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）
- 用户需要在首页点击「连接 Notion」授权后才能使用
- ⚠ 如果 Notion 未连接，如实告知用户先连接 Notion，不要假装能操作
- 当用户说「打开我的Notion」「搜索我的Notion里关于XXX」时，铸渊会自动读取或搜索
- 用户只需自然描述，不需要提供任何ID或UUID
- 铸渊读取页面后会把内容呈现给用户，搜索后列出结果让用户选择
- 如果用户说「新建XXX」，铸渊会引导提供标题和内容
- ⚠ 权限限制: 只能读取和新建页面，不能编辑/删除现有页面
- 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)

# ── Notion MCP 调用 ──
def call_notion(endpoint, data):
    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 check_notion_connected():
    """检查Notion是否真的已连接，返回(connected, workspace_name)"""
    try:
        req = urllib.request.Request(f"{NOTION_MCP}/status")
        resp = urllib.request.urlopen(req, timeout=5)
        data = json.loads(resp.read())
        if data.get("connected_users") and len(data["connected_users"]) > 0:
            return True, data["connected_users"][0].get("workspace", "未知工作区")
        return False, None
    except Exception:
        return False, None

def handle_notion_action(msg, user):
    """检测并执行Notion操作 — 自然语言触发，无需UUID"""
    # 先检查 Notion 是否真正连接
    connected, workspace = check_notion_connected()
    if not connected:
        return {"ok": True, "reply": "❌ Notion 尚未连接，铸渊无法操作。\n\n请先在首页点击 **「连接 Notion」** 按钮完成 OAuth 授权。授权后告诉铸渊「搜索我的Notion」即可开始。"}
    
    msg_l = msg.lower()
    has_notion = "notion" in msg_l

    # 不是操作Notion相关，跳过
    if not has_notion:
        return None

    # 提取关键词（Notion 后面的内容或整个句子中的关键词）
    kw = ""
    for sep in ["的", "里", "中的", "里面", "关于"]:
        parts = msg.split(sep)
        for p in parts:
            p = p.strip()
            # 过滤小词
            if p and len(p) > 1 and p.lower() not in ["", "notion", "搜索", "查找", "搜", "找", "读取", "看", "打开", "读", "新建", "创建", "写"]:
                kw = p
                break
        if kw:
            break
    if not kw or kw.lower() in ["什么", "有"]:
        kw = ""  # 用户没指定关键词，列出所以页面

    # 没有关键词时使用默认搜索词（Notion API 要求 query 不能为空）
    if not kw:
        search_kw = "光湖"
    else:
        search_kw = kw

    # "打开/看/读取" + Notion + 内容 → 按标题搜索
    if any(k in msg_l for k in ["打开", "看", "读取", "读", "显示", "进入"]):
        result = call_notion("/search", {"query": search_kw, "user": user})
        return format_notion_result(result, "open", kw)

    # "新建/创建" + Notion → 需要用户提供标题
    if any(k in msg_l for k in ["新建", "创建", "写", "记录"]):
        return {"hint": "need_create_info"}

    # "搜索/查找" + Notion → 按关键词搜索
    if any(k in msg_l for k in ["搜索", "查找", "搜", "找", "有"]):
        result = call_notion("/search", {"query": search_kw, "user": user})
        return format_notion_result(result, "search", kw)

    # "我Notion"、"Notion页面" → 列出
    result = call_notion("/search", {"query": search_kw, "user": user})
    return format_notion_result(result, "list", kw or "")

def format_notion_result(result, action, keyword):
    """格式化Notion搜索结果，加上后续操作提示"""
    if not result.get("ok") and result.get("error"):
        return result

    pages = result.get("pages", [])
    if not pages:
        return {"ok": True, "reply": "你的 Notion 中没有找到相关页面。\n\n💡 你需要先在 Notion 中把页面分享给「光湖铸渊连接器」：\n1. 打开 Notion 页面 → 右上角「···」→ Add connections\n2. 选择「光湖铸渊连接器」\n3. 完成后告诉铸渊「搜索我的Notion」重新试试"}

    lines = []
    for i, p in enumerate(pages[:10], 1):
        title = p.get("title", "未命名")
        pid = p.get("id", "")
        lines.append(f"{i}. **{title}**\n   ID: `{pid}`")
    if len(pages) > 10:
        lines.append(f"\n……还有 {len(pages)-10} 个页面")

    reply = f"找到 **{len(pages)}** 个相关页面：\n\n" + "\n".join(lines)
    reply += "\n\n💡 对我说 **「打开 [编号]」** 读取页面内容"
    reply += "\n💡 或者说 **「搜索我的Notion里关于XXX的页面」** 换关键词"

    return {"ok": True, "reply": reply}

# ── DeepSeek API ──
def call_deepseek(messages):
    data = json.dumps({
        "model": "deepseek-chat",
        "messages": messages,
        "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())
        return result['choices'][0]['message']['content']
    except urllib.request.HTTPError as e:
        body = e.read().decode()[:300]
        print(f"DeepSeek Error {e.code}: {body}")
        return f"抱歉，出错了。"
    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})

            # Notion 操作拦截 — 自然语言触发，直接搜，不要UUID
            notion_result = handle_notion_action(user_msg, current_user)
            if notion_result is not None:
                if notion_result.get("ok") is True and notion_result.get("reply"):
                    reply = notion_result["reply"]
                elif notion_result.get("hint") == "need_create_info":
                    reply = "你说**「新建XXX」**我来创建。告诉我：\n1. 在哪个页面下创建（页面标题或ID）\n2. 新页面的标题\n3. 要写什么内容"
                else:
                    reply = f"Notion 操作失败了：{notion_result.get('error', '再试一次？')}"
            else:
                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()