Repo Radar MCP

Discover, rank and compare GitHub repositories from any MCP-compatible AI client.

Repo Radar MCP is a Python-based MCP server that connects AI agents to the GitHub API, allowing them to search, filter, rank and evaluate open-source repositories by topic, language, stars, license, activity and relevance.

It is designed for developers, builders and AI agents that need a structured way to research open-source projects.

Why Repo Radar MCP?

AI agents can write code, but they also need good technical context.

Repo Radar MCP gives your MCP-compatible client a simple set of tools to answer questions like:

• What are the most popular Python repositories about RAG?

• Which MCP servers are worth studying?

• What GitHub projects are active, licensed and useful?

• Which repository should I use as a reference for my next project?

• How do several repositories compare by stars, forks, issues, license and recent activity?

Features

• Search GitHub repositories by topic.

• Filter by programming language.

• Filter by minimum stars.

• Rank repositories with a simple usefulness score.

• Analyze a single repository.

• Compare multiple repositories.

• Fetch repository README content.

• Return results as JSON or Markdown.

• Uses your GitHub token safely from environment variables.

• Works with MCP Inspector, Claude Desktop, Cursor and other MCP-compatible clients.

MCP Tools

Installation

Clone the repository:

git clone https://github.com/javiermorron/repo-radar-mcp.git
cd repo-radar-mcp

Create and activate a virtual environment.

Windows PowerShell

python -m venv .venv
.\.venv\Scripts\activate

macOS / Linux

python -m venv .venv
source .venv/bin/activate

Install the project:

pip install -e .

Copy the environment file:

cp .env.example .env

On Windows PowerShell:

copy .env.example .env

Edit .env and add your GitHub token:

GITHUB_TOKEN=your_github_token_here
GITHUB_API_VERSION=2022-11-28
GITHUB_USER_AGENT=repo-radar-mcp/1.0.0

Never commit your real .env file.

Run with MCP Inspector

From the project root:

mcp dev server.py

If the Inspector does not find uv, use this configuration in the Inspector:

Transport Type: STDIO
Command: python
Arguments: server.py

Then open the Tools tab and run:

{
  "topic": "mcp server",
  "language": "Python",
  "limit": 5,
  "min_stars": 10
}

Claude Desktop Example

A sample configuration is available in:

examples/claude_desktop_config.example.json

Example Windows entry:

{
  "mcpServers": {
    "repo-radar-mcp": {
      "command": "C:\\Users\\YOUR_USER\\repo-radar-mcp\\.venv\\Scripts\\python.exe",
      "args": [
        "C:\\Users\\YOUR_USER\\repo-radar-mcp\\server.py"
      ]
    }
  }
}

Example Prompts

Search the 5 most popular Python repositories about "mcp server" and explain which one is best to study.

Compare these repositories: freqtrade/freqtrade, ccxt/ccxt, hummingbot/hummingbot.

Find popular repositories about "rag chatbot" in Python with more than 500 stars and rank them by usefulness.

Analyze microsoft/autogen and tell me if it is active, useful and worth studying.

More prompts are available in:

examples/prompts.md

Repository Score

Repo Radar MCP includes a simple scoring system based on:

• Stars

• Forks

• Recent activity

• License availability

• Open issues

• Archived status

The score is not meant to replace human judgment. It is a quick signal to help agents and developers prioritize what to inspect first.

Project Structure

repo-radar-mcp/
├── src/
│   └── repo_radar_mcp/
│       ├── __init__.py
│       ├── server.py
│       ├── github_client.py
│       ├── scoring.py
│       ├── formatters.py
│       └── models.py
├── examples/
│   ├── claude_desktop_config.example.json
│   └── prompts.md
├── tests/
│   └── test_scoring.py
├── .env.example
├── .gitignore
├── LICENSE
├── README.md
├── CHANGELOG.md
├── requirements.txt
├── pyproject.toml
└── server.py

Roadmap

• Add repository release analysis.

• Add issue quality analysis.

• Add contributor activity metrics.

• Add support for GitHub topics recommendations.

• Add repository health report.

• Add CSV and JSON export helpers.

• Add Docker support.

• Add GitHub Actions for tests and linting.

Security

Repo Radar MCP uses a GitHub token from environment variables.

Do not commit:

• .env

• Personal access tokens

• Private API keys

• Local virtual environments

The .gitignore file already excludes common sensitive and generated files.

Contributing

Contributions are welcome.

Good first issues:

• Improve scoring logic.

• Add more output formats.

• Add tests.

• Improve MCP client examples.

• Add Docker support.

License

This project is licensed under the MIT License.

Created by Javier Morrón. Connect with me on LinkedIn: https://www.linkedin.com/in/javiermorron.