Skip to main content

Prerequisites

Before installing Pylance MCP Pro, ensure you have:
  • Python 3.11+ installed
  • Node.js 18+ (for MCP client setup)
  • A supported MCP client (Claude Desktop, Cursor, Continue.dev, Windsurf)

Installation Methods

Install in an isolated environment:
pipx install pylance-mcp-pro
Recommended for avoiding dependency conflicts.
Clone and install from GitHub:
git clone https://github.com/JarrodAI/pylance-mcp.git
cd pylance-mcp
pip install -e .
Useful for development or customization.

Verify Installation

Check that the server is installed correctly:
pylance-mcp-pro --version
You should see the version number printed.

Configuration

1. Set Workspace Root

Set the WORKSPACE_ROOT environment variable to your Python project:
export WORKSPACE_ROOT=/path/to/your/project
Or configure it in your MCP client settings (see client-specific guides).

2. API Key (Optional)

If you purchased a subscription, set your API key:
export PYLANCE_MCP_API_KEY=plmc_your_api_key_here
Free tier is available without an API key with rate limits.

3. Configure MCP Client

Add the server to your MCP client configuration:
{
  "mcpServers": {
    "pylance-mcp-pro": {
      "command": "python",
      "args": ["server.py"],
      "env": {
        "WORKSPACE_ROOT": "${workspaceFolder}",
        "PYLANCE_MCP_API_KEY": "plmc_your_api_key_here"
      }
    }
  }
}
The ${workspaceFolder} variable will be automatically replaced with your current workspace path in most MCP clients.

Client-Specific Setup

Testing the Installation

Once configured, test the server with a simple command in your MCP client:
Can you analyze the Python files in my workspace and show me any type errors?
The AI should use the get_diagnostics tool and return type checking results.

Troubleshooting

Issue: MCP client can’t find the pylance-mcp commandSolution:
  • Ensure Python is in your PATH
  • Try using the full path: /path/to/python -m pylance_mcp
  • Verify installation: pip show pylance-mcp
Issue: Server reports empty workspaceSolution:
  • Verify WORKSPACE_ROOT points to a valid directory
  • Ensure directory contains .py files
  • Check file permissions
Issue: Diagnostics return empty resultsSolution:
  • Ensure Python environment is activated
  • Check for pyrightconfig.json or pyproject.toml
  • Verify Python interpreter path with get_python_environment tool
Issue: Receiving rate limit errorsSolution:
  • Sign up for a paid plan at pylancemcp.com
  • Add your API key to the environment configuration
  • Check usage limits in your dashboard

Next Steps