// Software stack
Deep research agent
Build your own web research assistant: self-hosted search, page fetching, Markdown conversion, and an agent framework that writes the report.
5 pieces4 setup stepsscenario: Connect tools to agents (MCP, CLI, auth)
What goes in it
- 1 ★ 30k
gpt-researcher Agent frameworks & SDKs Open source
Open-source deep research agent
An autonomous agent that conducts deep research on any data using any LLM providers
- 2 ★ 39k
searxng Document parsing & web scraping Open source
Self-hosted metasearch engine
SearXNG is a free internet metasearch engine which aggregates results from various search services and databases. Users are neither tracked nor profiled.
- 3 ★ 86k
crawl4ai Document parsing & web scraping Open source
LLM-friendly web crawler
Open-source web crawler and scraper for LLMs and AI agents: any website into clean, LLM-ready Markdown. Run it yourself, or use Crawl4AI Cloud with one key.
- 4 ★ 191k
MarkItDown Document parsing & web scraping Open source
Converts files to Markdown
Microsoft tool converting office documents to Markdown for LLMs.
- 5 ★ 44k
langgraph optional Agent frameworks & SDKs Open source
Orchestrates multi-step research flows
Build resilient agents.
How to set it up
- 01Try built-in deep research in ChatGPT, Gemini or Perplexity first to confirm the need
- 02To self-host: SearXNG for search, Crawl4AI for fetching, MarkItDown for files
- 03Run GPT Researcher as is, or orchestrate search → read → write → cite-check yourself with LangGraph
- 04Every claim in the report needs a source link; spot-check by hand
✓ Good fit
- You need controlled sources or internal data
✗ Look elsewhere if
- Occasional research: use an off-the-shelf product