What is Rfc Mcp

rfc-mcp is a project that creates a simple Retrieval-Augmented Generation (RAG) application to store embeddings of text documents in Qdrant, allowing users to query this Vector Database through an MCP server and interact with it using a client powered by Claude.

Use cases

Use cases for rfc-mcp include academic research for retrieving relevant documents, customer support for quick information retrieval, and any application that requires efficient searching and interaction with large text datasets.

How to use

To use rfc-mcp, follow these steps: 1. Use the Jupyter notebook ‘idx_qdrant.ipynb’ to index your documents into Qdrant. 2. Copy the ‘.env.example’ file to ‘.env’ and fill in the necessary values. 3. Run the MCP server with the command ‘./run_server.sh’. 4. In a separate terminal, execute the MCP client using ‘./run_client.sh’.

Key features

Key features of rfc-mcp include the ability to store and query text document embeddings, integration with Qdrant for efficient vector storage and retrieval, and a user-friendly client interface powered by Claude for interaction.

Where to use

rfc-mcp can be used in various fields such as natural language processing, information retrieval, and machine learning applications where efficient document querying and embedding storage are required.

RFC MCP

This is a weekend project to create a simple RAG application that stores embeddings of text documents using Qdrant, and a MCP server to query this Vector Database, and a MCP client using Claude to interact.

Steps

1. Jupyter notebook idx_qdrant.ipynb has the Python code to index documents into Qdrant.
2. Copy .env.example to .env and fill in the values.
3. Run the MCP server using ./run_server.sh.
4. In another terminal, run the MCP client using ./run_client.sh.