docs: Claude Code v2.1.220 - deep subagent stream forwarding, MCP CI error detection, hook security

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

@@ -31,6 +31,8 @@ If you use Claude Code over SSH, set the environment variable or setting on the
 When the mode is on, the first thing Claude Code prints is a confirmation line naming the method that turned it on: `[Screen Reader Mode: on via flag]`, `[Screen Reader Mode: on via env]`, or `[Screen Reader Mode: on via settings]`. The method-naming format requires Claude Code v2.1.206 or later. When Claude Code relaunches itself, for example to finish installing an update, the new process inherits the mode through the `CLAUDE_AX_SCREEN_READER` environment variable, so its confirmation line reads `[Screen Reader Mode: on via env]` regardless of which method you used.
 {/* max-version: 2.1.205 */}Earlier versions print `[Accessible screen reader mode: on]`.
 
+After printing the confirmation line, Claude Code holds the rest of the interface back for three seconds so your screen reader can finish speaking the line, then renders the first prompt. Press any key to end the hold early. To change the hold's length, set the `CLAUDE_AX_STARTUP_QUIET_MS` environment variable to a number of milliseconds. The default is `3000`; set it to `0` to skip the hold. Claude Code caps the hold at `600000` milliseconds, 10 minutes. Requires Claude Code v2.1.217 or later.
+
 ## Turn off screen reader mode
 
 Reverse whichever method turned the mode on: start without the flag, unset the environment variable, or set `axScreenReader` to `false`. Setting `CLAUDE_AX_SCREEN_READER=0` keeps the mode off even when the setting is `true`.
@@ -62,6 +64,8 @@ Each message in the transcript starts with a label your screen reader announces,
 
 The terminal cursor follows the input caret, so a screen reader's read-current-line command answers "where am I" with the prompt you're editing.
 
+{/* min-version: 2.1.219 */}As you type at the end of the input line, Claude Code writes only the characters you type, so your screen reader echoes only those characters. Requires Claude Code v2.1.219 or later; earlier versions rewrite the whole input line on every keystroke, so the screen reader re-reads it as you type.
+
 {/* min-version: 2.1.218 */}When you delete a word or a line in the input, Claude Code announces the deleted text. Requires Claude Code v2.1.218 or later. The announcement covers:
 
 * Deleting a word with `Ctrl+W`, `Option+Delete` on macOS, or `Ctrl+Backspace` on Windows

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

@@ -14,7 +14,9 @@ When you start an agent, the SDK runs the same [execution loop that powers Claud
 
 Every agent session follows the same cycle:
 
-<img src="https://mintcdn.com/claude-code/ikqp3_70mqIahteV/images/agent-loop-diagram.svg?fit=max&auto=format&n=ikqp3_70mqIahteV&q=85&s=1c6e8f28d80dba14a7287419656f1237" alt="Diagram of the agent loop: your prompt enters the agentic loop, where Claude evaluates and either requests tool calls, whose results feed back into another evaluation, or returns the final answer" width="720" height="212" data-path="images/agent-loop-diagram.svg" />
+<img src="https://mintcdn.com/claude-code/ikqp3_70mqIahteV/images/agent-loop-diagram.svg?fit=max&auto=format&n=ikqp3_70mqIahteV&q=85&s=1c6e8f28d80dba14a7287419656f1237" className="dark:hidden" alt="Diagram of the agent loop: your prompt enters the agentic loop, where Claude evaluates and either requests tool calls, whose results feed back into another evaluation, or returns the final answer" width="720" height="212" data-path="images/agent-loop-diagram.svg" />
+
+<img src="https://mintcdn.com/claude-code/_xqph1dUOslCOwsj/images/agent-loop-diagram-dark.svg?fit=max&auto=format&n=_xqph1dUOslCOwsj&q=85&s=afe723c52a324d3c61fa72fb02432ab6" className="hidden dark:block" alt="Diagram of the agent loop: your prompt enters the agentic loop, where Claude evaluates and either requests tool calls, whose results feed back into another evaluation, or returns the final answer" width="720" height="212" data-path="images/agent-loop-diagram-dark.svg" />
 
