Happy slice of mcp pizzaMCP.pizza

local_faiss_mcp

MCP.Pizza Chef: nonatofabio

This server lets you turn your local documents—like PDFs, text files, and more—into a searchable knowledge base that your AI assistant can understand and query. By running it on your computer, you can ask natural language questions and get relevant answers or summaries from your own files. It works with Claude Desktop and other MCP-compatible AI apps, requiring some setup involving Python but no external accounts or keys.

Files/PDF
Other
Web/Research

Use This MCP server To

Search my local documents using natural language questions Ingest PDFs and text files for AI-powered semantic search Get concise answers with source citations from my documents Summarize multiple documents on a topic quickly Use Claude Desktop to interact with my local document knowledge base Customize search by selecting different embedding models Run a local AI memory store without internet dependency

README

Local FAISS MCP Server

License: MIT Python 3.10+ Tests PyPI version

A Model Context Protocol (MCP) server that provides local vector database functionality using FAISS for Retrieval-Augmented Generation (RAG) applications.

demo

Features

Core Capabilities

  • Local Vector Storage: Uses FAISS for efficient similarity search without external dependencies
  • Document Ingestion: Automatically chunks and embeds documents for storage
  • Semantic Search: Query documents using natural language with sentence embeddings
  • Persistent Storage: Indexes and metadata are saved to disk
  • MCP Compatible: Works with any MCP-compatible AI agent or client

v0.2.0 Highlights

  • CLI Tool: local-faiss command for standalone indexing and search
  • Document Formats: Native PDF/TXT/MD support, DOCX/HTML/EPUB with pandoc
  • Re-ranking: Two-stage retrieve and rerank for better results
  • Custom Embeddings: Choose any Hugging Face embedding model
  • MCP Prompts: Built-in prompts for answer extraction and summarization

Quickstart

local_faiss_mcp FAQ

Can I use this to search and summarize my local PDF documents?
Yes — you can ingest PDFs and other document types, then query them using natural language to get summaries or specific answers with Claude Desktop or any MCP-compatible AI.
Does this work with Claude Desktop?
Yes — local_faiss_mcp is designed to integrate with Claude Desktop and other MCP clients for local semantic search.
Do I need an API key or account to use this?
No — it runs locally on your machine without requiring any external API keys or accounts.
How hard is it to set up?
It requires developer-level setup, including installing Python packages and running the server from the command line.
What document formats can I use?
You can index TXT, MD, PDF natively, and DOCX, HTML, EPUB, and 40+ formats with pandoc installed.
Can I customize the search quality?
Yes — you can choose different embedding models and enable re-ranking for better search results.