coding-agent/s07_skill_loading/code.py

#!/usr/bin/env python3
"""
s07: 技能加载 — 两级按需知识注入。

  第 1 层（便宜，始终存在）：
    SYSTEM 提示词包含技能名称 + 单行描述（每个技能约 100 tokens）
    "可用技能： agent-builder, code-review, mcp-builder, pdf"

  第 2 层（昂贵，按需加载）：
    Agent 调用 load_skill("code-review") → 完整 SKILL.md 内容
    通过 tool_result 注入（每个技能约 2000 tokens）

  skills/
    agent-builder/SKILL.md
    code-review/SKILL.md
    mcp-builder/SKILL.md
    pdf/SKILL.md

相对 s06 的变化：
  + build_system() — 启动时扫描 skills/ 目录，把目录注入 SYSTEM
  + load_skill(name) — 通过 tool_result 返回完整 SKILL.md 内容
  + SKILLS_DIR 配置
  循环不变：load_skill 通过 TOOL_HANDLERS 自动分发。

运行: python s07_skill_loading/code.py
需要: pip install anthropic python-dotenv pyyaml + .env 中配置 ANTHROPIC_API_KEY
"""

import ast, json, os, subprocess
from pathlib import Path
import yaml

try:
    import readline
    readline.parse_and_bind('set bind-tty-special-chars off')
except ImportError:
    pass

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()
SKILLS_DIR = WORKDIR / "skills"
client = Anthropic(base_url=os.getenv("ANTHROPIC_BASE_URL"))
MODEL = os.environ["MODEL_ID"]
CURRENT_TODOS: list[dict] = []

# s07: 技能目录扫描 (供下方 build_system 使用)
def _parse_frontmatter(text: str) -> tuple[dict, str]:
    """解析 SKILL.md 的 YAML frontmatter，返回 (meta, body)。"""
    if not text.startswith("---"):
        return {}, text
    parts = text.split("---", 2)
    if len(parts) < 3:
        return {}, text
    try:
        meta = yaml.safe_load(parts[1]) or {}
    except yaml.YAMLError:
        meta = {}
    return meta, parts[2].strip()

# 启动时构建技能注册表 (用于在 load_skill 中安全查找)
SKILL_REGISTRY: dict[str, dict] = {}

def _scan_skills():
    """扫描 skills/ 目录，把名称、描述和内容写入 SKILL_REGISTRY。"""
    if not SKILLS_DIR.exists():
        return
    for d in sorted(SKILLS_DIR.iterdir()):
        if not d.is_dir():
            continue
        manifest = d / "SKILL.md"
        if manifest.exists():
            raw = manifest.read_text()
            meta, body = _parse_frontmatter(raw)
            name = meta.get("name", d.name)
            desc = meta.get("description", raw.split("\n")[0].lstrip("#").strip())
            SKILL_REGISTRY[name] = {"name": name, "description": desc, "content": raw}

_scan_skills()

def list_skills() -> str:
    """列出所有技能：名称 + 单行描述。"""
    if not SKILL_REGISTRY:
        return "（未找到技能）"
    return "\n".join(f"- **{s['name']}**: {s['description']}" for s in SKILL_REGISTRY.values())

# s07: SYSTEM 包含技能目录 (成本低 — 只有名称和描述)
def build_system() -> str:
    """构建 SYSTEM 提示词，并注入启动时扫描到的技能目录。"""
    catalog = list_skills()
    return (
        f"你是位于 {WORKDIR}. "
        f"可用技能：\n{catalog}\n"
        "需要时使用 load_skill 获取完整详情。"
    )

SYSTEM = build_system()

# s07: 子 Agent 使用自己的系统提示词 — 不加载技能，也没有 task
SUB_SYSTEM = (
    f"你是位于 {WORKDIR}. "
    "完成交给你的任务，然后返回简洁摘要。"
    "不要继续委派。"
)


