Data Dictionary MCP

A Model Context Protocol (MCP) server that coordinates AI agents to transform database tables into Wikipedia-style data dictionaries.

Overview

The Data Dictionary MCP project automates the conversion of various database formats into comprehensive, human-readable data dictionaries using AI-powered analysis and description. It leverages the Model Context Protocol (MCP) to coordinate AI agents for analyzing, describing, and verifying database structures.

Features

• Multi-Format Support: Process JSON, CSV, and Plain Text files (with more formats planned)
• AI-Powered Analysis: Generate field descriptions and identify relationships
• MCP Integration: Coordinate AI agents using the Model Context Protocol
• Schema Extraction: Extract database schemas from various formats into a unified representation
• Wikipedia-Style Output: Present data dictionaries in a familiar, accessible format

Project Status

This project is in active development. See the Project Roadmap for details.

Getting Started

Prerequisites

• Python 3.9+
• Git
• pip or poetry for dependency management

Installation

1. Clone the repository:

   git clone https://github.com/jonahkeegan/data-dictionary-mcp.git
   cd data-dictionary-mcp
   

2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install dependencies:

   pip install -r requirements.txt
   

4. Run the application:

   python src/main.py
   

Project Structure

data-dictionary-mcp/
├── docs/                  # Documentation
├── src/                   # Source code
│   ├── mcp/               # MCP server components
│   ├── analyzers/         # Format analyzers
│   ├── agents/            # Agent coordination
│   └── dictionary/        # Dictionary generation
├── tests/                 # Test suite
├── memory-bank/           # Cline memory bank
├── .gitignore
├── .clinerules            # Cline rules
├── README.md
└── requirements.txt

Project Roadmap

Milestone 1: MCP Server Foundation and Format Analyzers

• Implement MCP server with basic tool definitions
• Develop format analyzers for JSON, CSV, and Plain Text
• Create schema extraction system
• Implement unit tests for core components

Milestone 2: AI Agent Coordination and Field Description

• Implement agent coordination system
• Develop field description generation
• Create task distribution and result aggregation
• Add integration tests

Milestone 3: Content Verification and Publishing

• Implement content validation
• Develop Wikipedia-style formatting
• Create export capabilities
• Add end-to-end tests

Milestone 4: User Interface and Deployment

• Develop web interface
• Implement search capabilities
• Add user feedback system
• Create deployment infrastructure

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is open source and available under the MIT License.