ai-coding-gym-mcp 0.1.0

MCP server for AI Coding Gym - fetch and submit coding challenges

AI Coding Gym - MCP Server

Local MCP server for interacting with the AI Coding Gym platform. Provides tools to fetch coding problems and submit solutions.

Features

• /fetch: Fetch a coding problem and clone the repository to your local machine
• /submit: Submit your solution by committing and pushing changes

Quick Start

Installation

Option 1: Install from PyPI

pip install ai-coding-gym-mcp

Option 2: Install from GitHub

pip install git+https://github.com/yourusername/ai-coding-gym-mcp.git

Option 3: Install from source

git clone https://github.com/yourusername/ai-coding-gym-mcp.git
cd ai-coding-gym-mcp
pip install -e .

Configure Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on Mac):

{
  "mcpServers": {
    "ai-coding-gym": {
      "command": "python",
      "args": ["-m", "server"],
      "env": {
        "AI_CODING_GYM_SERVER": "https://your-server-url.com"
      }
    }
  }
}

See INSTALLATION.md for detailed setup instructions for Claude Desktop and VS Code.

Usage

Running the MCP Server

The server uses stdio for communication with MCP clients:

python server.py

Or configure it in your MCP client settings (e.g., Claude Desktop).

Tool: /fetch

Fetches a problem from the backend and clones the repository locally.

Parameters:

• problem_id (required): Problem identifier (e.g., "django__django-10097")
• user_id (required): Your user ID for authentication
• server_url (optional): Backend server URL (default: "https://api.example.com")
• workspace_dir (optional): Local workspace directory (default: "./workspace")

Example:

{
  "problem_id": "django__django-10097",
  "user_id": "user_123",
  "server_url": "https://ai-coding-gym.example.com"
}

What it does:

1. Calls backend API to get repository URL and deployment key
2. Clones the repository to workspace/{problem_id}/
3. Checks out the problem-specific branch
4. Saves the problem statement to PROBLEM_STATEMENT.md

Tool: /submit

Submits your solution by committing changes and pushing to the remote repository.

Parameters:

• problem_id (required): Problem identifier
• user_id (required): Your user ID for authentication
• server_url (optional): Backend server URL (default: "https://api.example.com")
• commit_message (optional): Custom commit message

Example:

{
  "problem_id": "django__django-10097",
  "user_id": "user_123",
  "commit_message": "Fixed the authentication bug"
}

What it does:

1. Stages all changes in the working directory (git add -A)
2. Commits with the provided or auto-generated message
3. Pushes to the remote branch using deployment key
4. Notifies the backend server about the submission

Backend API Endpoints

The MCP server expects the following backend API endpoints:

POST /api/fetch-problem

Request:

{
  "problem_id": "django__django-10097",
  "user_id": "user_123"
}

Response:

{
  "repo_url": "[e-mail]:org/repo.git",
  "branch": "django__django-10097-user_123",
  "deploy_key": "-----BEGIN OPENSSH PRIVATE KEY-----\n...\n-----END OPENSSH PRIVATE KEY-----",
  "problem_statement": "# Problem Description\n\n..."
}

POST /api/submit

Request:

{
  "problem_id": "django__django-10097",
  "user_id": "user_123",
  "commit_hash": "abc123def456...",
  "branch": "django__django-10097-user_123",
  "timestamp": "2026-02-03T10:30:00"
}

Response:

{
  "status": "success",
  "message": "Submission received"
}

Security

• Deployment keys are stored in ~/.mcp-keys/ with 600 permissions
• Keys are scoped per problem and managed by the backend
• SSH host key checking is disabled for convenience (consider enabling in production)
• Credentials are cached in memory during the MCP server session

Configuration

Default server URL is https://api.example.com. You can override it by passing server_url parameter to each tool call, or set it via environment variable:

export AI_CODING_GYM_SERVER="https://your-server.com"

Troubleshooting

"No credentials found for problem_id"

• Run /fetch first to download the problem and credentials

"Git clone/push failed"

• Check network connectivity
• Verify deployment key is valid
• Ensure SSH agent isn't interfering

"Directory already exists"

• Remove the existing directory or use a different workspace location

Publishing

See PUBLISHING.md for instructions on:

• Publishing to PyPI
• Publishing to GitHub
• Version management
• Release workflow

Development

The server uses:

• mcp: Model Context Protocol SDK
• requests: HTTP client for backend API calls
• subprocess: Git command execution with SSH key management

Local Development

# Clone the repository
git clone https://github.com/yourusername/ai-coding-gym-mcp.git
cd ai-coding-gym-mcp

# Install in development mode
pip install -e .

# Run tests (if available)
pytest

# Test the server locally
python server.py

Contributing

Contributions welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

License

MIT