deepsearch-service

Lin-A1's avatarfrom Lin-A1

深度搜索服务，基于LLM的迭代式搜索-推理服务，能分解复杂问题并生成综合报告

⭐1stars🔀0forks📁View on GitHub🕐Updated Jan 8, 2026
[Web Research]

When & Why to Use This Skill

The Deep Search Service is an advanced LLM-powered agent skill designed for iterative search and complex reasoning. Unlike standard search tools, it automatically decomposes intricate queries into multiple sub-tasks, performs multi-round information retrieval, and evaluates data quality in real-time. It synthesizes information from diverse sources to generate comprehensive, structured analysis reports, making it an essential tool for deep-dive investigations where accuracy and depth are paramount.

Use Cases

  • Comprehensive Market Analysis: Automatically gathering and synthesizing market trends, competitor strategies, and consumer insights from various web sources to create a strategic business report.
  • Technical Research & Architecture Design: Investigating emerging technologies or complex software patterns by cross-referencing official documentation, expert blogs, and community discussions to provide a verified best-practice guide.
  • Fact-Checking and Evidence Verification: Validating complex claims or news by iteratively searching for supporting or contradicting evidence across high-authority domains to ensure information integrity.
  • Academic Literature Review: Streamlining the process of identifying, summarizing, and connecting scholarly articles and research papers on specific scientific or social topics.
namedeepsearch-service
description基于 LLM 的深度迭代搜索与推理服务。擅长处理复杂问题，通过自动分解查询、多轮迭代检索、信息评估与验证，最终生成全面且结构化的深度分析报告。

功能

基于 LLM 的迭代式深度搜索服务，能够：

  1. 将复杂问题分解为多个子查询
  2. 迭代搜索收集多源信息
  3. 评估信息充分性，动态调整搜索策略
  4. 综合生成结构化分析报告

适用场景

  • 复杂问题需要多角度分析
  • 需要综合多个来源的信息
  • 要求生成完整的研究报告

调用方式

from services.deepsearch_service.client import DeepSearchClient

client = DeepSearchClient()

# 健康检查
status = client.health_check()

# 深度搜索（默认参数）
result = client.search("Python异步编程的最佳实践有哪些？")

# 自定义参数
result = client.search(
    query="如何设计一个高可用的微服务架构？",
    max_iterations=3,          # 最大迭代次数 (1-5)
    queries_per_iteration=3,   # 每轮查询数 (1-5)
    depth_level="deep"         # 搜索深度: quick/normal/deep
)

# 获取报告和来源
print(result["report"])
for source in result["sources"]:
    print(f"- {source['title']}: {source['url']}")

返回格式

{
  "query": "Python异步编程的最佳实践有哪些？",
  "report": "# Python异步编程最佳实践\n\n## 1. 核心概念...",
  "sources": [
    {
      "title": "Python官方asyncio文档",
      "url": "https://docs.python.org/...",
      "relevance": 0.95,
      "snippet": "asyncio是Python标准库中的异步I/O框架..."
    }
  ],
  "iterations": [
    {
      "iteration": 1,
      "queries": ["Python asyncio 教程", "async await 用法"],
      "results_count": 6,
      "key_findings": ["asyncio是标准库组件"]
    }
  ],
  "total_iterations": 2,
  "total_sources": 8,
  "elapsed_seconds": 45.32,
  "search_timestamp": "2025-12-28T18:30:00"
}

与 websearch_service 的区别

特性 WebSearch DeepSearch
搜索模式 单轮搜索 迭代多轮搜索
查询策略 直接使用输入 LLM 分解/优化
输出格式 结构化列表 综合报告 + 来源
响应时间 10-30s 30-120s