mcp-local-rag and supernova-mcp-rag
These two tools are complements, as nkapila6/mcp-local-rag provides a "primitive" RAG-like web search model context protocol (MCP) server that runs locally, while shabib87/supernova-mcp-rag is a practical proof-of-concept demonstrating how to build and run a local MCP server with RAG, implying the latter could leverage or build upon the functionality provided by the former.
About mcp-local-rag
nkapila6/mcp-local-rag
"primitive" RAG-like web search model context protocol (MCP) server that runs locally. ✨ no APIs ✨
This tool helps large language models (LLMs) like Claude perform deep, up-to-date web research and answer questions using current information. It takes your natural language questions, searches across many web sources, extracts the most relevant details, and feeds them back to the LLM. Researchers, analysts, or anyone needing accurate, real-time information from an AI assistant would find this useful.
About supernova-mcp-rag
shabib87/supernova-mcp-rag
A practical POC demonstrating how to build and run a local MCP server with Retrieval-Augmented Generation (RAG) for semantic search over internal documentation. Leverages Node.js, TypeScript, Hugging Face embeddings, and an in-memory vector store to enable fast, context-aware answers in tools like Cursor.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work