Linear MCP server: setup, auth, production

What is the Linear MCP server?

A Linear MCP server implements the Model Context Protocol and exposes Linear's capabilities as typed tools an agent can discover and call in the middle of a task. Issue tracking and project planning for engineering teams.

Connecting the tracker closes the loop between a code change and the work item behind it: the agent reads the spec it is implementing and updates status without a human relaying it. OAuth per user and team scoping keep it honest.

A real Linear workflow

The agent opens the Linear issue it’s working, reads the acceptance criteria and linked designs, implements against them, then updates the issue to In Review with a summary — so the tracker reflects reality without anyone relaying it.

What an agent can do with Linear

• Let an agent create and update issues from a code change
• Pull the spec and acceptance criteria for the issue it’s working on
• Summarize the current sprint or a project’s status

Read-only is the right default: let the agent observe and reason first, then grant specific write actions deliberately — each behind audit logging and, for anything high-impact, human approval.

Setting up the Linear MCP server

Prefer Linear OAuth over a personal API key so actions are attributable to a user and scoped to the right teams — a personal key inherits that person’s full access, which is rarely what you want for an agent.

Auth: Linear OAuth or a personal API key, scoped to the right teams. Credentials belong in the environment or a secret store — never in the repo or the agent's sandbox. Wiring a server into Claude Code or Cursor is the same for every tool; the step-by-step lives in the MCP connection guide.

The risk to watch with Linear

A personal API key gives the agent one human’s entire Linear footprint. Use OAuth and team scoping, and log agent-created issues so the board doesn’t fill with unattributed noise.

More broadly, a production-grade Linear integration needs oAuth per user, team scoping, and audit logs for agent-created issues. The MCP security checklist applies to every integration — we build servers with these controls baked in.