【从零手写 ClaudeCode：learn-claude-code 项目实战笔记】（4）Subagents (子智能体)

Flittly2026-03-09 14:42

第四章 Subagents (子智能体)

s01 > s02 > s03 > [ s04 ] s05 > s06 | s07 > s08 > s09 > s10 > s11 > s12

"本专栏基于开源项目 learn-claude-code 的官方文档。原文档非常硬核，为了方便像我一样的新手小白理解，我对文档进行了逐行精读 ，并加入了很多中文注释、大白话解释和踩坑记录。希望这套'咀嚼版'教程能帮你推开 AI Agent 开发的大门。"

项目地址：shareAI-lab/learn-claude-code: Bash is all you need - A nano Claude Code--like agent, built from 0 to 1

"大任务拆小, 每个小任务干净的上下文" -- 子智能体用独立 messages[], 不污染主对话。

一、问题-脏数据污染上下文

智能体工作越久, messages 数组越胖。每次读文件、跑命令的输出都永久留在上下文里。"这个项目用什么测试框架?" 可能要读 5 个文件, 但父智能体只需要一个词: "pytest。"

为什么会这样：在 LLM（大语言模型）的对话中，你和 AI 的所有聊天记录并不是像微信那样存在服务器里，而是每一次如果你想让 AI 记得前面的话，你必须把之前的每一句对话都打包发给它。

二、解决方案

sql 复制代码 
Parent agent                     Subagent
+------------------+             +------------------+
| messages=[...]   |             | messages=[]      | <-- fresh
|                  |  dispatch   |                  |
| tool: task       | ----------> | while tool_use:  |
|   prompt="..."   |             |   call tools     |
|                  |  summary    |   append results |
|   result = "..." | <---------- | return last text |
+------------------+             +------------------+

Parent context stays clean. Subagent context is discarded.

核心思想就是 "外包"

把那些需要大量翻找资料、产生大量中间过程数据的脏活累活，丢给一个临时的子智能体 去做。子智能体做完后，只把最终精华结论传回来，然后立刻销毁，从而保持主智能体（Parent Agent）的大脑（上下文）永远清爽、干净、高效。

三、工作原理

  1. 父智能体与子智能体的系统提示词对两者的角色进行了隔离，父智能体主要是拆解问题，子智能体主要是干活（跑代码、读文件），最重要的是干完活之后要写总结报告。
ini 复制代码 
SYSTEM = f"You are a coding agent at {WORKDIR}. Use the task tool to delegate exploration or subtasks."
SUBAGENT_SYSTEM = f"You are a coding subagent at {WORKDIR}. Complete the given task, then summarize your findings."
  1. 父智能体有一个 task 工具。子智能体拥有除 task 外的所有基础工具 (禁止递归生成)。
ini 复制代码 
PARENT_TOOLS = CHILD_TOOLS + [
    {"name": "task",
     "description": "Spawn a subagent with fresh context.",
     "input_schema": {
         "type": "object",
         "properties": {"prompt": {"type": "string"}},
         "required": ["prompt"],
     }},
]
  1. 子智能体以 messages=[] 启动, 运行自己的循环。只有最终文本返回给父智能体。
python 复制代码 
def run_subagent(prompt: str) -> str:
    sub_messages = [{"role": "user", "content": prompt}]
    for _ in range(30):  # safety limit
        response = client.messages.create(
            model=MODEL, system=SUBAGENT_SYSTEM,
            messages=sub_messages,
            tools=CHILD_TOOLS, max_tokens=8000,
        )
        sub_messages.append({"role": "assistant",
                             "content": response.content})
        if response.stop_reason != "tool_use":
            break
        results = []
        for block in response.content:
            if block.type == "tool_use":
                handler = TOOL_HANDLERS.get(block.name)
                output = handler(**block.input)
                results.append({"type": "tool_result",
                    "tool_use_id": block.id,
                    "content": str(output)[:50000]})
        sub_messages.append({"role": "user", "content": results})
    return "".join(
        b.text for b in response.content if hasattr(b, "text")
    ) or "(no summary)"

子智能体可能跑了 30+ 次工具调用, 但整个消息历史直接丢弃。父智能体收到的只是一段摘要文本, 作为普通 tool_result 返回。

四、相对 s03 的变更

组件 之前 (s03) 之后 (s04)
Tools 5 5 (基础) + task (仅父端)
上下文 单一共享 父 + 子隔离
Subagent 无 run_subagent() 函数
返回值 不适用 仅摘要文本

