Skip to content

Welcome to enrichmcp

Transform Your Data Model into an MCP API

enrichmcp brings the power of type-safe, relationship-aware data models to AI agents. Built on top of FastMCP, it provides the missing data layer that enables Agentic Enrichment - giving AI agents the ability to discover, understand, and navigate your data through intelligent schema introspection.

Quick Example

from enrichmcp import EnrichMCP, EnrichModel, Relationship
from pydantic import Field

# Create your API
app = EnrichMCP("Customer API", "Customer data for AI agents")


# Define your data model
@app.entity
class Customer(EnrichModel):
    """A customer in our system."""

    id: int = Field(description="Unique customer ID")
    name: str = Field(description="Customer name")
    email: str = Field(description="Contact email")

    orders: list["Order"] = Relationship(description="Orders placed by this customer")


@app.entity
class Order(EnrichModel):
    """A customer order."""

    id: int = Field(description="Order ID")
    total: float = Field(description="Order total in USD")
    status: str = Field(description="Order status")

    customer: Customer = Relationship(description="Customer who placed this order")


# Define how relationships are resolved
@Customer.orders.resolver
async def get_customer_orders(customer_id: int) -> list[Order]:
    # Your database logic here
    return await db.get_orders_for_customer(customer_id)


# Create entry points for AI
@app.retrieve
async def get_customer(customer_id: int) -> Customer:
    """Get a customer by ID."""
    return await db.get_customer(customer_id)


# Run it!
app.run()

# Serve over HTTP instead of stdio
app.run(transport="streamable-http")

Already using SQLAlchemy? See how to turn existing models into an API with just a few lines.

Why enrichmcp?

🤖 Built for AI Agents

Unlike traditional APIs designed for humans, enrichmcp APIs are designed for AI:

  • Self-Describing: Every element includes rich descriptions
  • Discoverable: AI can explore your entire data model with one call
  • Navigable: Relationships allow natural data traversal

🛡️ Type Safety First

Built on Pydantic, enrichmcp provides:

  • Automatic Validation: Invalid data never reaches your code
  • IDE Support: Full autocomplete and type hints
  • Runtime Safety: Catch errors before they happen

🚀 Zero Boilerplate

Focus on your data model, not protocol details:

  • Automatic Tool Generation: Decorators handle all MCP setup
  • Smart Defaults: Sensible conventions out of the box
  • Clean APIs: Intuitive patterns inspired by GraphQL and ORMs
  • Mutable Fields: Opt‑in mutability with auto-generated patch models

How It Works

  1. Define Your Model: Use Pydantic models with the @app.entity decorator
  2. Add Relationships: Connect entities with typed relationships
  3. Implement Resolvers: Define how relationships load data
  4. Create Resources: Add entry points for AI agents
  5. Run: Start your MCP server

AI agents can then: - Call explore_data_model() to understand your schema - Use resources to access data - Follow relationships to traverse your data graph - All with full type safety and validation

Installation

pip install enrichmcp

Next Steps

Support

Built with ❤️ by Featureform