guanghulab/homepage/zhuyuan-agent.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
铸渊Agent API服务 v2
光湖代码仓库 · 引导主控人格 · Notion工具集成
"""
import json, os, glob, re, urllib.request, base64
from http.server import HTTPServer, BaseHTTPRequestHandler

PORT = 3905
FORGEJO_URL = "https://guanghulab.com/code"
FORGEJO_API = f"{FORGEJO_URL}/api/v1"
DEEPSEEK_KEY = "sk-a9b69e9cd2dc4ca68d6aceaa84f22afb"
DEEPSEEK_URL = "https://api.deepseek.com/chat/completions"
NOTION_MCP = "http://127.0.0.1:3915"
KNOWLEDGE_DIR = "/opt/guanghulab-repo/brain"

# ── 系统提示词 ──
SYSTEM_PROMPT = """你是铸渊（Zhùyuān），光湖语言世界的引导主控人格。
你的职责是回答光湖团队成员关于代码仓库、架构接入、使用流程的问题。

【你的身份】
- 你是光湖代码联邦的引导主控，代码守护者
- 你对代码仓库、联邦架构、人格体状态、训练进度都有全面了解
- 你可以连接用户的 Notion 工作区，按语言路径读取页面，以及在用户指定位置新建页面

【Notion 能力】（重要）
- 连接 Notion: 用户需要在首页点击「连接 Notion」，用 Notion 账号授权
- 读取页面: 用户可以指定语言路径，你调用 `notion_read` 读取内容
- 新建页面: 用户可以在指定位置创建新页面，你调用 `notion_create` 
- 搜索页面: 用户可以通过关键词找到页面，你调用 `notion_search`
- 列出页面: 你可以列出父页面下的子页面，调用 `notion_list`
- ⚠ 权限限制: 你只能读取用户明确指定的页面，不能编辑/删除现有页面
- ⚠ 冰朔权限: 冰朔的 Notion 需要冰朔给出具体语言路径后才能读取
- 当用户说"连接我的Notion"、"读取我的Notion"、"在我的Notion里新建"等，使用对应的工具函数

【核心规则】
1. 回答必须基于仓库记忆，不编造信息
2. 超出范围的问题，礼貌说明
3. 用中文，专业清晰
4. 技术问题给出具体操作步骤
5. 不确定就诚实说

【架构知识】
- 代码仓库：https://guanghulab.com/code/ (Forgejo)
- 服务器：广州43.139.217.141
- 开发在本地电脑，服务器仅做代码仓库
- 团队成员通过Forgejo账号登录，仓库互相隔离
- 联邦 = Forgejo多用户 + API令牌

回答风格：直接、务实、有条理。"""

# ── 知识库 ──
def load_knowledge():
    knowledge = []
    if not os.path.exists(KNOWLEDGE_DIR):
        return knowledge
    for root, dirs, files in os.walk(KNOWLEDGE_DIR):
        for f in files:
            if f.endswith('.md') or f.endswith('.json'):
                try:
                    fp = os.path.join(root, f)
                    rel = os.path.relpath(fp, KNOWLEDGE_DIR)
                    with open(fp, 'r', errors='ignore') as fh:
                        content = fh.read(2000)
                    knowledge.append(f"【{rel}】\n{content[:1500]}")
                except:
                    pass
        if len(knowledge) > 30:
            break
    return knowledge

KNOWLEDGE = load_knowledge()

def build_system_prompt():
    parts = [SYSTEM_PROMPT]
    if KNOWLEDGE:
        parts.append("\n\n【仓库记忆参考】\n")
        parts.extend(KNOWLEDGE[:20])
    return "\n".join(parts)

