Linear MCP Server

Linear MCP Server

Linear MCP Server is a Model Context Protocol (MCP) integration for Linear, enabling language models to interact programmatically with Linear's issue tracking system. It allows AI tools and agents to create, update, search, and manage Linear issues and comments, as well as query user and team information, through standardized protocol endpoints.

Author: jerhadf


View Protocol

What is Linear MCP Server?

Linear MCP Server is a lightweight server that acts as a bridge between AI applications (such as Claude Desktop) and the Linear issue tracking API. By adhering to the Model Context Protocol, it exposes Linear functionalities—like issue creation, updates, comments, and search—as tools and resources that are easily accessible by large language models. This streamlines workflows, automates project management, and enhances productivity by integrating AI seamlessly with Linear.

How to Configure

  1. Obtain a Linear API key by visiting your Linear team’s API settings page.
  2. For integration with Claude Desktop, add the following configuration in your claude_desktop_config.json (located at ~/Library/Application Support/Claude/ on MacOS):
{
  "mcpServers": {
    "linear": {
      "command": "npx",
      "args": [
        "-y",
        "linear-mcp-server"
      ],
      "env": {
        "LINEAR_API_KEY": "your_linear_api_key_here"
      }
    }
  }
}
  1. Alternatively, install automatically via Smithery with:
npx @smithery/cli install linear-mcp-server --client claude
  1. Ensure your LINEAR_API_KEY is securely provided via environment variables as shown in the configuration.

How to Use

After configuring and starting the server:

  • Use your AI application (e.g., Claude Desktop) to query Linear via natural language. The app will automatically discover available Linear tools and resources through MCP.
  • Example actions include creating new issues, updating issue status, searching for tasks by keyword or priority, retrieving assigned issues, or adding comments.
  • Prompts like "Show me all my high-priority issues" or "Create a bug report for login failures" will trigger the appropriate MCP tool invocation.
  • Developers can directly use MCP endpoints such as tools/call to invoke Linear functions, or browse available options using tools/list.

Key Features

  • Seamless connectivity between AI applications and Linear via the Model Context Protocol.
  • No custom integration coding required; all major issue operations are available as standardized tools.
  • Predefined tools for issue creation, updates, searches, user queries, and commenting.
  • Supports secure local environment configuration, keeping API keys within your infrastructure.
  • Resource endpoints enable direct retrieval or aggregation of Linear entities such as issues and team task lists.
  • Open Source (MIT License) for modification and self-hosting.

Use Cases

  • Automating the creation, triage, and tracking of engineering tickets from plain English bug reports or feature requests.
  • Quickly surfacing assigned tasks or high-priority issues for team members during standups.
  • Enabling AI agents to monitor, summarize, and update progress on ongoing sprints directly inside Linear.
  • Fast, filtered search for issues by team, assignee, label, or custom criteria using natural language.
  • Adding comments or clarifications to Linear issues as part of conversational workflows.
  • Analyzing workload and backlog distribution across different teams or users.

FAQ

Q1: Is Linear MCP Server still maintained?
A1: No, this project is deprecated. For the latest official Linear MCP integration, refer to https://linear.app/changelog/2025-05-01-mcp or https://mcp.linear.app/sse.

Q2: Can I use Linear MCP Server with other LLM tools besides Claude Desktop?
A2: Yes, you can use it with any Model Context Protocol compatible host, though the provided config examples target Claude Desktop specifically.

Q3: Do I need to write code to interact with Linear MCP Server?
A3: Not necessarily. If your AI application supports MCP, you can interact with Linear via natural language. Developers can invoke tool endpoints directly if desired.

Q4: Where should I store my Linear API key?
A4: Store your API key in environment variables or your application's configuration file as shown. Do not share or expose it publicly.

Q5: What happens if my API key expires or is revoked?
A5: The MCP server will lose access to Linear and all tool invocations will fail until a valid API key is provided.