Overview
An MCP server implementing the Linear GraphQL API to enable AI assistants to interact with Linear as a project management system. MCP Linear exposes typical MCP capabilities to retrieve data (issues, projects, teams) and perform actions such as creating and updating issues, changing issue status, assigning issues to teammates, adding comments, and creating projects and teams. It relies on a Linear API token for authentication and can be deployed via Smithery (recommended) or configured manually in your MCP settings. The README provides installation steps for Cursor and Claude Desktop, example manual configuration blocks, and environment-variable usage (LINEAR_API_TOKEN). You can install globally with npm and run with mcp-linear --token YOUR_LINEAR_API_TOKEN, or export LINEAR_API_TOKEN and run mcp-linear. The server supports per-client configurations and includes guidance on client-specific config locations. Additional documentation is linked (TOOLS.md and DEVELOPMENT.md) for a full list of available tools and local development notes. The project is MIT-licensed and focuses on bridging AI workflows with Linear through the MCP protocol.
Features
Retrieve data from Linear (issues, projects, teams, and more)
Fetch Linear data via the MCP server, including issues, projects, teams, and related entities.
Create and update issues
Create new issues and update existing ones in Linear through MCP prompts.
Change issue status
Update the status of Linear issues (e.g., In Progress, Done) via MCP.
Assign issues to team members
Assign Linear issues to teammates directly from MCP interactions.
Add comments
Add comments to Linear issues to capture notes and updates.
Create projects and teams
Create new projects and teams within Linear through the MCP server.