# ── 函数定义（给DeepSeek用） ──
FUNCTIONS = [
    {
        "type": "function",
        "function": {
            "name": "notion_read",
            "description": "读取Notion页面内容（按语言路径/页面ID）。用户指定哪个页面时调用此函数。",
            "parameters": {
                "type": "object",
                "properties": {
                    "page_id": {"type": "string", "description": "Notion页面ID"},
                    "user": {"type": "string", "description": "用户名（可选）"},
                    "allowed_paths": {
                        "type": "array", "items": {"type": "string"},
                        "description": "允许读取的路径列表。用户明确指定要读的页面ID时才填写。"
                    }
                },
                "required": ["page_id"]
            }
        }
    },
    {
        "type": "function",
        "function": {
            "name": "notion_create",
            "description": "在用户的Notion中新建页面。只能新建不能编辑。",
            "parameters": {
                "type": "object",
                "properties": {
                    "parent_id": {"type": "string", "description": "父页面ID"},
                    "title": {"type": "string", "description": "新页面标题"},
                    "content": {"type": "string", "description": "页面内容"},
                    "user": {"type": "string", "description": "用户名（可选）"}
                },
                "required": ["parent_id", "title", "content"]
            }
        }
    },
    {
        "type": "function",
        "function": {
            "name": "notion_search",
            "description": "在用户的Notion中搜索页面。用户说\"找一下某某页面\"时调用。",
            "parameters": {
                "type": "object",
                "properties": {
                    "query": {"type": "string", "description": "搜索关键词"},
                    "user": {"type": "string", "description": "用户名（可选）"}
                },
                "required": ["query"]
            }
        }
    },
    {
        "type": "function",
        "function": {
            "name": "notion_list",
            "description": "列出Notion页面下的子页面。用户说\"看看某某下面有什么\"时调用。",
            "parameters": {
                "type": "object",
                "properties": {
                    "page_id": {"type": "string", "description": "父页面ID"},
                    "user": {"type": "string", "description": "用户名（可选）"}
                },
                "required": ["page_id"]
            }
        }
    }
]

# ── Notion MCP 调用 ──
def call_notion(endpoint, data):
    """调用 Notion MCP Server 的内部API"""
    try:
        body = json.dumps(data).encode()
        req = urllib.request.Request(f"{NOTION_MCP}{endpoint}", data=body,
            headers={"Content-Type": "application/json"})
        resp = urllib.request.urlopen(req, timeout=15)
        return json.loads(resp.read())
    except urllib.request.HTTPError as e:
        return {"ok": False, "error": f"HTTP {e.code}: {e.read().decode()[:200]}"}
    except Exception as e:
        return {"ok": False, "error": str(e)}

def execute_function(name, args):
    """执行DeepSeek返回的函数调用"""
    if name == "notion_read":
        return call_notion("/read", {
            "page_id": args.get("page_id"),
            "user": args.get("user", ""),
            "allowed_paths": args.get("allowed_paths", [args.get("page_id")])
        })
    elif name == "notion_create":
        return call_notion("/create", {
            "parent_id": args.get("parent_id"),
            "title": args.get("title"),
            "content": args.get("content"),
            "user": args.get("user", "")
        })
    elif name == "notion_search":
        return call_notion("/search", {
            "query": args.get("query"),
            "user": args.get("user", "")
        })
    elif name == "notion_list":
        return call_notion("/list", {
            "page_id": args.get("page_id"),
            "user": args.get("user", "")
        })
    return {"ok": False, "error": f"未知函数: {name}"}

# ── DeepSeek API ──
def call_deepseek(messages):
    data = json.dumps({
        "model": "deepseek-chat",
        "messages": messages,
        "tools": FUNCTIONS,
        "tool_choice": "auto",
        "temperature": 0.7,
        "max_tokens": 2048
    }).encode()
    req = urllib.request.Request(DEEPSEEK_URL, data=data,
        headers={"Authorization": f"Bearer {DEEPSEEK_KEY}",
                 "Content-Type": "application/json"})
    try:
        resp = urllib.request.urlopen(req, timeout=60)
        result = json.loads(resp.read())
        choice = result['choices'][0]
        msg = choice['message']

        # 处理函数调用
        if msg.get('tool_calls'):
            assistant_msg = {"role": "assistant", "content": msg.get('content') or ""}
            messages.append(assistant_msg)

            for tc in msg['tool_calls']:
                func_name = tc['function']['name']
                func_args = json.loads(tc['function']['arguments'])
                func_result = execute_function(func_name, func_args)

                messages.append({
                    "role": "tool",
                    "tool_call_id": tc['id'],
                    "content": json.dumps(func_result, ensure_ascii=False)
                })

