Deep Web Research MCP
Published:
Timeline: March 2026
Project Description: A local web-research system built around an MCP (Model Context Protocol) server and a terminal client. Given a query, it searches with SearXNG, extracts article text with Trafilatura, cleans it, and ranks the results with BM25. Results can go to any MCP-compatible AI tool, or be piped into a local LLM such as Ollama. Built with Python and FastMCP.
Project Github Link: deep-web-research-mcp
