Input and output

Handle approvals and user input

Handle approvals and user input 这一页讲的，就是 Handle approvals and user input 这件事在 Claude Code 里到底怎么用。

页面信息

对应原页

页面性质

第三方中文解释页

使用建议

先看人话解释，再对照原页命令和代码

这页不是官方原文，而是顺着官方文档结构做的中文解释版。命令、参数、配置名这些硬东西尽量保留，解释部分则尽量讲成人能照着做的话。

如果你碰到特别敏感的配置、权限或企业环境差异，最好顺手点上面的“查看原始文档”再核一遍。

这一页先讲明白

这页主要讲 Handle approvals and user input：Surface Claude's approval requests and clarifying questions to users, then return their decisions to the SDK.

你可以把它当成"Input and output"这块里专门管这一摊事的说明书。

你可以把"Handle approvals and user input"理解成 Input and output 这一栏里的一把专门工具。这页不是让你背书，而是教你什么时候该把这把工具拿出来。

原文这页大多会按 Detect when Claude needs input、Handle tool approval requests、Respond to tool requests、Handle clarifying questions 这些环节往下讲。

翻成人话，大概就是：Detect when Claude needs input

第一，先别一上来全开全配。先按最小一步试通，确认没跑偏，再继续往下加。

第二，命令、配置名、参数名这些硬东西尽量保留原样。人话解释是帮你听懂，不是帮你改关键字。

第三，照着原文这几个环节挨个过：Detect when Claude needs input -> Handle tool approval requests -> Respond to tool requests -> Handle clarifying questions。像下地先看水路、再试机器、再正式开干，一步一步最稳。

关键片段

原页关键片段：Detect when Claude needs input

这一段要真抓重点，通常就抓下面这块原文。

async def handle_tool_request(tool_name, input_data, context):
    # Prompt user and return allow or deny
    ...


options = ClaudeAgentOptions(can_use_tool=handle_tool_request)

关键片段

原页关键片段：Handle tool approval requests

先看下面这块原始片段，等会儿再回头看解释会顺得多。

import asyncio

from claude_agent_sdk import ClaudeAgentOptions, ResultMessage, query
from claude_agent_sdk.types import (
    HookMatcher,
    PermissionResultAllow,
    PermissionResultDeny,
    ToolPermissionContext,
)


async def can_use_tool(
    tool_name: str, input_data: dict, context: ToolPermissionContext
) -> PermissionResultAllow | PermissionResultDeny:
    # Display the tool request
    print(f"\nTool: {tool_name}")
    if tool_name == "Bash":
        print(f"Command: {input_data.get('command')}")
        if input_data.get("description"):
            print(f"Description: {input_data.get('description')}")
    else:
        print(f"Input: {input_data}")

    # Get user approval
    response = input("Allow this action? (y/n): ")

    # Return allow or deny based on user's response
    if response.lower() == "y":
        # Allow: tool executes with the original (or modified) input
        return PermissionResultAllow(updated_input=input_data)
    else:
        # Deny: tool doesn't execute, Claude sees the message
        return PermissionResultDeny(message="User denied this action")


# Required workaround: dummy hook keeps the stream open for can_use_tool
async def dummy_hook(input_data, tool_use_id, context):
    return {"continue_": True}


async def prompt_stream():
    yield {
        "type": "user",
        "message": {
            "role": "user",
            "content": "Create a test file in /tmp and then delete it",
        },
    }


async def main():
    async for message in query(
        prompt=prompt_stream(),
        options=ClaudeAgentOptions(
            can_use_tool=can_use_tool,
            hooks={"PreToolUse": [HookMatcher(matcher=None, hooks=[dummy_hook])]},
        ),
    ):
        if isinstance(message, ResultMessage) and message.subtype == "success":
            print(message.result)


asyncio.run(main())

关键片段

原页关键片段：Respond to tool requests 1

这一段要真抓重点，通常就抓下面这块原文。

from claude_agent_sdk.types import PermissionResultAllow, PermissionResultDeny

# Allow the tool to execute
return PermissionResultAllow(updated_input=input_data)

# Block the tool
return PermissionResultDeny(message="User rejected this action")

关键片段

原页关键片段：Respond to tool requests 2

这一段要真抓重点，通常就抓下面这块原文。

async def can_use_tool(tool_name, input_data, context):
    print(f"Claude wants to use {tool_name}")
    approved = await ask_user("Allow this action?")

    if approved:
        return PermissionResultAllow(updated_input=input_data)
    return PermissionResultDeny(message="User declined")