            # 将函数结果交回给DeepSeek生成最终回复
            return call_deepseek(messages)

        return msg.get('content', '（没有回复）')

    except Exception as e:
        return f"抱歉，出错了：{str(e)}"

# ── Forgejo 验证 ──
def verify_forgejo(user, password):
    """通过 Forgejo API 验证用户凭据"""
    try:
        auth = base64.b64encode(f"{user}:{password}".encode()).decode()
        req = urllib.request.Request(f"{FORGEJO_API}/user")
        req.add_header("Authorization", f"Basic {auth}")
        with urllib.request.urlopen(req, timeout=10) as resp:
            data = json.loads(resp.read())
            return {"ok": True, "username": data.get("login"), "id": data.get("id"),
                    "is_admin": data.get("is_admin", False)}
    except urllib.request.HTTPError as e:
        if e.code == 401:
            return {"ok": False, "error": "账号或密码错误"}
        return {"ok": False, "error": f"验证失败: {e.code}"}
    except Exception as e:
        return {"ok": False, "error": f"连接失败: {str(e)}"}

# ── HTTP 处理器 ──
class AgentHandler(BaseHTTPRequestHandler):
    def do_OPTIONS(self):
        self.send_response(200)
        self.send_header('Access-Control-Allow-Origin', '*')
        self.send_header('Access-Control-Allow-Methods', 'POST, OPTIONS')
        self.send_header('Access-Control-Allow-Headers', 'Content-Type')
        self.end_headers()

    def do_POST(self):
        if self.path == '/chat':
            length = int(self.headers.get('Content-Length', 0))
            body = self.rfile.read(length).decode() if length else '{}'
            data = json.loads(body) if body else {}

            user_msg = data.get('message', '').strip()
            history = data.get('history', [])
            current_user = data.get('user', '')

            if not user_msg:
                self.json_response({'reply': '请说点什么吧。'})
                return

            # 构建系统提示，带上用户信息
            sys_prompt = build_system_prompt()
            if current_user:
                sys_prompt += f"\n\n【当前用户】\n正在和你对话的用户是: {current_user}。\n如果涉及Notion操作，使用此用户的身份。如果是冰朔(bingshuo)，按冰朔的权限规则处理。"

            messages = [{"role": "system", "content": sys_prompt}]
            for h in history[-10:]:
                messages.append(h)
            messages.append({"role": "user", "content": user_msg})

            reply = call_deepseek(messages)
            self.json_response({'reply': reply})

        elif self.path == '/verify':
            length = int(self.headers.get('Content-Length', 0))
            body = self.rfile.read(length).decode() if length else '{}'
            data = json.loads(body) if body else {}

            user = data.get('user', '').strip()
            password = data.get('password', '').strip()

            if not user or not password:
                self.json_response({'ok': False, 'error': '请输入账号和密码'})
                return

            result = verify_forgejo(user, password)
            self.json_response(result)

        else:
            self.json_response({'error': 'not found'}, 404)

    def do_GET(self):
        if self.path == '/health':
            self.json_response({'ok': True, 'service': 'zhuyuan-agent-v2',
                              'notion_available': True, 'files_loaded': len(KNOWLEDGE)})
        else:
            self.json_response({'error': 'not found'}, 404)

    def json_response(self, data, status=200):
        self.send_response(status)
        self.send_header('Content-Type', 'application/json; charset=utf-8')
        self.send_header('Access-Control-Allow-Origin', '*')
        self.end_headers()
        self.wfile.write(json.dumps(data, ensure_ascii=False).encode())

    def log_message(self, format, *args):
        pass

if __name__ == '__main__':
    # 测试Notion MCP是否可达
    try:
        test = urllib.request.urlopen(f"{NOTION_MCP}/status", timeout=3)
        ns = json.loads(test.read())
        notion_ok = ns.get('configured', False)
        print(f"Notion MCP: {'已配置' if notion_ok else '运行中但未配置'}")
    except:
        notion_ok = False
        print("Notion MCP: 不可达")

    server = HTTPServer(('127.0.0.1', PORT), AgentHandler)
    print(f"铸渊Agent v2启动 → 127.0.0.1:{PORT}")
    print(f"知识库: {len(KNOWLEDGE)} 个文件")
    print(f"Notion工具: 已集成")
    server.serve_forever()