mcp-app

The MCP APP is a robust Model Context Protocol server application designed to enhance the capabilities of Language Learning Models (LLMs) by integrating Retrieval-Augmented Generation (RAG) and web searching tools. The RAG tools enable the LLM to not only retrieve knowledge from a vectorstore but also to add documents, thereby expanding the knowledge base that the LLM can utilize. The application is built using a technology stack that includes SQLAlchemy for database interaction, OpenAI for embedding vectorstore, PostgreSQL as the database, and PGVector as the vectorstore. This setup allows for efficient data management and retrieval, making it a powerful tool for applications requiring dynamic and extensive data handling.

Features

• MCP Server Application: Implements the MCP server for efficient data handling.
• SQLAlchemy ORM: Facilitates interaction with the SQL database.
• OpenAI Integration: Provides embedding capabilities for vectorstore.
• PostgreSQL Database: Serves as the primary SQL database.
• PGVector: Utilized as the vectorstore for data retrieval and storage.

Usage with Different Platforms

QuickStart

sh
uv sync
source .venv/bin/activate
mcp dev run

Combining Claude Desktop with MCP APP

sh
# You must install all dependencies written in pyproject.toml
mcp install server.py --env-file .env --with sqlalchemy --with pgvector --with openai --with "psycopg[binary]" --with pydantic --with python-dotenv --with tavily-python