coding-agent/s10_system_prompt/code.py

#!/usr/bin/env python3
"""
s10: 系统提示词 — 运行时组装提示词并缓存。

运行:  python s10_system_prompt/code.py
需要: pip install anthropic python-dotenv + .env 中配置 ANTHROPIC_API_KEY

相对 s09 的变化：
  - PROMPT_SECTIONS：按主题作为 key 的提示词片段字典
  - assemble_system_prompt(context)：根据真实状态选择并拼接片段
  - get_system_prompt(context)：通过 json.dumps 实现确定性的缓存
  - agent_loop 使用 get_system_prompt(context)，不再使用硬编码 SYSTEM

当 .memory/MEMORY.md 存在时加载记忆片段（基于真实状态，不靠关键词）。
"""

import os, subprocess, json
from pathlib import Path

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()
MEMORY_DIR = WORKDIR / ".memory"
MEMORY_INDEX = MEMORY_DIR / "MEMORY.md"
client = Anthropic(base_url=os.getenv("ANTHROPIC_BASE_URL"))
MODEL = os.environ["MODEL_ID"]


# ── 提示词片段 ──

PROMPT_SECTIONS = {
    "identity": "你是一个编码 Agent。直接行动，不要只解释。",
}


def assemble_system_prompt(context: dict) -> str:
    """根据当前上下文选择并拼接提示词片段。"""
    sections = []

    # 始终加载 — 身份
    sections.append(PROMPT_SECTIONS["identity"])

    # 动态加载 — 来自上下文的工具和工作区
    tools = ", ".join(context.get("enabled_tools", []))
    if tools:
        sections.append(f"可用工具：{tools}.")
    sections.append(f"工作目录：{context.get('workspace', WORKDIR)}")

    # 条件加载 — MEMORY.md 存在且有内容时加载记忆
    memories = context.get("memories", "")
    if memories:
        sections.append(f"相关记忆：\n{memories}")

    return "\n\n".join(sections)


_last_context_key = None
_last_prompt = None


def get_system_prompt(context: dict) -> str:
    """提示词缓存包装器：只有上下文变化时才重新组装。

    这里使用 json.dumps 做确定性序列化，不使用 Python 的 hash()，
    因为 hash() 有进程级随机化，而且嵌套 dict/list 也不适用。
    这个缓存只避免同一进程内重复拼接字符串。
    真实 Claude Code 还会通过稳定的片段顺序和
    SYSTEM_PROMPT_DYNAMIC_BOUNDARY 保护 API 级提示词缓存。
    """
    global _last_context_key, _last_prompt
    key = json.dumps(context, sort_keys=True, ensure_ascii=False, default=str)
    if key == _last_context_key and _last_prompt:
        print("  \033[90m[缓存命中] 系统提示词未变化\033[0m")
        return _last_prompt
    _last_context_key = key
    _last_prompt = assemble_system_prompt(context)

    loaded = ["identity", "tools", "workspace"]
    if context.get("memories"):
        loaded.append("memory")
    print(f"  \033[32m[已组装] 片段：{', '.join(loaded)}\033[0m")
    return _last_prompt


# ── 工具 ──

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}"


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"]}},
]

TOOL_HANDLERS = {"bash": run_bash, "read_file": run_read, "write_file": run_write}


# ── 上下文 ──

def update_context(context: dict, messages: list) -> dict:
    """从真实状态推导上下文：有哪些工具、是否存在记忆文件。"""
    memories = ""
    if MEMORY_INDEX.exists():
        content = MEMORY_INDEX.read_text().strip()
        if content:
            memories = content
    return {
        "enabled_tools": list(TOOL_HANDLERS.keys()),
        "workspace": str(WORKDIR),
        "memories": memories,
    }


# ── Agent 循环 ──

def agent_loop(messages: list, context: dict):
    """主循环：使用组装后的系统提示词，而不是硬编码 SYSTEM。"""
    system = get_system_prompt(context)
    while True:
        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":
            return

        results = []
        for block in response.content:
            if block.type != "tool_use":
                continue
            print(f"\033[36m> {block.name}\033[0m")
            handler = TOOL_HANDLERS.get(block.name)
            output = handler(**block.input) if handler else f"未知工具：{block.name}"
            print(str(output)[:200])
            results.append({"type": "tool_result",
                            "tool_use_id": block.id, "content": output})
        messages.append({"role": "user", "content": results})

        # 每轮工具调用后重新评估上下文和提示词
        context = update_context(context, messages)
        system = get_system_prompt(context)


if __name__ == "__main__":
    print("s10: 系统提示词 — 运行时组装")
    print("输入问题后按回车发送。输入 q 退出。\n")
    history = []
    context = update_context({}, [])
    while True:
        try:
            query = input("\033[36ms10 >> \033[0m")
        except (EOFError, KeyboardInterrupt):
            break
        if query.strip().lower() in ("q", "exit", ""):
            break
        history.append({"role": "user", "content": query})
        agent_loop(history, context)
        context = update_context(context, history)
        for block in history[-1]["content"]:
            if getattr(block, "type", None) == "text":
                print(block.text)
        print()