# ═══════════════════════════════════════════════════════════
#  来自 s02-s06 （未改动）: 工具实现
# ═══════════════════════════════════════════════════════════

def safe_path(p: str) -> Path:
    path = (WORKDIR / p).resolve()
    if not path.is_relative_to(WORKDIR):
        raise ValueError(f"路径逃逸出工作区：{p}")
    return path

def run_bash(command: str) -> str:
    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 "（无输出）"
    except subprocess.TimeoutExpired:
        return "错误：执行超时（120 秒）"

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

def run_write(path: str, content: str) -> str:
    try:
        file_path = safe_path(path)
        file_path.parent.mkdir(parents=True, exist_ok=True)
        file_path.write_text(content)
        return f"已写入 {len(content)} 字节到 {path}"
    except Exception as e:
        return f"错误：{e}"

def run_edit(path: str, old_text: str, new_text: str) -> str:
    try:
        file_path = safe_path(path)
        text = file_path.read_text()
        if old_text not in text:
            return f"错误：在文件中未找到目标文本：{path}"
        file_path.write_text(text.replace(old_text, new_text, 1))
        return f"已编辑 {path}"
    except Exception as e:
        return f"错误：{e}"

def run_glob(pattern: str) -> str:
    import glob as g
    try:
        results = []
        for match in g.glob(pattern, root_dir=WORKDIR):
            if (WORKDIR / match).resolve().is_relative_to(WORKDIR):
                results.append(match)
        return "\n".join(results) if results else "（无匹配）"
    except Exception as e:
        return f"错误：{e}"

def _normalize_todos(todos):
    if isinstance(todos, str):
        try:
            todo_list = json.loads(todos)
        except json.JSONDecodeError:
            try:
                todo_list = ast.literal_eval(todos)
            except (SyntaxError, ValueError):
                return None, "错误：todos 必须是列表或 JSON 数组字符串"
    if not isinstance(todos, list):
        return None, "错误：todos 必须是列表"
    for i, t in enumerate(todos):
        if not isinstance(t, dict):
            return None, f"错误：todos[{i}] 必须是对象"
        if "content" not in t or "status" not in t:
            return None, f"错误：todos[{i}] 缺少 'content' 或 'status'"
        if t["status"] not in ("pending", "in_progress", "completed"):
            return None, f"错误：todos[{i}] 包含无效状态 '{t['status']}'"
    return todo_list, None

def run_todo_write(todos: list) -> str:
    global CURRENT_TODOS
    todo_list, error = _normalize_todos(todos)
    if error:
        return error
    CURRENT_TODOS = todo_list
    lines = ["\n\033[33m## 当前任务\033[0m"]
    for t in CURRENT_TODOS:
        icon = {"pending": " ", "in_progress": "\033[36m▸\033[0m", "completed": "\033[32m✓\033[0m"}[t["status"]]
        lines.append(f"  [{icon}] {t['content']}")
    print("\n".join(lines))
    return f"已更新 {len(CURRENT_TODOS)} 个任务"

def extract_text(content) -> str:
    if not isinstance(content, list):
        return str(content)
    return "\n".join(getattr(b, "text", "") for b in content if getattr(b, "type", None) == "text")


# ═══════════════════════════════════════════════════════════
#  来自 s06 （未改动）: 子 Agent
# ═══════════════════════════════════════════════════════════

SUB_TOOLS = [
    {"name": "bash", "description": "运行一条 shell 命令。",
     "input_schema": {"type": "object", "properties": {"command": {"type": "string"}}, "required": ["command"]}},
    {"name": "read_file", "description": "读取文件内容。",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}}, "required": ["path"]}},
    {"name": "write_file", "description": "向文件写入内容。",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "content": {"type": "string"}}, "required": ["path", "content"]}},
    {"name": "edit_file", "description": "在文件中替换一次完全匹配的文本。",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "old_text": {"type": "string"}, "new_text": {"type": "string"}}, "required": ["path", "old_text", "new_text"]}},
    {"name": "glob", "description": "查找匹配 glob 模式的文件。",
     "input_schema": {"type": "object", "properties": {"pattern": {"type": "string"}}, "required": ["pattern"]}},
]
SUB_HANDLERS = {"bash": run_bash, "read_file": run_read, "write_file": run_write,
                "edit_file": run_edit, "glob": run_glob}

