RAGmonsters-mcp-pg
If you are the rightful owner of RAGmonsters-mcp-pg and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This repository demonstrates a custom MCP server for integrating LLMs with databases using the RAGmonsters dataset.
The Custom PostgreSQL MCP Server for RAGmonsters is an advanced approach to integrating Large Language Models (LLMs) with databases using the Model Context Protocol (MCP). Unlike generic MCP PostgreSQL servers that allow LLMs to explore databases through raw SQL queries, this project creates a custom MCP server providing a domain-specific API tailored to the application's needs. Utilizing FastMCP, a high-performance MCP implementation, this server enhances efficiency and reliability for tool-based interactions with LLMs. The project is based on the RAGmonsters dataset, an open-source fictional dataset designed for testing Retrieval-Augmented Generation (RAG) systems. The custom MCP server abstracts database complexity, provides domain-specific operations, optimizes common queries, enforces business rules, and improves security by limiting direct SQL access.
Features
- Custom MCP Server with FastMCP: High-performance domain-specific API for RAGmonsters data
- Optimized Queries: Pre-built efficient database operations
- Business Logic Layer: Domain rules and constraints embedded in the API
- Structured Response Format: Consistent JSON responses for LLM consumption
- Comprehensive Logging: Detailed logging for debugging and monitoring
Tools
getMonsters
Get a list of monsters, supporting filtering, sorting and pagination
getMonsterById
Get the details of the monster with ID
add
Simple addition tool (for testing)
