after the command, including `key=value`, is ignored. To change a setting for a cloud session, set an [environment variable](/docs/en/cloud-environments#set-environment-variables) on the environment. In a session with one repository outside a [project](/docs/en/claude-projects), commit the key to that repository's `.claude/settings.json` instead. [Settings in cloud sessions](/docs/en/settings#settings-in-cloud-sessions) lists what each session reads.
 
 For context management specifically:
 
@@ -278,7 +278,7 @@ Auto-compaction runs automatically when the context window approaches capacity.
 
 To change the auto-compact window instead, set [`CLAUDE_CODE_AUTO_COMPACT_WINDOW`](/docs/en/env-vars) in your environment variables, or run [`/autocompact`](/docs/en/commands#all-commands) with a token count in a session where the variable isn't set.
 
-[Subagents](/docs/en/sub-agents) work the same way they do locally. Claude can spawn them with the Agent tool to offload research or parallel work into a separate context window, keeping the main conversation lighter. Subagents defined in your repo's `.claude/agents/` are picked up automatically.
+[Subagents](/docs/en/sub-agents) work the same way they do locally. Claude can spawn them with the Agent tool to offload research or parallel work into a separate context window, keeping the main conversation lighter. Outside a [project](/docs/en/claude-projects#what-threads-pick-up-from-your-repositories)'s threads, subagents defined in your repo's `.claude/agents/` load automatically.
 
 [Agent teams](/docs/en/agent-teams) are off by default but can be enabled by adding `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` to your [environment variables](/docs/en/cloud-environments#set-environment-variables).
 

content/en/docs/claude-code/claude-projects.md

@@ -52,7 +52,7 @@ A project is one coordinating conversation with Claude plus the threads it start
 * **Threads**: the workers. Each is a separate session with its own context window that does one piece of work and reports back to the conversation when it finishes. A cloud thread works on its own branch and opens a pull request when the work calls for one.
 * **What every cloud thread starts with**:
   * The project's repositories and files, plus its [instructions and memory](#give-a-project-standing-context)
-  * The `CLAUDE.md` and skills in [each of the project's repositories](#what-threads-pick-up-from-your-repositories), and in a project with one repository, that repository's permission rules and hooks too
+  * The `CLAUDE.md` in [each of the project's repositories](#what-threads-pick-up-from-your-repositories), but not the repositories' skills, `.claude/settings.json`, or `.mcp.json`
   * The [connectors](#get-skills-plugins-connectors-and-tools-into-threads) on your claude.ai account
   * A [cloud environment](#choose-an-environment-for-threads) that sets its network access, environment variables, network secrets, and installed tools
 * **The Overview pane**: where you [see all the threads at once](#see-what-needs-you-in-overview) and which of them need you. Its other tabs are **Library** for the files you added and the files threads produced, **Pull requests** for the ones threads opened, and **Scheduled tasks** for scheduled work in the project.
@@ -239,13 +239,7 @@ Claude saves preferences like these to [project memory](#give-a-project-standing
 
 Threads run in [auto mode](/docs/en/permission-modes#eliminate-prompts-with-auto-mode) when the thread's model supports it, so most tool calls run without asking you. When a thread needs your approval, the prompt is inside that thread and the thread waits until you answer it there. Telling Claude in the project conversation to go ahead doesn't reach it.
 
-Each approval covers that prompt, or the rest of that thread if you choose the broader option.
-
-To let every thread run certain commands without asking, or to block some, add [permission rules](/docs/en/permissions) to the repository's `.claude/settings.json`. Check that cloud threads in your project apply them:
-
-* **One repository**: cloud threads apply the rules. See [What threads pick up from your repositories](#what-threads-pick-up-from-your-repositories).
-* **Several repositories, Anthropic-hosted environment**: no repository's permission rules reach a cloud thread, so you rely on auto mode and on the approvals you give inside each thread.
-* **Several repositories, self-hosted environment**: see [which repository's settings apply](/docs/en/self-hosted-environments-configuration#repository-settings-in-sessions-with-several-repositories).
+Each approval covers that prompt, or the rest of that thread if you choose the broader option. Cloud threads don't read [permission rules](/docs/en/permissions) from a repository's `.claude/settings.json`, so you rely on auto mode and on the approvals you give inside each thread.
 