关键片段

原页关键片段：Respond to tool requests 3

这一段要真抓重点，通常就抓下面这块原文。

async def can_use_tool(tool_name, input_data, context):
    if tool_name == "Bash":
        # User approved, but scope all commands to sandbox
        sandboxed_input = {**input_data}
        sandboxed_input["command"] = input_data["command"].replace(
            "/tmp", "/tmp/sandbox"
        )
        return PermissionResultAllow(updated_input=sandboxed_input)
    return PermissionResultAllow(updated_input=input_data)

关键片段

原页关键片段：Handle clarifying questions 1

先看下面这块原始片段，等会儿再回头看解释会顺得多。

async for message in query(
    prompt="Analyze this codebase",
    options=ClaudeAgentOptions(
        # Include AskUserQuestion in your tools list
        tools=["Read", "Glob", "Grep", "AskUserQuestion"],
        can_use_tool=can_use_tool,
    ),
):
    print(message)

关键片段

原页关键片段：Handle clarifying questions 2

先看下面这块原始片段，等会儿再回头看解释会顺得多。

async def can_use_tool(tool_name: str, input_data: dict, context):
    if tool_name == "AskUserQuestion":
        # Your implementation to collect answers from the user
        return await handle_clarifying_questions(input_data)
    # Handle other tools normally
    return await prompt_for_approval(tool_name, input_data)

改配置

原页关键片段：Handle clarifying questions 3

这一段说完，最后还得写到配置里才算真的生效。

{
  "questions": [
    {
      "question": "How should I format the output?",
      "header": "Format",
      "options": [
        { "label": "Summary", "description": "Brief overview" },
        { "label": "Detailed", "description": "Full explanation" }
      ],
      "multiSelect": false
    },
    {
      "question": "Which sections should I include?",
      "header": "Sections",
      "options": [
        { "label": "Introduction", "description": "Opening context" },
        { "label": "Conclusion", "description": "Final summary" }
      ],
      "multiSelect": true
    }
  ]
}

改配置

原页关键片段：Question format 1

想把这条规矩固定住，就把下面这块老老实实写进去。

{
  "questions": [
    {
      "question": "How should I format the output?",
      "header": "Format",
      "options": [
        { "label": "Summary", "description": "Brief overview of key points" },
        { "label": "Detailed", "description": "Full explanation with examples" }
      ],
      "multiSelect": false
    }
  ]
}

终端里敲

原页关键片段：Question format 2

看到这里，别光点头，下面这条命令先跑起来再说。

import { query } from "@anthropic-ai/claude-agent-sdk";

for await (const message of query({
  prompt: "Help me choose a card layout",
  options: {
    toolConfig: {
      askUserQuestion: { previewFormat: "html" }
    },
    canUseTool: async (toolName, input) => {
      // input.questions[].options[].preview is an HTML string or undefined
      return { behavior: "allow", updatedInput: input };
    }
  }
})) {
  // ...
}

改配置

原页关键片段：Question format 3

想把这条规矩固定住，就把下面这块老老实实写进去。

{
  "label": "Compact",
  "description": "Title and metric value only",
  "preview": "<div style=\"padding:12px;border:1px solid #ddd;border-radius:8px\"><div style=\"font-size:12px;color:#666\">Active users</div><div style=\"font-size:28px;font-weight:600\">1,284</div></div>"
}

改配置

原页关键片段：Response format

这会儿轮到改配置了，字段名和关键字别自己乱换。

{
  "questions": [
    // ...
  ],
  "answers": {
    "How should I format the output?": "Summary",
    "Which sections should I include?": ["Introduction", "Conclusion"]
  }
}

预留广告位

正文中段响应式广告 等你后面真接 AdSense，这里再放正式广告。

Documentation Index

这里不是让你背"Documentation Index"这个词，而是让你看它真干活时怎么使。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

Detect when Claude needs input

这一段主要是在把"Detect when Claude needs input"讲实，不是只摆个标题给你看。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

关键片段

Detect when Claude needs input

这一段要真抓重点，通常就抓下面这块原文。

async def handle_tool_request(tool_name, input_data, context):
    # Prompt user and return allow or deny
    ...


options = ClaudeAgentOptions(can_use_tool=handle_tool_request)

Handle tool approval requests

看到这里，就把"Handle tool approval requests"当成一件真要上手的活来看。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

