An MCP server implementation that integrates the SearXNG API for powerful web search capabilities and uses @missionsquad/puppeteer-scraper to read and process live web content.
Deep Research infrastructure for AI agents. Search the web, read pages incrementally, follow source links, and collect relevant evidence across multiple sources. Web search runs through your own SearXNG instance, while the agent remains in control of research planning and reasoning. Cluefinch runs locally over MCP stdio.