 1. **Receive prompt.** Claude receives your prompt, along with the system prompt, tool definitions, and conversation history. The SDK yields a [`SystemMessage`](#message-types) with subtype `"init"` containing session metadata.
 2. **Evaluate and respond.** Claude evaluates the current state and determines how to proceed. It may respond with text, request one or more tool calls, or both. The SDK yields an [`AssistantMessage`](#message-types) containing the text and any tool call requests.

content/en/docs/claude-code/agent-sdk/cost-tracking.md

@@ -37,7 +37,9 @@ Cost tracking depends on understanding how the SDK scopes usage data:
 
 The following diagram shows the message stream from a single `query()` call, with token usage reported at each step and the cumulative estimate at the end:
 
-<img src="https://mintcdn.com/claude-code/ikqp3_70mqIahteV/images/agent-sdk/message-usage-flow.svg?fit=max&auto=format&n=ikqp3_70mqIahteV&q=85&s=68497aee338e01cc745323af7aea378e" alt="Diagram showing a query producing two steps of messages. Step 1 has four assistant messages sharing the same ID and usage (count once), Step 2 has one assistant message with a new ID, and the final result message shows the estimated total_cost_usd." width="760" height="520" data-path="images/agent-sdk/message-usage-flow.svg" />
+<img src="https://mintcdn.com/claude-code/ikqp3_70mqIahteV/images/agent-sdk/message-usage-flow.svg?fit=max&auto=format&n=ikqp3_70mqIahteV&q=85&s=68497aee338e01cc745323af7aea378e" className="dark:hidden" alt="Diagram showing a query producing two steps of messages. Step 1 has four assistant messages sharing the same ID and usage (count once), Step 2 has one assistant message with a new ID, and the final result message shows the estimated total_cost_usd." width="760" height="520" data-path="images/agent-sdk/message-usage-flow.svg" />
+
+<img src="https://mintcdn.com/claude-code/_xqph1dUOslCOwsj/images/agent-sdk/message-usage-flow-dark.svg?fit=max&auto=format&n=_xqph1dUOslCOwsj&q=85&s=8ea95085abc0a6b7f55ecef498bd4d14" className="hidden dark:block" alt="Diagram showing a query producing two steps of messages. Step 1 has four assistant messages sharing the same ID and usage (count once), Step 2 has one assistant message with a new ID, and the final result message shows the estimated total_cost_usd." width="760" height="520" data-path="images/agent-sdk/message-usage-flow-dark.svg" />
 
 <Steps>
   <Step title="Each step produces assistant messages">

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

@@ -20,7 +20,9 @@ If you do not need infrastructure control, custom isolation, or your own data pl
 
 Every hosting decision on this page follows from how the SDK runs the agent. When your code calls `query()`, the SDK spawns a separate `claude` CLI process and talks to it over stdio. That subprocess owns the shell, the working directory, and the JSONL session transcripts on local disk.
 