关键片段

Handle tool approval requests

先看下面这块原始片段，等会儿再回头看解释会顺得多。

import asyncio

from claude_agent_sdk import ClaudeAgentOptions, ResultMessage, query
from claude_agent_sdk.types import (
    HookMatcher,
    PermissionResultAllow,
    PermissionResultDeny,
    ToolPermissionContext,
)


async def can_use_tool(
    tool_name: str, input_data: dict, context: ToolPermissionContext
) -> PermissionResultAllow | PermissionResultDeny:
    # Display the tool request
    print(f"\nTool: {tool_name}")
    if tool_name == "Bash":
        print(f"Command: {input_data.get('command')}")
        if input_data.get("description"):
            print(f"Description: {input_data.get('description')}")
    else:
        print(f"Input: {input_data}")

    # Get user approval
    response = input("Allow this action? (y/n): ")

    # Return allow or deny based on user's response
    if response.lower() == "y":
        # Allow: tool executes with the original (or modified) input
        return PermissionResultAllow(updated_input=input_data)
    else:
        # Deny: tool doesn't execute, Claude sees the message
        return PermissionResultDeny(message="User denied this action")


# Required workaround: dummy hook keeps the stream open for can_use_tool
async def dummy_hook(input_data, tool_use_id, context):
    return {"continue_": True}


async def prompt_stream():
    yield {
        "type": "user",
        "message": {
            "role": "user",
            "content": "Create a test file in /tmp and then delete it",
        },
    }


async def main():
    async for message in query(
        prompt=prompt_stream(),
        options=ClaudeAgentOptions(
            can_use_tool=can_use_tool,
            hooks={"PreToolUse": [HookMatcher(matcher=None, hooks=[dummy_hook])]},
        ),
    ):
        if isinstance(message, ResultMessage) and message.subtype == "success":
            print(message.result)


asyncio.run(main())

Respond to tool requests

这一段主要是在把"Respond to tool requests"讲实，不是只摆个标题给你看。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

关键片段

Respond to tool requests 1

这一段要真抓重点，通常就抓下面这块原文。

from claude_agent_sdk.types import PermissionResultAllow, PermissionResultDeny

# Allow the tool to execute
return PermissionResultAllow(updated_input=input_data)

# Block the tool
return PermissionResultDeny(message="User rejected this action")

关键片段

Respond to tool requests 2

这一段要真抓重点，通常就抓下面这块原文。

async def can_use_tool(tool_name, input_data, context):
    print(f"Claude wants to use {tool_name}")
    approved = await ask_user("Allow this action?")

    if approved:
        return PermissionResultAllow(updated_input=input_data)
    return PermissionResultDeny(message="User declined")

关键片段

Respond to tool requests 3

这一段要真抓重点，通常就抓下面这块原文。

async def can_use_tool(tool_name, input_data, context):
    if tool_name == "Bash":
        # User approved, but scope all commands to sandbox
        sandboxed_input = {**input_data}
        sandboxed_input["command"] = input_data["command"].replace(
            "/tmp", "/tmp/sandbox"
        )
        return PermissionResultAllow(updated_input=sandboxed_input)
    return PermissionResultAllow(updated_input=input_data)

关键片段

Respond to tool requests 4

这一段要真抓重点，通常就抓下面这块原文。

async def can_use_tool(tool_name, input_data, context):
    approved = await ask_user(f"Allow {tool_name}?")

    if not approved:
        return PermissionResultDeny(message="User rejected this action")
    return PermissionResultAllow(updated_input=input_data)

Handle clarifying questions

这一段更像在讲判断条件，什么时候该上，什么时候先别急。把触发条件看清，比背标题更重要。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

关键片段

Handle clarifying questions 1

先看下面这块原始片段，等会儿再回头看解释会顺得多。

async for message in query(
    prompt="Analyze this codebase",
    options=ClaudeAgentOptions(
        # Include AskUserQuestion in your tools list
        tools=["Read", "Glob", "Grep", "AskUserQuestion"],
        can_use_tool=can_use_tool,
    ),
):
    print(message)

关键片段

Handle clarifying questions 2

先看下面这块原始片段，等会儿再回头看解释会顺得多。

async def can_use_tool(tool_name: str, input_data: dict, context):
    if tool_name == "AskUserQuestion":
        # Your implementation to collect answers from the user
        return await handle_clarifying_questions(input_data)
    # Handle other tools normally
    return await prompt_for_approval(tool_name, input_data)

改配置

Handle clarifying questions 3

这一段说完，最后还得写到配置里才算真的生效。

{
  "questions": [
    {
      "question": "How should I format the output?",
      "header": "Format",
      "options": [
        { "label": "Summary", "description": "Brief overview" },
        { "label": "Detailed", "description": "Full explanation" }
      ],
      "multiSelect": false
    },
    {
      "question": "Which sections should I include?",
      "header": "Sections",
      "options": [
        { "label": "Introduction", "description": "Opening context" },
        { "label": "Conclusion", "description": "Final summary" }
      ],
      "multiSelect": true
    }
  ]
}

