mbs_watcher.py

#!/usr/bin/env python3
"""
MBS auto-watcher — 自动捕获 Hermes 会话记忆
被 cron 定时调用，检查新会话，提取关键信息存入 MBS
"""
import json
import os
import sqlite3
import subprocess
import time
from datetime import datetime, timedelta

SESSIONS_DIR = os.path.expanduser("~/.hermes/sessions")
MEMORY_DB = os.path.expanduser("~/.memory-bridge/memory.db")
STATE_FILE = os.path.expanduser("~/.memory-bridge/last_check.txt")

def get_latest_sessions(minutes_back=60):
    """获取最近N分钟内的会话文件"""
    now = time.time()
    cutoff = now - (minutes_back * 60)
    sessions = []
    if not os.path.exists(SESSIONS_DIR):
        return sessions
    for f in sorted(os.listdir(SESSIONS_DIR), reverse=True):
        fpath = os.path.join(SESSIONS_DIR, f)
        if not f.endswith(".json"):
            continue
        mtime = os.path.getmtime(fpath)
        if mtime < cutoff:
            break  # 文件按修改时间倒序，超时了就不用继续
        sessions.append(fpath)
    return sessions

def extract_text_from_session(path):
    """从会话JSON中提取用户和AI的对话文本"""
    try:
        with open(path) as f:
            data = json.load(f)
    except (json.JSONDecodeError, FileNotFoundError):
        return ""

    texts = []
    messages = data if isinstance(data, list) else data.get("messages", [])

    for msg in messages:
        role = msg.get("role", "")
        content = msg.get("content", "")
        if isinstance(content, list):
            content = " ".join(c.get("text", "") for c in content if isinstance(c, dict))
        if role in ("user", "assistant") and content:
            texts.append(f"{role}: {content[:500]}")
    return "\n".join(texts[-20:])  # 取最近20条消息

def already_processed(path):
    """检查会话是否已处理过"""
    if not os.path.exists(MEMORY_DB):
        return False
    try:
        db = sqlite3.connect(MEMORY_DB)
        name = os.path.basename(path)
        count = db.execute(
            "SELECT COUNT(*) FROM observations WHERE narrative LIKE ?",
            (f"%{name}%",)
        ).fetchone()[0]
        db.close()
        return count > 0
    except:
        return False

def main():
    sessions = get_latest_sessions(180)  # 查最近3小时
    new_count = 0

    for s in sessions[:5]:  # 最多处理5个
        if already_processed(s):
            continue
        text = extract_text_from_session(s)
        if len(text) < 50:
            continue

        # 通过管道传给 mbs auto
        result = subprocess.run(
            ["mbs", "auto"],
            input=text,
            capture_output=True,
            text=True,
            timeout=30
        )
        if result.returncode == 0 and "✔" in result.stdout:
            new_count += 1
            print(f"  ✔ {os.path.basename(s)}")

    if new_count == 0:
        print("无新会话需要处理")
    else:
        print(f"处理了 {new_count} 个新会话")

if __name__ == "__main__":
    main()