EnrichMCP is a Python framework designed to create a semantic layer for AI agents to interact with data, acting as an ORM for AI. It transforms data models into typed, discoverable tools, enabling agents to understand and navigate complex data structures, relationships, and business logic.

How It Works

EnrichMCP builds upon the Model Context Protocol (MCP) by adding semantic understanding and data interaction capabilities. It automatically generates typed tools from data models, handles relationships between entities, provides schema discovery, and validates all inputs/outputs using Pydantic. This approach allows AI agents to query, fetch, and manipulate data through a consistent, type-safe interface, abstracting away backend complexities.

Quick Start & Requirements

Install with pip: pip install enrichmcp. For SQLAlchemy support: pip install enrichmcp[sqlalchemy]. Requires Python 3.7+.

Highlighted Details

• Automatic schema discovery for AI agents.
• Seamless relationship navigation between entities.
• Full Pydantic type safety and validation for all interactions.
• Support for CRUD operations, pagination, request caching, and context management.

Maintenance & Community

Built by Featureform. Contributions are welcome via CONTRIBUTING.md.

Licensing & Compatibility

Licensed under Apache 2.0. Compatible with commercial and closed-source applications.

Limitations & Caveats

The framework is primarily focused on Python environments. While it supports various backends, integration complexity may vary. Specific performance characteristics depend heavily on the underlying data sources and AI agent implementation.