Back to MCP Catalog

Eunomia Data Governance MCP Server

Data Science ToolsPython
Data governance orchestration for LLM-based applications

Eunomia Data Governance is an extension of the Eunomia framework that connects data governance instruments with MCP servers. It provides a robust way to enforce policies like PII detection, user access control, and other governance requirements on text flowing through LLM pipelines. With Eunomia, developers can seamlessly integrate data governance into their existing MCP ecosystem, orchestrating multiple servers while ensuring compliance with privacy and security standards. The framework uses a modular "instrument" approach that makes it easy to apply specific policies to text streams without disrupting the underlying application architecture.

Overview

Eunomia Data Governance allows you to enforce data governance policies on text-based pipelines, particularly those using Large Language Models. It integrates with the Model Context Protocol (MCP) ecosystem, enabling you to orchestrate multiple servers while maintaining control over sensitive data.

Installation

To get started with Eunomia Data Governance:

  1. Clone the repository:
git clone https://github.com/whataboutyou-ai/eunomia-mcp-server.git
  1. Install the package and its dependencies using your preferred Python package manager (pip, uv, etc.)

Basic Usage

Eunomia uses the concept of "instruments" to define data governance policies. These instruments are organized in an Orchestra that processes text streams flowing through your MCP-based servers.

Setting Up Your Configuration

Create a settings file that defines your MCP server configuration and the instruments you want to apply:

from pydantic_settings import BaseSettings
from eunomia.orchestra import Orchestra
from eunomia.instruments import PiiInstrument

class Settings(BaseSettings):
    APP_NAME: str = "eunomia-mcp-server"
    APP_VERSION: str = "0.1.0"
    LOG_LEVEL: str = "info"
    
    # Define the MCP servers you want to connect to
    MCP_SERVERS: dict = {
        "web-browser-mcp-server": {
            "command": "uv",
            "args": [
                "tool",
                "run",
                "web-browser-mcp-server"
            ],
            "env": {
                "REQUEST_TIMEOUT": "30"
            }
        }
    }
    
    # Define your Orchestra with the instruments you want to apply
    ORCHESTRA: Orchestra = Orchestra(
        instruments=[
            # This instrument detects and replaces email addresses and person names
            PiiInstrument(entities=["EMAIL_ADDRESS", "PERSON"], edit_mode="replace"),
            # Add more instruments as needed
        ]
    )

Running the Server

Once your settings are defined, run the Eunomia MCP server:

uv --directory "path/to/server/" run orchestra_server

This will:

  1. Load your settings
  2. Launch the Eunomia MCP Server
  3. Apply your defined instruments to incoming text
  4. Orchestrate communication with your external MCP servers

Available Instruments

Eunomia comes with several built-in instruments:

  • PiiInstrument: Detects and handles personally identifiable information
  • IdbacInstrument: Implements identity-based access control
  • And more from the core Eunomia framework

You can also create custom instruments by extending the base classes provided by Eunomia.

Advanced Configuration

For more complex setups, you can:

  • Chain multiple instruments together
  • Create custom instruments for specific governance needs
  • Configure detailed logging and monitoring
  • Set up environment-specific configurations

Refer to the Eunomia documentation for more detailed information on advanced usage patterns.

Related MCPs

Vega-Lite Data Visualization
Data Science ToolsPython

Create interactive data visualizations using Vega-Lite syntax

Open Data
Data Science ToolsPython

Connect any Open Data to any LLM with Model Context Protocol

Tinybird
Data Science ToolsPython

Query and interact with Tinybird workspaces from any MCP client

About Model Context Protocol

Model Context Protocol (MCP) allows AI models to access external tools and services, extending their capabilities beyond their training data.

Generate Cursor Documentation

Save time on coding by generating custom documentation and prompts for Cursor IDE.