MCP Server Dataset

🚀 MCP Server Dataset Builder

A comprehensive tool for building and maintaining datasets of Model Context Protocol (MCP) servers. This tool can automatically collect, classify, and update information about MCP servers from multiple sources.

🚀 Quick Start

The MCP Server Dataset Builder aims to streamline the process of gathering and organizing MCP server information. It extracts data from various sources, processes it, and generates a daily CSV file with detailed server information.

✨ Features

• Dual Data Sources: Combines data from a curated list and GitHub searches.
• Automatic Categorization: Assigns categories based on repository content.
• Tech Stack Detection: Identifies the programming languages and frameworks used.
• Emoji Marking: Adds visual indicators for quick identification.
• Daily Updates: Runs automatically to keep the dataset up - to - date.
• Data Persistence: Maintains historical data when adding new entries.

📦 Installation

To run the script locally:

# Install dependencies
pip install -r requirements.txt

# Run README extraction
python extract_mcp_servers.py

# Run daily updates
python daily.py

💻 Usage Examples

Basic Usage

The dataset is automatically updated via GitHub Actions without manual intervention.

Advanced Usage

You can manually trigger the workflow from the GitHub Actions tab:

1. Go to the "Actions" tab of the repository.
2. Select "Unified MCP Server Extraction".
3. Click "Run workflow".
4. Optional customizations:
   • GitHub search keywords
   • Minimum star and fork thresholds
   • Which extraction methods to run

📚 Documentation

Dataset Structure

The generated CSV file contains the following fields:

Environment Variables

You can use the following environment variables to customize the behavior:

Data Sources

1. awesome - mcp - servers Repository

The tool extracts MCP server information from the [awesome - mcp - servers](https://github.com/punkpeye/awesome - mcp - servers) repository, which contains a curated list of MCP servers.

2. GitHub Search

The tool searches for other MCP server repositories on GitHub to ensure comprehensive coverage of the ecosystem.

Classification System

Repositories are classified based on their content and purpose:

• Framework: Core MCP server implementations
• Tool: Auxiliary tools and utilities
• Client: Client libraries and applications
• Tutorial: Learning resources and examples
• Database: Database integrations
• API: API implementations
• Storage: Storage solutions
• AI: AI and LLM integrations
• Chat: Chat and messaging features
• Search: Search functionality

Tech Stack Detection

The tool identifies the following:

• Language: Programming languages used
• Framework: Web frameworks used
• Database: Integrated database systems
• Other Technologies: Such as APIs, third - party services, etc.

Project Description

The goal of this project is to automate the collection and organization of MCP server - related information for researchers and developers. By retrieving information from public repositories on GitHub and combining keywords extracted from READMEs, it generates structured CSV data for subsequent data analysis and processing.

How to Contribute

You are welcome to fork this repository and submit issues or pull requests to improve the tool or fix problems.

📄 License

[Add license information here]