# 666ghj/DeepSearchAgent-Demo

从0实现一个简洁清晰的Deep Search Agent

Repository: https://github.com/666ghj/DeepSearchAgent-Demo
Canonical: https://ross.abutalabs.com/products/deepsearchagent-demo
Language: Python
License: MIT
License Family: permissive
Last push: 2025-08-19T12:37:50+00:00

## Health v2 (maintenance only)
Score: 34/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 37, release rhythm 35, longevity 27
- inputs: {"age_days": 381, "days_push": 379, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1223, forks 309 (observed 2026-08-28T04:04:02.675929+00:00)

## What it is
A framework-free Python implementation of a deep search AI agent that generates high-quality research reports through multi-round web searching and reflection. It integrates LLMs like DeepSeek and OpenAI with the Tavily search engine and includes a Streamlit web interface.

## Use cases
- build a deep research agent from scratch without langchain
- generate research reports automatically from a query
- implement multi-round search and reflection with an LLM
- compare deepseek and openai models for research tasks
- run a streamlit app for AI-powered web research
- learn how deep search agents work internally

## When to choose
- you want a minimal, dependency-light deep research agent you can read and modify
- you need multi-LLM support with Tavily web search
- you want a working Streamlit UI for research report generation
- you are learning agent architecture like reflection loops and state management

## When to avoid
- you need a production-grade, battle-tested deep research pipeline
- you require heavy framework integrations like LangChain or LangGraph
- you need search providers other than Tavily without writing adapters
- you need non-Python environments or offline research without an LLM API

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, rag, search-engine, llm-inference, web-scraping
- domain: artificial-intelligence, large-language-models
- platform: python, cross-platform
- tags: deep-search, deep-research, reflection-loop, framework-free, tavily, deepseek, streamlit, research-reports, multi-step-agent, ai-agents, retrieval-augmented-generation, search, natural-language-processing, web-server

## Member repositories
- 666ghj/DeepSearchAgent-Demo (main) score 34

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:02.675929+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:15:30.322396+00:00, confidence not recorded.
  - readme: https://github.com/666ghj/DeepSearchAgent-Demo (fetched 2026-08-28T04:04:02.675929+00:00, sha ef414e8ba58c)
- Data as of 2026-08-30T08:39:29.467469+00:00.