mcp-test-server

mcp-test-server

0
developer_toolsai_chatbotknowledge_and_memoryfile_systems

The MCP Test Server is a Python-based implementation leveraging the Model Context Protocol to enable a RAG system with OpenAI's vector store capabilities. It allows for creating and querying vector databases from local directories, integrating seamlessly with other applications like Claude.

MCP Test Server

A Python server using the Model Context Protocol (MCP) that provides a Retrieval Augmented Generation (RAG) system leveraging OpenAI's vector store features.

Features

  • Create Vector DB: Constructs a vector database from files in a specified directory (text, PDF, DOCX, markdown).
  • Query Vector DB: Executes queries on the constructed vector DB to retrieve relevant information.

Requirements

  • Python 3.11 or higher
  • OpenAI API Key

Installation

1. Project Initialization and Virtual Environment Creation

2. Install Required Packages

3. Set Environment Variables

Testing the MCP Server (Optional)

Testing with MCP Inspector

Integration with the Claude App

Provided Tools

  1. create_vector_db_from_directory - Creates a vector DB from directory files.
  2. query_vector_db - Searches the vector DB to retrieve information.