RAG-Powered MCP Server

Example Use Cases

• 1

  Answer questions based on private PDF document collections.

• 2

  That require source-grounded, non-hallucinated answers from indexed documents.

• 3

  Integrate document Q&A capabilities into Claude Desktop via MCP.

Description

This MCP server implements a Retrieval-Augmented Generation pipeline that indexes PDFs semantically and allows Claude Desktop to query them in real-time. It uses FAISS for vector similarity search, Sentence Transformers for embeddings, and Groq LLaMA for fast, source-grounded answer generation. The server provides a Streamlit chat interface for interactive querying and supports adjustable retrieval depth to improve answer relevance.