MCP ConfigsApr 6, 2026·2 min read

Linear MCP — Project Management for AI Agents

MCP server that connects AI agents to Linear for issue tracking, project management, and workflow automation. Create issues, update status, and query projects via natural language. 1,500+ stars.

TL;DR
Linear MCP lets AI agents create issues, update status, and query projects in Linear.
§01

What it is

Linear MCP is a Model Context Protocol server that connects AI agents to Linear, the project management tool. Through this MCP integration, an AI assistant like Claude Code or Cursor can create issues, update statuses, query projects, and manage workflows in Linear using natural language commands.

This integration targets development teams already using Linear who want their AI coding assistants to interact with the issue tracker without switching context. Instead of opening Linear in a browser, you tell your AI agent to create a bug report or check sprint progress.

§02

Why it saves time or tokens

Context switching between an AI coding tool and a project management UI breaks flow. Linear MCP eliminates this by letting the agent read and write issues directly. Creating an issue from a code review finding, linking a commit to a ticket, or checking what is assigned to you all happen in the same conversation. This reduces the overhead of manual copy-paste between tools.

§03

How to use

  1. Install the Linear MCP server in your AI tool's MCP configuration
  2. Authenticate with your Linear API key
  3. Ask your AI agent to create issues, query projects, or update statuses in natural language
§04

Example

{
  'mcpServers': {
    'linear': {
      'command': 'npx',
      'args': ['-y', '@anthropic/linear-mcp-server'],
      'env': {
        'LINEAR_API_KEY': 'lin_api_xxxxx'
      }
    }
  }
}

With this configuration, your AI agent can execute commands like 'create a high-priority bug in the Backend project titled API timeout on large payloads' and Linear MCP translates that into the correct API call.

ActionWhat the Agent Does
Create issueCalls Linear API with title, description, priority, team
Update statusMoves issue to In Progress, Done, etc.
Query projectLists issues by filter: assignee, status, label
Add commentPosts a comment on an existing issue
§05

Related on TokRepo

§06

Common pitfalls

  • The Linear API key grants full access to your workspace; use a service account with limited permissions rather than your personal key
  • MCP servers run locally and need network access to api.linear.app; corporate firewalls may block this
  • Natural language commands can be ambiguous about which team or project to target; be specific when you have multiple teams in Linear

Frequently Asked Questions

What AI tools work with Linear MCP?+

Linear MCP works with any AI tool that supports the Model Context Protocol, including Claude Code, Claude Desktop, and Cursor. The MCP server runs as a local process that these tools connect to via the standard MCP protocol. Any tool implementing the MCP client specification can use it.

Do I need a Linear paid plan to use Linear MCP?+

You need a Linear account with API access. Linear provides API keys on all plans including their free tier. The MCP server itself is open source and free. The only requirement is a valid LINEAR_API_KEY with sufficient permissions for the operations you want to perform.

Can Linear MCP create issues with custom fields?+

Yes, Linear MCP can set standard fields like title, description, priority, assignee, labels, and project. Support for custom fields depends on the MCP server implementation version. Check the server documentation for the specific fields available in your version.

Is Linear MCP secure for team use?+

Linear MCP runs locally on your machine and communicates directly with Linear's API over HTTPS. Your API key stays in your local MCP configuration file. The server does not send data to any third party. However, treat your API key as a secret and do not commit it to version control.

How does Linear MCP compare to Linear's native integrations?+

Linear's native integrations (GitHub, Slack, Figma) are event-driven: they react to webhooks. Linear MCP is agent-driven: an AI assistant actively creates and queries issues on demand. They complement each other. Use native integrations for automated workflows and MCP for interactive AI-assisted project management.

Citations (3)
🙏

Source & Thanks

Built on Linear API. Licensed under MIT.

mcp-linear — ⭐ 1,500+

Thanks for connecting AI agents to where the work gets tracked.

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.