#!/usr/bin/env python3
"""
Deep Research调试脚本
用于复现和调试语义分簇JSON解析错误
"""
import asyncio
import sys
import os
# 添加项目根目录到路径
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from agents.deep_research.director import ContextEngineeringDirector
from core.models import DeepResearchRequest
from loguru import logger
async def debug_deep_research():
"""调试Deep Research功能"""
try:
# 创建Director实例
director = ContextEngineeringDirector()
# 创建一个简单的测试请求
# 使用真实的页面ID
request = DeepResearchRequest(
page_id="22dccc690d8280e48e28cc45b8596548", # 从Notion链接提取的页面ID
purpose="调试测试语义分簇JSON解析",
max_pages=5,
research_complexity="standard"
)
logger.info("开始调试Deep Research...")
# 尝试执行研究流程
try:
result = await director.orchestrate_research(request)
logger.info("Deep Research调试完成")
print(f"结果类型: {type(result)}")
except Exception as e:
logger.error(f"Deep Research执行失败: {e}")
logger.exception("详细错误堆栈:")
except Exception as e:
logger.error(f"初始化失败: {e}")
logger.exception("详细错误堆栈:")
if __name__ == "__main__":
print("🔍 开始Deep Research调试...")
asyncio.run(debug_deep_research())
print("✅ 调试完成")
