Linear

You can run a Linear MCP Server that exposes Linear’s issue tracking and project update features through a standardized MCP interface. This enables you to create, read, update, and delete issues, manage labels and assignees, search with advanced filters, and track project updates from a local or remote MCP client.

How to use

To use the Linear MCP Server, you connect an MCP client to the server to perform common Linear operations via the MCP interface. You can create issues and subissues, update fields, add comments, filter and search issues with Linear’s filtering capabilities, manage labels, and retrieve project updates. The server handles authentication using your Linear API key and provides robust error handling, rate limit awareness, and data cleaning to keep your AI context and downstream workflows efficient.

How to install

Prerequisites: you need Bun runtime (v1.0.0 or higher) and a Linear account with API access.

1. Install Bun if not already installed.

2. Clone the Linear MCP project and navigate into it.

3. Install dependencies and build the project.

4. Configure the MCP server in your local environment to connect to Linear using your API key.

5. Start the MCP server through your chosen MCP client interface and reload the client to recognize the new server.

Configuration and run details

The server is designed to run as an MCP stdio server driven by a local node process. You provide the command to start the server and any environment variables needed for authentication.

{
  "mcpServers": {
    "linear": {
      "command": "node",
      "args": ["/absolute/path/to/linear-mcp/build/index.js"],
      "env": {
        "LINEAR_API_KEY": "your_api_key"
      }
    }
  }
}

Environment and start configuration

The following environment variable is required for authentication and should be provided in the MCP server configuration: LINEAR_API_KEY. This key is used by the server to authenticate requests against Linear’s API.

Development

Development commands include running the development server and building the project. Use Bun to install dependencies, build, and run tests or a local dev server as you iterate on features.

Notes

The MCP server leverages Linear’s official SDK for type-safe operations and includes comprehensive error handling, rate limit handling, and data cleaning to optimize content for AI contexts.

Available tools

create_issue

Create a new Linear issue or subissue with labels, assignees, and optional parent relationship.

update_issue

Update an existing Linear issue with modified fields including title, description, status, priority, assignees, labels, and cycle.

get_issue

Retrieve detailed information about a specific Linear issue with optional relationship data.

search_issues

Search for Linear issues using a query string and advanced filters with support for complex logical expressions and relationships.

get_teams

Get a list of Linear teams with optional name or key filtering.

delete_issue

Delete an existing Linear issue after validation.

create_comment

Create a new comment on a Linear issue and support markdown.

get_projects

Get a list of Linear projects with filtering and pagination.

get_project_updates

Get updates for a given Linear project with optional filtering.

create_project_update

Create a new update for a Linear project with health status.