MCP - AI Agent Example

This example demonstrates how to enable MCP (Model Context Protocol) support in Feast, allowing AI agents and applications to interact with your features through standardized MCP interfaces.

Prerequisites

  1. Python 3.8+

  2. Feast installed

  3. FastAPI MCP library

Installation

  1. Install Feast with MCP support:

pip install feast[mcp]

Alternatively, you can install the dependencies separately:

pip install feast
pip install fastapi_mcp

Setup

  1. Navigate to this example directory within your cloned Feast repository:

cd examples/mcp_feature_store
  1. Initialize a Feast repository in this directory. We'll use the existing feature_store.yaml that's already configured for MCP:

feast init . 

This will create a data subdirectory and a feature_repo subdirectory if they don't exist, and will use the feature_store.yaml present in the current directory (examples/mcp_feature_store).

  1. Apply the feature store configuration:

cd feature_repo 
feast apply
cd .. # Go back to examples/mcp_feature_store for the next steps

Starting the MCP-Enabled Feature Server

Start the Feast feature server with MCP support:

feast serve --host 0.0.0.0 --port 6566

If MCP is properly configured, you should see a log message indicating that MCP support has been enabled:

INFO:feast.feature_server:MCP support has been enabled for the Feast feature server

Available MCP Tools

The fastapi_mcp integration automatically exposes your Feast feature server's FastAPI endpoints as MCP tools. This means AI assistants can:

  • Call /get-online-features to retrieve features from the feature store

  • Use /health to check server status

Configuration Details

The key configuration that enables MCP support:

feature_server:
    type: mcp                    # Use MCP feature server type
    enabled: true               # Enable feature server
    mcp_enabled: true           # Enable MCP protocol support
    mcp_server_name: "feast-feature-store"
    mcp_server_version: "1.0.0"

Last updated

Was this helpful?