-<img src="https://mintcdn.com/claude-code/ikqp3_70mqIahteV/images/agent-sdk/hosting-subprocess.svg?fit=max&auto=format&n=ikqp3_70mqIahteV&q=85&s=9dac857ca9d3b1410c3734900c386004" alt="Request flow: client to your app, which spawns a claude CLI subprocess over stdio inside the container; the subprocess writes to local disk and calls api.anthropic.com over HTTPS" width="920" height="220" data-path="images/agent-sdk/hosting-subprocess.svg" />
+<img src="https://mintcdn.com/claude-code/ikqp3_70mqIahteV/images/agent-sdk/hosting-subprocess.svg?fit=max&auto=format&n=ikqp3_70mqIahteV&q=85&s=9dac857ca9d3b1410c3734900c386004" className="dark:hidden" alt="Request flow: client to your app, which spawns a claude CLI subprocess over stdio inside the container; the subprocess writes to local disk and calls api.anthropic.com over HTTPS" width="920" height="220" data-path="images/agent-sdk/hosting-subprocess.svg" />
+
+<img src="https://mintcdn.com/claude-code/_xqph1dUOslCOwsj/images/agent-sdk/hosting-subprocess-dark.svg?fit=max&auto=format&n=_xqph1dUOslCOwsj&q=85&s=3fdeff3d7f44b2b67762668acfbb25f5" className="hidden dark:block" alt="Request flow: client to your app, which spawns a claude CLI subprocess over stdio inside the container; the subprocess writes to local disk and calls api.anthropic.com over HTTPS" width="920" height="220" data-path="images/agent-sdk/hosting-subprocess-dark.svg" />
 
 One agent session maps to one subprocess. Running N concurrent sessions means N subprocesses, each with its own process tree and transcript file. By default they all inherit your application's working directory, so pass `cwd` on each `query()` call when sessions need separate filesystems:
 

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

@@ -46,7 +46,9 @@ When Claude requests a tool, the SDK checks permissions in this order:
   </Step>
 </Steps>
 
-<img src="https://mintcdn.com/claude-code/jYgs7qigNjO1Badj/images/agent-sdk/permissions-flow.svg?fit=max&auto=format&n=jYgs7qigNjO1Badj&q=85&s=c771ad9085b1277d3708027a49c744bc" alt="Diagram of the six-step permission evaluation flow matching the steps above: a tool request passes through hooks, deny rules, ask rules, permission mode, allow rules, and canUseTool. Hooks, deny rules, and canUseTool can route down to Blocked; permission mode bypass, allow rules, and canUseTool can route up to Execute; ask rules route to canUseTool." width="1180" height="260" data-path="images/agent-sdk/permissions-flow.svg" />
+<img src="https://mintcdn.com/claude-code/jYgs7qigNjO1Badj/images/agent-sdk/permissions-flow.svg?fit=max&auto=format&n=jYgs7qigNjO1Badj&q=85&s=c771ad9085b1277d3708027a49c744bc" className="dark:hidden" alt="Diagram of the six-step permission evaluation flow matching the steps above: a tool request passes through hooks, deny rules, ask rules, permission mode, allow rules, and canUseTool. Hooks, deny rules, and canUseTool can route down to Blocked; permission mode bypass, allow rules, and canUseTool can route up to Execute; ask rules route to canUseTool." width="1180" height="260" data-path="images/agent-sdk/permissions-flow.svg" />
+
+<img src="https://mintcdn.com/claude-code/_xqph1dUOslCOwsj/images/agent-sdk/permissions-flow-dark.svg?fit=max&auto=format&n=_xqph1dUOslCOwsj&q=85&s=e53a91e9059cbf51852b7cedb4dd4251" className="hidden dark:block" alt="Diagram of the six-step permission evaluation flow matching the steps above: a tool request passes through hooks, deny rules, ask rules, permission mode, allow rules, and canUseTool. Hooks, deny rules, and canUseTool can route down to Blocked; permission mode bypass, allow rules, and canUseTool can route up to Execute; ask rules route to canUseTool." width="1180" height="260" data-path="images/agent-sdk/permissions-flow-dark.svg" />
 
 As of v2.1.198, if you pass a `canUseTool` callback that this evaluation order can never reach, the TypeScript SDK emits a Node.js process warning once when the query is constructed. The warning's code is `CLAUDE_SDK_CAN_USE_TOOL_SHADOWED`. Two configurations trigger it:
 