关键片段

Handle clarifying questions 4

先看下面这块原始片段，等会儿再回头看解释会顺得多。

return PermissionResultAllow(
    updated_input={
        "questions": input_data.get("questions", []),
        "answers": {
            "How should I format the output?": "Summary",
            "Which sections should I include?": ["Introduction", "Conclusion"],
        },
    }
)

Question format

这一段不只是挂个标题，它是在说明"Question format"这一块到底负责什么。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

改配置

Question format 1

想把这条规矩固定住，就把下面这块老老实实写进去。

{
  "questions": [
    {
      "question": "How should I format the output?",
      "header": "Format",
      "options": [
        { "label": "Summary", "description": "Brief overview of key points" },
        { "label": "Detailed", "description": "Full explanation with examples" }
      ],
      "multiSelect": false
    }
  ]
}

终端里敲

Question format 2

看到这里，别光点头，下面这条命令先跑起来再说。

import { query } from "@anthropic-ai/claude-agent-sdk";

for await (const message of query({
  prompt: "Help me choose a card layout",
  options: {
    toolConfig: {
      askUserQuestion: { previewFormat: "html" }
    },
    canUseTool: async (toolName, input) => {
      // input.questions[].options[].preview is an HTML string or undefined
      return { behavior: "allow", updatedInput: input };
    }
  }
})) {
  // ...
}

改配置

Question format 3

想把这条规矩固定住，就把下面这块老老实实写进去。

{
  "label": "Compact",
  "description": "Title and metric value only",
  "preview": "<div style=\"padding:12px;border:1px solid #ddd;border-radius:8px\"><div style=\"font-size:12px;color:#666\">Active users</div><div style=\"font-size:28px;font-weight:600\">1,284</div></div>"
}

Response format

这一段主要是在把"Response format"讲实，不是只摆个标题给你看。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

改配置

Response format

这会儿轮到改配置了，字段名和关键字别自己乱换。

{
  "questions": [
    // ...
  ],
  "answers": {
    "How should I format the output?": "Summary",
    "Which sections should I include?": ["Introduction", "Conclusion"]
  }
}

Complete example

这一块主要是在说"Complete example"真到手上该怎么用，哪里最容易踩坑。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

关键片段

Complete example

"Complete example"这一段里最要紧的原始写法在下面，先看它怎么落地。

import asyncio

from claude_agent_sdk import ClaudeAgentOptions, ResultMessage, query
from claude_agent_sdk.types import HookMatcher, PermissionResultAllow


def parse_response(response: str, options: list) -> str:
    """Parse user input as option number(s) or free text."""
    try:
        indices = [int(s.strip()) - 1 for s in response.split(",")]
        labels = [options[i]["label"] for i in indices if 0 <= i < len(options)]
        return ", ".join(labels) if labels else response
    except ValueError:
        return response


async def handle_ask_user_question(input_data: dict) -> PermissionResultAllow:
    """Display Claude's questions and collect user answers."""
    answers = {}

    for q in input_data.get("questions", []):
        print(f"\n{q['header']}: {q['question']}")

        options = q["options"]
        for i, opt in enumerate(options):
            print(f"  {i + 1}. {opt['label']} - {opt['description']}")
        if q.get("multiSelect"):
            print("  (Enter numbers separated by commas, or type your own answer)")
        else:
            print("  (Enter a number, or type your own answer)")

        response = input("Your choice: ").strip()
        answers[q["question"]] = parse_response(response, options)

    return PermissionResultAllow(
        updated_input={
            "questions": input_data.get("questions", []),
            "answers": answers,
        }
    )


async def can_use_tool(
    tool_name: str, input_data: dict, context
) -> PermissionResultAllow:
    # Route AskUserQuestion to our question handler
    if tool_name == "AskUserQuestion":
        return await handle_ask_user_question(input_data)
    # Auto-approve other tools for this example
    return PermissionResultAllow(updated_input=input_data)


async def prompt_stream():
    yield {
        "type": "user",
        "message": {
            "role": "user",
            "content": "Help me decide on the tech stack for a new mobile app",
        },
    }