 ### Run a thread on your own computer
 
@@ -313,10 +307,10 @@ Rules about one repository, such as its build commands, belong in that repositor
 
 ### Decide which repositories to add
 
-The repositories you add to a project come with everything in them, their code, `CLAUDE.md`, and skills, in every cloud thread. Repositories you don't add are still within reach: a cloud thread can add one to itself when its task needs it. Most projects use both:
+The repositories you add to a project come with their code and `CLAUDE.md` in every cloud thread, but [not their skills, `.claude/settings.json`, or `.mcp.json`](#what-threads-pick-up-from-your-repositories). Repositories you don't add are still within reach: a cloud thread can add one to itself when its task needs it. Most projects use both:
 
-* **Add it to the project**, in the **New project** dialog, in **Project settings > Environment**, or by asking Claude in the conversation to add it to the project. Every cloud thread from then on clones it and starts with its `CLAUDE.md` and skills loaded, whether or not the task touches it. Going from one repository to several also changes what threads take from each repository's `.claude/settings.json`; see [What threads pick up from your repositories](#what-threads-pick-up-from-your-repositories).
-* **Leave it off and let threads add it when needed.** A cloud thread whose task needs a repository the project doesn't have can add it to itself, and a note in the thread says it was added to this thread only. The clone happens partway through the task, so that repository's `CLAUDE.md` and skills weren't there when the thread started. The next thread starts without it. A repository added this way needs the same [prerequisites](#check-the-prerequisites) as a project repository: the Claude GitHub App installed on it and push access from your GitHub account.
+* **Add it to the project**, in the **New project** dialog, in **Project settings > Environment**, or by asking Claude in the conversation to add it to the project. Every cloud thread from then on clones it and starts with its `CLAUDE.md` loaded, whether or not the task touches it.
+* **Leave it off and let threads add it when needed.** A cloud thread whose task needs a repository the project doesn't have can add it to itself, and a note in the thread says it was added to this thread only. The clone happens partway through the task, so that repository's `CLAUDE.md` wasn't there when the thread started. The next thread starts without it. A repository added this way needs the same [prerequisites](#check-the-prerequisites) as a project repository: the Claude GitHub App installed on it and push access from your GitHub account.
 
 A project doesn't need a repository at all. Its cloud threads can still research, write documents, and write and run code in their own sandbox, and they deliver files to the **Library** tab. Any of its cloud threads can still add a repository to itself when a task calls for one.
 
@@ -335,16 +329,18 @@ Add the files and folders you want threads to read in the **New project** dialog
 
 ### What threads pick up from your repositories
 
-Each cloud thread clones every repository in the project and loads `CLAUDE.md` and skills from all of them. Permission rules, hooks, and `env` come only from the `.claude/settings.json` in the directory the thread starts in.
+Cloud threads load each repository's `CLAUDE.md` instructions but not its skills, `.mcp.json` servers, or `.claude/settings.json`, whether the project has one repository or several.
 
-| In each repository | One repository | Several repositories |
-| :- | :- | :- |
-| `CLAUDE.md` | Loaded when the thread starts | Loaded from every repository when the thread starts |
-| Skills, agents, and commands under `.claude/` | Loaded | Loaded from every repository |
-| Plugins enabled in `.claude/settings.json` | Not loaded. Add the plugin in **Project settings > Plugins** instead | Not loaded. Add the plugin in **Project settings > Plugins** instead |
-| Permission rules, hooks, and `env` defined in `.claude/settings.json` | Apply to the thread, except the `env` keys that [no cloud session honors](/docs/en/cloud-environments#what-carries-over-from-your-setup) | Don't apply in an Anthropic-hosted environment. For a self-hosted environment, see [which repository's settings apply](/docs/en/self-hosted-environments-configuration#repository-settings-in-sessions-with-several-repositories) |
-
-In a project with several repositories, put standing rules in project instructions and give threads environment variables through the [cloud environment](#choose-an-environment-for-threads).
+| In each repository | In a cloud thread |
+| :- | :- |
+| `CLAUDE.md`, `.claude/CLAUDE.md`, and `.claude/rules/` at the repository root | Loaded from every repository when the thread starts. `CLAUDE.md` files in subdirectories and [path-specific rules](/docs/en/memory#path-specific-rules) aren't loaded |
+| Skills under `.claude/skills/` | Not loaded. Enable the skill for your claude.ai account instead, as [Get skills, plugins, connectors, and tools into threads](#get-skills-plugins-connectors-and-tools-into-threads) describes |
+| Agents and commands under `.claude/` | Not loaded. To use them, [run the thread on your own computer](#run-a-thread-on-your-own-computer) |
+| MCP servers in `.mcp.json` | Not loaded. Use the connectors on your claude.ai account instead, as [Get skills, plugins, connectors, and tools into threads](#get-skills-plugins-connectors-and-tools-into-threads) describes |
+| Plugins enabled in `.claude/settings.json` | Not loaded. Add the plugin in **Project settings > Plugins** instead |
+| Permission rules in `.claude/settings.json` | Not applied. Approve tool calls inside the thread, as [Unblock a thread waiting on approval](#unblock-a-thread-waiting-on-approval) describes, or [run the thread on your own computer](#run-a-thread-on-your-own-computer) |
+| Hooks in `.claude/settings.json` | Not run. To use your hooks, [run the thread on your own computer](#run-a-thread-on-your-own-computer) |
+| `env` in `.claude/settings.json` | Not applied. Set environment variables in the [cloud environment](#choose-an-environment-for-threads) |
 
