Cursor MCP Setup Guide

This guide provides step-by-step instructions for setting up the Registry MCP with Cursor.

Quick Setup

Step 1: Install Dependencies

# Clone and setup the repository
git clone https://github.com/slolab/registry-mcp.git
cd registry-mcp

# Install dependencies with uv
uv sync

# Install the package in development mode
uv pip install -e .

Step 2: Find Your Python Path

# Get the Python executable path
uv run which python

This will output something like: /Users/username/GitHub/registry-mcp/.venv/bin/python

Step 3: Configure Cursor

Create or edit your Cursor MCP configuration file:

macOS/Linux: ~/.cursor/mcp.json Windows: %APPDATA%\.cursor\mcp.json

{
  "mcpServers": {
    "registry-mcp": {
      "command": "/Users/username/GitHub/registry-mcp/.venv/bin/python",
      "args": ["-u", "-m", "registry_mcp.main"],
      "workingDirectory": "/Users/username/GitHub/registry-mcp",
      "env": { "PYTHONUNBUFFERED": "1" }
    }
  }
}

Important: Replace /Users/username/GitHub/registry-mcp with your actual path.

Step 4: Test the Configuration

# Test the command manually
/path/to/your/registry-mcp/.venv/bin/python -u -m registry_mcp.main --help

Step 5: Restart Cursor

Close and restart Cursor completely for the changes to take effect.

Automated Setup Script

You can use this script to automate the setup:

#!/bin/bash
# setup-cursor-mcp.sh

# Get the current directory (should be registry-mcp)
REGISTRY_MCP_DIR="$(pwd)"
PYTHON_PATH="$REGISTRY_MCP_DIR/.venv/bin/python"
CURSOR_CONFIG_DIR="$HOME/.cursor"

# Create Cursor config directory if it doesn't exist
mkdir -p "$CURSOR_CONFIG_DIR"

# Create the configuration
cat > "$CURSOR_CONFIG_DIR/mcp.json" << EOF
{
  "mcpServers": {
    "registry-mcp": {
      "command": "$PYTHON_PATH",
      "args": ["-u", "-m", "registry_mcp.main"],
      "workingDirectory": "$REGISTRY_MCP_DIR",
      "env": { "PYTHONUNBUFFERED": "1" }
    }
  }
}
EOF

echo "✅ MCP configuration created at: $CURSOR_CONFIG_DIR/mcp.json"
echo "📝 Configuration:"
echo "   Command: $PYTHON_PATH"
echo "   Working Directory: $REGISTRY_MCP_DIR"
echo ""
echo "🔄 Please restart Cursor to apply the changes."
echo ""
echo "🧪 Test the configuration:"
echo "   $PYTHON_PATH -u -m registry_mcp.main --help"

Save this as setup-cursor-mcp.sh, make it executable, and run it:

chmod +x setup-cursor-mcp.sh
./setup-cursor-mcp.sh

Testing the Connection

After restarting Cursor, test the connection by asking Cursor to:

1. "Analyze my current project directory"
2. "Get registry workflow guidance"
3. "Show me example submissions"
4. "Help me create a YAML specification for my component"

If the connection works, Cursor should be able to use the Registry MCP tools to help you with registry submissions.

Troubleshooting

Common Issues

1. "No module named registry_mcp"
2. Make sure you're using the Python from the .venv/bin/python path

3. Verify the package is installed with uv pip install -e .

4. "Command not found"

5. Use absolute paths in the configuration

6. Check that the Python executable exists at the specified path

7. "Permission denied"

8. Ensure the virtual environment is properly set up

9. Run uv sync to recreate the environment if needed

10. Cursor doesn't recognize the MCP

11. Restart Cursor completely
12. Check the configuration file syntax (valid JSON)
13. Verify the file is in the correct location

Debugging Steps

1. Test the command manually:

   /path/to/registry-mcp/.venv/bin/python -u -m registry_mcp.main --help
   

2. Check the configuration file:

   cat ~/.cursor/mcp.json
   

3. Verify the Python path:

   cd /path/to/registry-mcp
   uv run which python
   

4. Check Cursor's logs for MCP-related errors

Next Steps

Once the MCP is connected, you can:

1. Follow the Usage Guide for step-by-step instructions
2. Use the Quick Reference for tool parameters
3. Start annotating your components for registry submission

The Registry MCP provides 10 specialized tools to help you create, validate, and submit schema.org-compatible specifications to the BioContextAI registry.