local_faiss_mcp
Verified Safeby nonatofabio
Overview
Provides local vector database functionality using FAISS for Retrieval-Augmented Generation (RAG) applications, enabling semantic search and document ingestion for AI agents.
Installation
local-faiss-mcp --index-dir ./my_vector_storeSecurity Notes
The server operates via standard I/O (stdio) and manages local files within a user-specified directory. It loads embedding models from Hugging Face, which is a standard practice but introduces a potential supply chain risk if a loaded model itself were malicious. However, the server code does not exhibit direct code execution vulnerabilities, 'eval' usage, 'os.system' calls, or hardcoded secrets. File operations are confined to the designated index directory.
Similar Servers
mcp-local-rag
Provides a local RAG-like web search capability for LLMs through the Model Context Protocol without external APIs.
qdrant-loader
The QDrant Loader MCP Server provides advanced Retrieval-Augmented Generation (RAG) capabilities to AI development tools by bridging a QDrant knowledge base. It offers intelligent search through semantic, hierarchy-aware, and attachment-focused tools, integrating seamlessly with MCP-compatible AI tools to provide context-aware code assistance, documentation lookup, and intelligent suggestions.
Nova-LLM-mCP-memory-system
Production-grade GPU-accelerated vector memory for AI applications, providing secure and high-performance vector search.
mcp-rag-server
Provides a local Retrieval-Augmented Generation (RAG) server for any code repository, integrating with clients that speak the Model Context Protocol (MCP) like GitHub Copilot Agent.