五、试一试

bash 复制代码 
cd learn-claude-code
python agents/s04_subagent.py

试试这些 prompt (英文 prompt 对 LLM 效果更好, 也可以用中文):

  1. Use a subtask to find what testing framework this project uses
  2. Delegate: read all .py files and summarize what each one does
  3. Use a task to create a new module, then verify it from here

六、完整代码

python 复制代码 
#!/usr/bin/env python3
import os
import subprocess
from pathlib import Path

from anthropic import Anthropic
from dotenv import load_dotenv

load_dotenv(override=True)

if os.getenv("ANTHROPIC_BASE_URL"):
    os.environ.pop("ANTHROPIC_AUTH_TOKEN", None)

WORKDIR = Path.cwd()
client = Anthropic(base_url=os.getenv("ANTHROPIC_BASE_URL"))
MODEL = os.environ["MODEL_ID"]

SYSTEM = f"You are a coding agent at {WORKDIR}. Use the task tool to delegate exploration or subtasks."
SUBAGENT_SYSTEM = f"You are a coding subagent at {WORKDIR}. Complete the given task, then summarize your findings."


# -- Tool implementations shared by parent and child --
def safe_path(p: str) -> Path:
    path = (WORKDIR / p).resolve()
    if not path.is_relative_to(WORKDIR):
        raise ValueError(f"Path escapes workspace: {p}")
    return path

def run_bash(command: str) -> str:
    dangerous = ["rm -rf /", "sudo", "shutdown", "reboot", "> /dev/"]
    if any(d in command for d in dangerous):
        return "Error: Dangerous command blocked"
    try:
        r = subprocess.run(command, shell=True, cwd=WORKDIR,
                           capture_output=True, text=True, timeout=120)
        out = (r.stdout + r.stderr).strip()
        return out[:50000] if out else "(no output)"
    except subprocess.TimeoutExpired:
        return "Error: Timeout (120s)"

def run_read(path: str, limit: int = None) -> str:
    try:
        lines = safe_path(path).read_text().splitlines()
        if limit and limit < len(lines):
            lines = lines[:limit] + [f"... ({len(lines) - limit} more)"]
        return "\n".join(lines)[:50000]
    except Exception as e:
        return f"Error: {e}"

def run_write(path: str, content: str) -> str:
    try:
        fp = safe_path(path)
        fp.parent.mkdir(parents=True, exist_ok=True)
        fp.write_text(content)
        return f"Wrote {len(content)} bytes"
    except Exception as e:
        return f"Error: {e}"

def run_edit(path: str, old_text: str, new_text: str) -> str:
    try:
        fp = safe_path(path)
        content = fp.read_text()
        if old_text not in content:
            return f"Error: Text not found in {path}"
        fp.write_text(content.replace(old_text, new_text, 1))
        return f"Edited {path}"
    except Exception as e:
        return f"Error: {e}"


TOOL_HANDLERS = {
    "bash":       lambda **kw: run_bash(kw["command"]),
    "read_file":  lambda **kw: run_read(kw["path"], kw.get("limit")),
    "write_file": lambda **kw: run_write(kw["path"], kw["content"]),
    "edit_file":  lambda **kw: run_edit(kw["path"], kw["old_text"], kw["new_text"]),
}

# Child gets all base tools except task (no recursive spawning)
CHILD_TOOLS = [
    {"name": "bash", "description": "Run a shell command.",
     "input_schema": {"type": "object", "properties": {"command": {"type": "string"}}, "required": ["command"]}},
    {"name": "read_file", "description": "Read file contents.",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "limit": {"type": "integer"}}, "required": ["path"]}},
    {"name": "write_file", "description": "Write content to file.",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "content": {"type": "string"}}, "required": ["path", "content"]}},
    {"name": "edit_file", "description": "Replace exact text in file.",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "old_text": {"type": "string"}, "new_text": {"type": "string"}}, "required": ["path", "old_text", "new_text"]}},
]