def spawn_subagent(description: str) -> str:
    print(f"\n\033[35m[子 Agent 已启动]\033[0m")
    messages = [{"role": "user", "content": description}]
    for _ in range(30):
        response = client.messages.create(model=MODEL, system=SUB_SYSTEM,
            messages=messages, tools=SUB_TOOLS, max_tokens=8000)
        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":
                blocked = trigger_hooks("PreToolUse", block)
                if blocked:
                    results.append({"type": "tool_result", "tool_use_id": block.id,
                                    "content": str(blocked)})
                    continue
                handler = SUB_HANDLERS.get(block.name)
                output = handler(**block.input) if handler else f"未知工具：{block.name}"
                trigger_hooks("PostToolUse", block, output)
                print(f"  \033[90m[sub] {block.name}: {str(output)[:100]}\033[0m")
                results.append({"type": "tool_result", "tool_use_id": block.id, "content": output})
        messages.append({"role": "user", "content": results})
    result = extract_text(messages[-1]["content"])
    if not result:
        for msg in reversed(messages):
            if msg["role"] == "assistant":
                result = extract_text(msg["content"])
                if result:
                    break
        if not result:
            result = "子 Agent 已等待 30 轮仍未给出最终回答，已停止。"
    print(f"\033[35m[子 Agent 已完成]\033[0m")
    return result


# ═══════════════════════════════════════════════════════════
#  新增于 s07: load_skill — 运行时加载完整内容
# ═══════════════════════════════════════════════════════════

def load_skill(name: str) -> str:
    """通过注册表加载完整技能内容，避免路径穿越。"""
    skill = SKILL_REGISTRY.get(name)
    if not skill:
        return f"未找到技能：{name}"
    return skill["content"]


# ═══════════════════════════════════════════════════════════
#  工具注册表 — 来自 s02-s07 的全部工具，怎么省token 怎么降低模型幻觉
# ═══════════════════════════════════════════════════════════

TOOLS = [
    {"name": "bash", "description": "运行一条 shell 命令。",
     "input_schema": {"type": "object", "properties": {"command": {"type": "string"}}, "required": ["command"]}},
    {"name": "read_file", "description": "读取文件内容。",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "limit": {"type": "integer"}}, "required": ["path"]}},
    {"name": "write_file", "description": "向文件写入内容。",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "content": {"type": "string"}}, "required": ["path", "content"]}},
    {"name": "edit_file", "description": "在文件中替换一次完全匹配的文本。",
     "input_schema": {"type": "object", "properties": {"path": {"type": "string"}, "old_text": {"type": "string"}, "new_text": {"type": "string"}}, "required": ["path", "old_text", "new_text"]}},
    {"name": "glob", "description": "查找匹配 glob 模式的文件。",
     "input_schema": {"type": "object", "properties": {"pattern": {"type": "string"}}, "required": ["pattern"]}},
    {"name": "todo_write", "description": "为当前编码会话创建并维护任务清单。",
     "input_schema": {"type": "object", "properties": {"todos": {"type": "array", "items": {"type": "object", "properties": {"content": {"type": "string"}, "status": {"type": "string", "enum": ["pending", "in_progress", "completed"]}}, "required": ["content", "status"]}}}, "required": ["todos"]}},
    {"name": "task", "description": "启动一个子 Agent 处理复杂子任务。只返回最终结论。",
     "input_schema": {"type": "object", "properties": {"description": {"type": "string"}}, "required": ["description"]}},
    # s07: 技能工具 (目录已在 SYSTEM 提示词中，此处加载完整内容)
    {"name": "load_skill", "description": "按名称加载某个技能的完整内容。",
     "input_schema": {"type": "object", "properties": {"name": {"type": "string"}}, "required": ["name"]}},
]