 ### Choose an environment for threads
 
@@ -356,9 +352,9 @@ If cloud threads need to reach an internal API or a private package registry, or
 
 Cloud threads don't have the skills, MCP servers, plugins, and tools installed only on your machine. A thread that Claude runs on your machine through [Remote Control](/docs/en/remote-control) uses what's installed there. To make each of these available to cloud threads:
 
-* Skills, subagents, and commands: commit them to a repository you added to the project, for example a skill at `.claude/skills/<skill-name>/SKILL.md`. Each cloud thread clones every repository in the project and loads `.claude/skills/`, `.claude/agents/`, and `.claude/commands/` from each of them, so a skill committed to one repository is available in every cloud thread. Cloud threads also load the [skills you enable for your claude.ai account](/docs/en/skills#skills-in-cowork-and-cloud-sessions).
+* Skills: enable them for your claude.ai account, and cloud threads load the [skills you enable there](/docs/en/skills#skills-in-cowork-and-cloud-sessions), apart from [some threads in a self-hosted environment](/docs/en/self-hosted-environments-configuration#how-each-session’s-config-is-assembled). Skills committed to a repository's `.claude/skills/` don't load in cloud threads.
 * Plugins: add them in **Project settings > Plugins**; they load into each new cloud thread. Plugins that a repository declares in its `.claude/settings.json` [don't load in cloud threads](/docs/en/cloud-environments#what-carries-over-from-your-setup).
-* MCP servers: cloud threads get their MCP tools from the connectors on your claude.ai account, which are MCP servers you connect once at [claude.ai/customize/connectors](https://claude.ai/customize/connectors). Every cloud thread can use all of them with no per-project setup. The project conversation itself has no connectors, so send work that needs one as a task for a cloud thread. In a project with one repository, cloud threads also load MCP servers from that repository's [`.mcp.json`](/docs/en/cloud-environments#what-carries-over-from-your-setup). [How connectors reach Claude Code](/docs/en/mcp#how-connectors-reach-claude-code) lists the rules for cloud sessions and the settings that turn connectors off.
+* MCP servers: cloud threads get their MCP tools from the connectors on your claude.ai account, which are MCP servers you connect once at [claude.ai/customize/connectors](https://claude.ai/customize/connectors). Every cloud thread can use all of them with no per-project setup. The project conversation itself has no connectors, so send work that needs one as a task for a cloud thread. [How connectors reach Claude Code](/docs/en/mcp#how-connectors-reach-claude-code) lists the rules for cloud sessions and the settings that turn connectors off.
 * Command-line tools and packages: install them in the environment's [setup script](/docs/en/cloud-environments#setup-scripts).
 
 To see which connectors a running cloud thread has at claude.ai/code, open the thread and select **Connectors** from the **+** menu beside its message box. Turning a connector off there removes it from that thread and saves that as your account default, so new threads and claude.ai chats start without it until you turn it back on. A cloud thread picks up a connector you add or reconnect after the next message you send it.