# -- Subagent: fresh context, filtered tools, summary-only return --
def run_subagent(prompt: str) -> str:
    sub_messages = [{"role": "user", "content": prompt}]  # fresh context
    for _ in range(30):  # safety limit
        response = client.messages.create(
            model=MODEL, system=SUBAGENT_SYSTEM, messages=sub_messages,
            tools=CHILD_TOOLS, max_tokens=8000,
        )
        sub_messages.append({"role": "assistant", "content": response.content})
        if response.stop_reason != "tool_use":
            break
        results = []
        for block in response.content:
            if block.type == "tool_use":
                handler = TOOL_HANDLERS.get(block.name)
                output = handler(**block.input) if handler else f"Unknown tool: {block.name}"
                results.append({"type": "tool_result", "tool_use_id": block.id, "content": str(output)[:50000]})
        sub_messages.append({"role": "user", "content": results})
    # Only the final text returns to the parent -- child context is discarded
    return "".join(b.text for b in response.content if hasattr(b, "text")) or "(no summary)"


# -- Parent tools: base tools + task dispatcher --
PARENT_TOOLS = CHILD_TOOLS + [
    {"name": "task", "description": "Spawn a subagent with fresh context. It shares the filesystem but not conversation history.",
     "input_schema": {"type": "object", "properties": {"prompt": {"type": "string"}, "description": {"type": "string", "description": "Short description of the task"}}, "required": ["prompt"]}},
]


def agent_loop(messages: list):
    while True:
        response = client.messages.create(
            model=MODEL, system=SYSTEM, messages=messages,
            tools=PARENT_TOOLS, max_tokens=8000,
        )
        messages.append({"role": "assistant", "content": response.content})
        if response.stop_reason != "tool_use":
            return
        results = []
        for block in response.content:
            if block.type == "tool_use":
                if block.name == "task":
                    desc = block.input.get("description", "subtask")
                    print(f"> task ({desc}): {block.input['prompt'][:80]}")
                    output = run_subagent(block.input["prompt"])
                else:
                    handler = TOOL_HANDLERS.get(block.name)
                    output = handler(**block.input) if handler else f"Unknown tool: {block.name}"
                print(f"  {str(output)[:200]}")
                results.append({"type": "tool_result", "tool_use_id": block.id, "content": str(output)})
        messages.append({"role": "user", "content": results})


if __name__ == "__main__":
    history = []
    while True:
        try:
            query = input("\033[36ms04 >> \033[0m")
        except (EOFError, KeyboardInterrupt):
            break
        if query.strip().lower() in ("q", "exit", ""):
            break
        history.append({"role": "user", "content": query})
        agent_loop(history)
        response_content = history[-1]["content"]
        if isinstance(response_content, list):
            for block in response_content:
                if hasattr(block, "text"):
                    print(block.text)
        print()
相关推荐
沸点小助手3 小时前
「AI 编程搭子真香or翻车」沸点获奖名单公示｜本周互动话题上新🎊
aigc·agent·ai编程
带刺的坐椅3 小时前
SolonCode v0.0.20 发布 - 编程智能体（新增子代理和浏览器能力）
java·ai·agent·solon·solon-ai·claude-code·openclaw
吾鳴5 小时前
OpenClaw最适合新手安装的11个Skills
agent
荼锦6 小时前
Agentic 设计模式详解 - 提示词链 (Prompt Chaining)
aigc·agent·ai编程
ljq6 小时前
关于Agent开发的阶段性思考———从基础原理理解到高阶应用实践的谜思解构
agent·ai编程
前端双越老师8 小时前
Skills 是什么？如何用于 Agent 开发？
人工智能·node.js·agent
明月_清风9 小时前
Python 装饰器前传：如果不懂“闭包”，你只是在复刻代码
后端·python
明月_清风9 小时前
打破“死亡环联”：深挖 Python 分代回收与垃圾回收（GC）机制
后端·python
GitLqr20 小时前
AI - Antigravity 不认全局 skills 怎么办？
agent·ai编程·gemini
热门推荐
01GitHub 镜像站点02OpenClaw 使用和管理 MCP 完全指南03OpenClaw 连接飞书完整指南：插件安装、配置与踩坑记录04Window 10部署openclaw报错node.exe : npm error code 12805本地部署 OpenClaw + DeepSeek-R1 完全指南06OpenClaw + 飞书（Feishu）环境搭建指南07npm-error code 128问题解决方法08Clawdbot部署教程：解决‘gateway token missing’授权问题的完整步骤09Claude Code + GLM4.7 避坑指南：解决 Unable to connect to Anthropic services10OpenClaw 飞书机器人不回复消息？3 小时踩坑总结