MCPEngine

What is MCPEngine?

MCPEngine (EnrichMCP) is a Python framework built on the Model Context Protocol (MCP) that helps AI agents understand and navigate your data. It adds a semantic layer that turns your data model into typed, discoverable tools—like an ORM for AI. It automatically generates typed tools from data models, handles entity relationships, provides schema discovery, validates all inputs/outputs with Pydantic, and works with any backend (databases, APIs, or custom logic). It is designed for developers who want to make their data AI-navigable with minimal boilerplate.

How to use MCPEngine?

Install via pip install enrichmcp or pip install enrichmcp[sqlalchemy] for SQLAlchemy support. To use, define data entities as Pydantic models with the EnrichModel base class and decorate them with @app.entity(). For SQLAlchemy users, add the EnrichSQLAlchemyMixin to your declarative base and call include_sqlalchemy_models(). Then run the MCP server with app.run(). AI agents can then call tools like explore_data_model(), list_*(), and navigate relationships automatically.

Key features of MCPEngine

• Automatic schema discovery – AI agents explore your entire data model with one call
• Relationship navigation – define relationships once, agents traverse naturally
• Type safety and validation – full Pydantic validation on every interaction
• Mutability and CRUD – auto-generated patch models for updates
• Built-in pagination – handle large datasets with PageResult
• Context and authentication – pass auth, database connections, or any context
• Request caching – reduce API overhead with per-request or global cache

Use cases of MCPEngine