MCP Linear Server

🚀 Linear MCP Integration Server

This server provides linear integration capabilities through the Model Context Protocol (MCP). It enables AI models to interact with Linear for issue tracking and project management.

✨ Features

The server offers the following tools via the MCP interface:

linear_create_issue

Create a new Linear issue, supporting the following parameters:

• title (required): Issue title
• teamId (required): ID of the team where the issue will be created
• description (optional): Issue description
• labels (optional): List of issue labels

linear_list_issues

List all issues in a specified team, supporting the following parameters:

• teamId (required): ID of the target team
• filter (optional): Filter criteria (e.g., status, labels, etc.)

linear_get_issue

Get detailed information about a single issue, supporting the following parameters:

• teamId (required): ID of the target team
• issueId (required): ID of the issue to retrieve

linear_update_issue

Update the information of an existing issue, supporting the following parameters:

• teamId (required): ID of the target team
• issueId (required): ID of the issue to update
• title (optional): New title
• description (optional): New description
• labels (optional): New list of labels

linear_delete_issue

Delete a specified issue, supporting the following parameters:

• teamId (required): ID of the target team
• issueId (required): ID of the issue to delete

📦 Installation

Development Setup

1. Clone the repository.
2. Install dependencies:

npm install

3. Create a .env file and add your Linear API key:

LINEAR_API_KEY=your_api_key_here

4. Run the server:

npm run dev

🔧 Technical Details

Project Architecture

• Written in TypeScript to ensure strong typing and maintainability of the code.
• Built on the Linear SDK, providing encapsulation and support for the Linear API.
• Works with the MCP protocol to achieve seamless interaction with AI models.

Core Features

• Create, read, update, and delete issues.
• Isolate issues and control permissions between teams.
• Support custom filters and sorting logic.
• Provide detailed error handling and logging.

📈 Performance and Reliability

Caching System

• Use in-memory caching to accelerate frequently accessed operations.
• Set reasonable expiration times to balance consistency and performance.
• Support manual cache refresh to ensure data consistency.

Error Handling

• Comprehensive error capture and handling mechanism.
• Friendly error message feedback.
• Automatic retry strategy to handle temporary failures.

Performance Optimization

• Manage thread pools and connection pools.
• Balance the load of concurrent requests.
• Optimize data access patterns for efficiency.
• Use pagination to handle large data sets.

📚 Dependencies

• @linear/sdk: Linear API client library
• @modelcontextprotocol/sdk: MCP server implementation
• zod: Runtime type checking and validation
• dotenv: Environment variable management tool
• TypeScript development suite and related tools

For the full list of dependencies, please refer to package.json.

🚀 Smithery Deployment

This server can be deployed on Smithery.ai using the provided configuration files:

Prerequisites

• A Smithery account
• A Linear API key

Deployment Steps

1. Add this repository to Smithery or claim an existing service.
2. Access the deployment tab (only available to authenticated owners).
3. Configure the deployment parameters and set the Linear API key.
4. Publish the server.

Configuration Files

Dockerfile

Defines the server's build process:

• Uses Node.js 18 Alpine as the base image.
• Installs dependencies and builds the TypeScript code.
• Sets the running command.

smithery.yaml

Configures the server startup parameters:

• Specifies the running command.
• Defines the required configuration parameters (Linear API key).
• Manages environment variables.

Local Testing

1. Build the Docker image:

docker build -t linear-mcp-server .

2. Run the container and set the environment variables:

docker run -e LINEAR_API_KEY=your_api_key_here linear-mcp-server

3. Or use the MCP inspection tool:

npm run inspect

In this way, you can successfully deploy the server to Smithery and conduct local tests.