Building RagRabbit, An Open Source RAG Search with Postgres as the Vector Store

RagRabbit is an open-source RAG toolkit built by Marco D'Alia that crawls websites, converts pages to Markdown, generates vector embeddings stored in Postgres via pgVector, and serves doc chunks through an MCP server. It uses Neon as the serverless Postgres host and deploys on Vercel with a single click. The tool eliminates the need for a separate vector database by leveraging pgVector for similarity search alongside standard Postgres full-text search. It supports OpenAI and Claude for embeddings and Q&A, and includes authentication options like GitHub OAuth and magic links.