GitHub - anoncam/linear-mcp: A Linear MCP implementation that handles all Linear resource types. (original) (raw)

Linear MCP Server

A Model Context Protocol (MCP) server for Linear, providing AI assistants with access to Linear's project management capabilities.

Features

Setup

Prerequisites

Installation

  1. Clone the repository:
    git clone https://github.com/yourusername/linear-mcp.git
    cd linear-mcp
  2. Install dependencies:
  3. Create a .env file with your Linear API key:
LINEAR_API_KEY=your_linear_api_key_here  
  1. Build the project:

Usage

Command-line (stdio)

To run the server with stdio transport (for use with CLI tools that support MCP):

Or:

HTTP Server (for remote connections)

To run the server as an HTTP service with Server-Sent Events (SSE):

This will start an HTTP server on port 3000 (configurable via PORT environment variable).

Development Mode

To run in development mode with automatic reloading:

Claude Desktop Integration

To use this server with Claude Desktop:

  1. Build the project:
  2. In Claude Desktop, go to Settings → Advanced → MCP Configuration.
  3. Add the following configuration (adjust paths to match your installation):
    {
    "mcpServers": {
    "linear": {
    "command": "node",
    "args": [
    "/path/to/linear-mcp/dist/index.js"
    ],
    "env": {
    "LINEAR_API_KEY": "your_linear_api_key_here"
    }
    }
    }
    }
  4. Save the configuration and restart Claude Desktop.

Alternatively, you can copy the provided claude-desktop-config.json file and modify the paths to match your installation.

Resources

The server provides access to all major Linear entities as resources:

Core Resources

Additional Resources

Specialized Resources

Tools

The server provides tools for:

Prompts

The server provides prompt templates for:

Testing

Test with the MCP Inspector:

npx @modelcontextprotocol/inspector stdio -- npm start

Or, if running in HTTP mode, open the MCP Inspector in your browser and connect to your server's URL.

Environment Variables

License

MIT