# Required workaround: dummy hook keeps the stream open for can_use_tool
async def dummy_hook(input_data, tool_use_id, context):
    return {"continue_": True}


async def main():
    async for message in query(
        prompt=prompt_stream(),
        options=ClaudeAgentOptions(
            can_use_tool=can_use_tool,
            hooks={"PreToolUse": [HookMatcher(matcher=None, hooks=[dummy_hook])]},
        ),
    ):
        if isinstance(message, ResultMessage) and message.subtype == "success":
            print(message.result)


asyncio.run(main())

Limitations

这一块主要是在说"Limitations"真到手上该怎么用，哪里最容易踩坑。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

Other ways to get user input

这段看着像个标题，其实是在说"Other ways to get user input"管到哪儿。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

Streaming input

这里不是让你背"Streaming input"这个词，而是让你看它真干活时怎么使。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

Custom tools

这一段主要是在把"Custom tools"讲实，不是只摆个标题给你看。

这里还牵扯作用域，意思就是这条规则到底管当前项目、你个人，还是只管这一趟会话。

Related resources

这一段是在把 permissions: set up permission modes and rules 配起来、配稳当。你主要盯住在哪儿改、怎么写、改完怎么看它真生效。

看这段时要特别盯工具和权限边界，别为了省事一把全开。

照着做一遍

如果你不想来回翻，就先照这几步顺着做。

每做完一步就看一下结果，再决定要不要继续往下。

关键片段

第 1 步：Detect when Claude needs input

这一段要真抓重点，通常就抓下面这块原文。

async def handle_tool_request(tool_name, input_data, context):
    # Prompt user and return allow or deny
    ...


options = ClaudeAgentOptions(can_use_tool=handle_tool_request)

关键片段

第 2 步：Handle tool approval requests

先看下面这块原始片段，等会儿再回头看解释会顺得多。

import asyncio

from claude_agent_sdk import ClaudeAgentOptions, ResultMessage, query
from claude_agent_sdk.types import (
    HookMatcher,
    PermissionResultAllow,
    PermissionResultDeny,
    ToolPermissionContext,
)


async def can_use_tool(
    tool_name: str, input_data: dict, context: ToolPermissionContext
) -> PermissionResultAllow | PermissionResultDeny:
    # Display the tool request
    print(f"\nTool: {tool_name}")
    if tool_name == "Bash":
        print(f"Command: {input_data.get('command')}")
        if input_data.get("description"):
            print(f"Description: {input_data.get('description')}")
    else:
        print(f"Input: {input_data}")

    # Get user approval
    response = input("Allow this action? (y/n): ")

    # Return allow or deny based on user's response
    if response.lower() == "y":
        # Allow: tool executes with the original (or modified) input
        return PermissionResultAllow(updated_input=input_data)
    else:
        # Deny: tool doesn't execute, Claude sees the message
        return PermissionResultDeny(message="User denied this action")


# Required workaround: dummy hook keeps the stream open for can_use_tool
async def dummy_hook(input_data, tool_use_id, context):
    return {"continue_": True}


async def prompt_stream():
    yield {
        "type": "user",
        "message": {
            "role": "user",
            "content": "Create a test file in /tmp and then delete it",
        },
    }


async def main():
    async for message in query(
        prompt=prompt_stream(),
        options=ClaudeAgentOptions(
            can_use_tool=can_use_tool,
            hooks={"PreToolUse": [HookMatcher(matcher=None, hooks=[dummy_hook])]},
        ),
    ):
        if isinstance(message, ResultMessage) and message.subtype == "success":
            print(message.result)


asyncio.run(main())

关键片段

第 3 步：Respond to tool requests 1

这一段要真抓重点，通常就抓下面这块原文。

from claude_agent_sdk.types import PermissionResultAllow, PermissionResultDeny

# Allow the tool to execute
return PermissionResultAllow(updated_input=input_data)

# Block the tool
return PermissionResultDeny(message="User rejected this action")

关键片段

第 4 步：Respond to tool requests 2

这一段要真抓重点，通常就抓下面这块原文。

async def can_use_tool(tool_name, input_data, context):
    print(f"Claude wants to use {tool_name}")
    approved = await ask_user("Allow this action?")

    if approved:
        return PermissionResultAllow(updated_input=input_data)
    return PermissionResultDeny(message="User declined")

一眼看懂这一页

先把这页到底在讲什么看明白，再去碰具体命令和配置，最不容易绕晕。

Handle approvals and user input
   |
   v
这是 Input and output 里的一摊要紧活
   |
   v
先弄懂，再下手

文末提醒

这站会按官方 docs 的导航和内容变化继续重生成，原站加页、删页、改页时，这里会跟着更新。

人话解释会尽量顺着原页往下讲，但命令、参数名、配置名这些硬东西还是保留原样，免得你抄过去跑不起来。