mcp-github-research-assistant

DecodertoMux/mcp-github-research-assistant

3.1

If you are the rightful owner of mcp-github-research-assistant and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

A modular MCP server that connects to the GitHub API to facilitate intelligent research and analysis of open-source repositories.

MCP GitHub Research Assistant

A modular MCP (Model Context Protocol) server that connects to the GitHub API to search, analyze, and compare repositories.
This project provides intelligent research tools that help you discover, inspect, and evaluate open-source projects with ease.

Features:

Search Repositories

  • Search GitHub by keyword, language, or topic
  • Filter by minimum stars and sort order (stars / forks / updated)
  • Returns key metrics like stars, forks, and last update date

Get Repository Details

  • View full metadata: stars, forks, license, open issues, and topics
  • Includes a README preview (decoded from base64)
  • Displays creation, update, and last push times in human-readable format

Compare Repositories

  • Compare 2–3 repositories side by side
  • Table includes stars, forks, language, size, open issues, license, and activity recency
  • Useful for benchmarking frameworks or libraries

Get Trending Repositories

  • Discover trending repos by language and time period (daily, weekly, monthly)
  • Automatically refreshes every 10 minutes via caching

Check Repository Activity

  • Check if a project is actively maintained
  • Displays last commit, push date, and activity rating:
    • Active (<30 days)
    • Moderate (<90 days)
    • Slow (<180 days)
    • Inactive (>6 months)
  • Includes indicators for issue tracking, wiki, and archived status

Technical Overview

  • Language: Python 3.10+
  • Framework: [mcp.server]
  • API: GitHub REST API v3
  • Caching: cachetools.TTLCache for efficient API use
  • Async runtime: asyncio
  • Logging: All logs printed to stderr to maintain clean MCP protocol output

Installation

1. Clone the Repository

git clone <repository-url>
cd mcp-github-research-assistant

2. Set Up Virtual Environment

# Create a virtual environment
python -m venv venv

# Activate it
source venv/bin/activate  # On macOS/Linux
# or
venv\Scripts\activate     # On Windows

3. Install Dependencies

pip install -r requirements.txt

4. Configure MCP Client

Add this server configuration to your MCP client config file:

{
  "mcpServers": {
    "github-research": {
      "command": "/path-to/mcp-github-research-assistant/venv/bin/python",
      "args": ["/path-to/mcp-github-research-assistant/server.py"]
    }
  }
}

5. (Optional) Add GitHub Token

For higher rate limits, add your GitHub token to :

GITHUB_TOKEN = "your_github_token_here"