TOOL_HANDLERS = {
    "bash": run_bash, "read_file": run_read, "write_file": run_write,
    "edit_file": run_edit, "glob": run_glob, "todo_write": run_todo_write,
    "task": spawn_subagent, "load_skill": load_skill,
}


# ═══════════════════════════════════════════════════════════
#  来自 s04 （未改动）: Hook 系统
# ═══════════════════════════════════════════════════════════

HOOKS = {"UserPromptSubmit": [], "PreToolUse": [], "PostToolUse": [], "Stop": []}

def register_hook(event: str, callback):
    HOOKS[event].append(callback)

def trigger_hooks(event: str, *args):
    for callback in HOOKS[event]:
        result = callback(*args)
        if result is not None:
            return result
    return None

DENY_LIST = ["rm -rf /", "sudo", "shutdown", "reboot", "mkfs", "dd if="]

def permission_hook(block):
    if block.name == "bash":
        for p in DENY_LIST:
            if p in block.input.get("command", ""):
                print(f"\n\033[31m⛔ 已拦截：'{p}'\033[0m")
                return "权限被拒绝"
    return None

def log_hook(block):
    print(f"\033[90m[钩子] {block.name}\033[0m")
    return None

def context_inject_hook(query: str):
    print(f"\033[90m[钩子] UserPromptSubmit: 工作目录：{WORKDIR}\033[0m")
    return None

def summary_hook(messages: list):
    tool_count = sum(1 for m in messages
                     for b in (m.get("content") if isinstance(m.get("content"), list) else [])
                     if isinstance(b, dict) and b.get("type") == "tool_result")
    print(f"\033[90m[钩子] Stop：会话使用了 {tool_count} 次工具调用\033[0m")
    return None

register_hook("UserPromptSubmit", context_inject_hook)
register_hook("PreToolUse", permission_hook)
register_hook("PreToolUse", log_hook)
register_hook("Stop", summary_hook)


# ═══════════════════════════════════════════════════════════
#  agent_loop — 与以下相同： s05-s06 + 催办提醒
# ═══════════════════════════════════════════════════════════

def agent_loop(messages: list):
    rounds_since_todo = 0
    while True:
        if rounds_since_todo >= 3 and messages:
            messages.append({"role": "user",
                             "content": "<reminder>请更新你的待办事项。</reminder>"})
            rounds_since_todo = 0
            
        response = client.messages.create(
            model=MODEL, system=SYSTEM, messages=messages,
            tools=TOOLS, max_tokens=8000,
        )
        messages.append({"role": "assistant", "content": response.content})

        if response.stop_reason != "tool_use":
            force = trigger_hooks("Stop", messages)
            if force:
                messages.append({"role": "user", "content": force})
                continue
            return

        rounds_since_todo += 1
        results = []
        for block in response.content:
            if block.type != "tool_use":
                continue

            blocked = trigger_hooks("PreToolUse", block)
            if blocked:
                results.append({"type": "tool_result", "tool_use_id": block.id,
                                "content": str(blocked)})
                continue

            handler = TOOL_HANDLERS.get(block.name)
            output = handler(**block.input) if handler else f"未知工具：{block.name}"

            trigger_hooks("PostToolUse", block, output)

            if block.name == "todo_write":
                rounds_since_todo = 0

            results.append({"type": "tool_result", "tool_use_id": block.id,
                            "content": output})

        messages.append({"role": "user", "content": results})


if __name__ == "__main__":
    print("s07: 技能加载 — 目录进 SYSTEM，内容按需加载")
    print("输入问题后按回车。输入 q 退出。\n")

    history = []
    while True:
        try:
            query = input("\033[36ms07 >> \033[0m")
        except (EOFError, KeyboardInterrupt):
            break
        if query.strip().lower() in ("q", "exit", ""):
            break
        trigger_hooks("UserPromptSubmit", query)
        history.append({"role": "user", "content": query})
        agent_loop(history)
        for block in history[-1]["content"]:
            if getattr(block, "type", None) == "text":
                print(block.text)
        print()