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

@@ -4303,6 +4303,8 @@ type SDKTaskNotificationMessage = {
 };
 ```
 
+Claude Code prepends a notice to every task notification it sends to the model. The notice states that no human input has occurred, so the model doesn't treat the notification as a user instruction or approval. To detect a task-notification turn, check `origin.kind === "task-notification"` on [`SDKUserMessage`](#sdkusermessage) or [`SDKResultMessage`](#sdkresultmessage) rather than matching on the notice text. Before v2.1.205, Claude Code left the notice off notifications that arrived while the session was idle.
+
 ### `SDKToolUseSummaryMessage`
 
 Summary of tool usage in a conversation.

content/en/docs/claude-code/channels-reference.md

@@ -34,7 +34,9 @@ A channel is an [MCP](https://modelcontextprotocol.io) server that runs on the s
 * **Chat platforms** (Telegram, Discord): your plugin runs locally and polls the platform's API for new messages. When someone DMs your bot, the plugin receives the message and forwards it to Claude. No URL to expose.
 * **Webhooks** (CI, monitoring): your server listens on a local HTTP port. External systems POST to that port, and your server pushes the payload to Claude.
 
-<img src="https://mintcdn.com/claude-code/9FG0ZKj9uKYiHmbi/images/channel-architecture.svg?fit=max&auto=format&n=9FG0ZKj9uKYiHmbi&q=85&s=9a037b7da80184ae49015c0256b21a1f" alt="Architecture diagram showing external systems connecting to your local channel server, which communicates with Claude Code over stdio" width="600" height="220" data-path="images/channel-architecture.svg" />
+<img src="https://mintcdn.com/claude-code/9FG0ZKj9uKYiHmbi/images/channel-architecture.svg?fit=max&auto=format&n=9FG0ZKj9uKYiHmbi&q=85&s=9a037b7da80184ae49015c0256b21a1f" className="dark:hidden" alt="Architecture diagram showing external systems connecting to your local channel server, which communicates with Claude Code over stdio" width="600" height="220" data-path="images/channel-architecture.svg" />
+
+<img src="https://mintcdn.com/claude-code/_xqph1dUOslCOwsj/images/channel-architecture-dark.svg?fit=max&auto=format&n=_xqph1dUOslCOwsj&q=85&s=ae1e494440806a6a5d74a1279e22e162" className="hidden dark:block" alt="Architecture diagram showing external systems connecting to your local channel server, which communicates with Claude Code over stdio" width="600" height="220" data-path="images/channel-architecture-dark.svg" />
 
 ## What you need
 
@@ -452,7 +454,9 @@ When a permission prompt opens, the relay loop has four steps:
 
 The local terminal dialog stays open through all of this. If someone at the terminal answers before the remote verdict arrives, that answer is applied instead and the pending remote request is dropped.
 
-<img src="https://mintcdn.com/claude-code/9FG0ZKj9uKYiHmbi/images/channel-permission-relay.svg?fit=max&auto=format&n=9FG0ZKj9uKYiHmbi&q=85&s=97d57f128f0da55f105ab1e3a7e10240" alt="Sequence diagram: Claude Code sends a permission_request notification to the channel server, the server formats and sends the prompt to the chat app, the human replies with a verdict, and the server parses that reply into a permission notification back to Claude Code" width="600" height="230" data-path="images/channel-permission-relay.svg" />
+<img src="https://mintcdn.com/claude-code/9FG0ZKj9uKYiHmbi/images/channel-permission-relay.svg?fit=max&auto=format&n=9FG0ZKj9uKYiHmbi&q=85&s=97d57f128f0da55f105ab1e3a7e10240" className="dark:hidden" alt="Sequence diagram: Claude Code sends a permission_request notification to the channel server, the server formats and sends the prompt to the chat app, the human replies with a verdict, and the server parses that reply into a permission notification back to Claude Code" width="600" height="230" data-path="images/channel-permission-relay.svg" />
+
+<img src="https://mintcdn.com/claude-code/_xqph1dUOslCOwsj/images/channel-permission-relay-dark.svg?fit=max&auto=format&n=_xqph1dUOslCOwsj&q=85&s=368c8d9119a9a9cff5d826d806724842" className="hidden dark:block" alt="Sequence diagram: Claude Code sends a permission_request notification to the channel server, the server formats and sends the prompt to the chat app, the human replies with a verdict, and the server parses that reply into a permission notification back to Claude Code" width="600" height="230" data-path="images/channel-permission-relay-dark.svg" />
 
 ### Permission request fields
 

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

@@ -60,7 +60,7 @@ Chrome integration also requires signing in with `/login`. If you authenticate w
   <Step title="Ask Claude to use the browser">
     This example navigates to a page, interacts with it, and reports what it finds, all from your terminal or editor:
 
-    ```text theme={null}
+    ```text wrap theme={null}
     Go to code.claude.com/docs, click on the search box,
     type "hooks", and tell me what results appear
     ```
@@ -118,7 +118,7 @@ These examples show common ways to combine browser actions with coding tasks. Ru
 
 When developing a web app, ask Claude to verify your changes work correctly:
 
-```text theme={null}
+```text wrap theme={null}
 I just updated the login form validation. Can you open localhost:3000,
 try submitting the form with invalid data, and check if the error
 messages appear correctly?
@@ -130,7 +130,7 @@ Claude navigates to your local server, interacts with the form, and reports what
 
 Claude can read console output to help diagnose problems. Tell Claude what patterns to look for rather than asking for all console output, since logs can be verbose:
 
-```text theme={null}
+```text wrap theme={null}
 Open the dashboard page and check the console for any errors when
 the page loads.
 ```
@@ -141,7 +141,7 @@ Claude reads the console messages and can filter for specific patterns or error
 
 Speed up repetitive data entry tasks:
 
-```text theme={null}
+```text wrap theme={null}
 I have a spreadsheet of customer contacts in contacts.csv. For each row,
 go to the CRM at crm.example.com, click "Add Contact", and fill in the
 name, email, and phone fields.
@@ -155,7 +155,7 @@ Claude can attach files from your machine to upload fields on a page. Claude Cod
 
 This example attaches a log file to a form:
 
-```text theme={null}
+```text wrap theme={null}
 Open the bug tracker at bugs.example.com, create a new issue,
 and attach logs/session.log to it
 ```
@@ -170,7 +170,7 @@ Three restrictions apply to uploads:
 
 Use Claude to write directly in your documents without API setup:
 
-```text theme={null}
+```text wrap theme={null}
 Draft a project update based on the recent commits and add it to my
 Google Doc at docs.google.com/document/d/abc123
 ```
@@ -181,7 +181,7 @@ Claude opens the document, clicks into the editor, and types the content. This w
 
 Pull structured information from websites:
 
-```text theme={null}
+```text wrap theme={null}
 Go to the product listings page and extract the name, price, and
 availability for each item. Save the results as a CSV file.
 ```
@@ -192,7 +192,7 @@ Claude navigates to the page, reads the content, and compiles the data into a st
 
 Coordinate tasks across multiple websites:
 
-```text theme={null}
+```text wrap theme={null}
 Check my calendar for meetings tomorrow, then for each meeting with
 an external attendee, look up their company website and add a note
 about what they do.
@@ -204,7 +204,7 @@ Claude works across tabs to gather information and complete the workflow.
 
 Create shareable recordings of browser interactions:
 
-```text theme={null}
+```text wrap theme={null}
 Record a GIF showing how to complete the checkout flow, from adding
 an item to the cart through to the confirmation page.
 ```
@@ -215,7 +215,7 @@ Claude records the interaction sequence and saves it as a GIF file. The recordin
 
 Ask Claude to keep a screenshot as a file:
 
-```text theme={null}
+```text wrap theme={null}
 Take a screenshot of the checkout page and save it to disk
 ```