Think inside your AI world.
How do I connect Linear to Claude?
'linear mcp claude' is one of the most-searched connector phrases for a reason: Linear runs an official hosted MCP server over OAuth 2.1, and the connection is genuinely quick.
Direct answer: Linear's official hosted MCP server connects over OAuth 2.1. In Claude: Settings → Connectors → add Linear, authorise, done. In Claude Code: one 'claude mcp add' and the same OAuth flow. Claude can then read issues with priority, labels, assignees and linked PRs, plus projects, cycles, milestones, initiatives, project updates and health, documents, roadmaps and releases.
What the workspace read covers
Sprint questions answer themselves from live state: what moved this cycle, which issues sit unassigned at high priority, how the milestone burndown looks, what the last project update claimed. Because linked PRs travel with issues, delivery questions stay joined-up rather than split across tools.
The rule the tracker doesn't carry
Suppose you ratified a cycle-scope rule: unplanned work stays under a fifth of any cycle, because two quarters of interrupt-driven drift once cost you a release. Linear labels the unplanned work faithfully — and holds no opinion on your fifth, since thresholds you set live outside every board that displays them.
The board, read measured
Through Unl the cycle read arrives pre-judged: unplanned at 28% of committed points — through your ceiling, which you set after the drift quarters. The stand-back question 'is this cycle healthy?' stops being a squint at the board and becomes your own rule reporting its state, with its memory intact.
Linear to Claude is an OAuth flow and a minute of setup; Linear through Unl reads the same board inside the delivery rules you ratified, so cycle health arrives as a verdict you already agreed with yourself about.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
What is the Linear MCP server?
Linear's official hosted endpoint, authenticated over OAuth 2.1, exposing the workspace read surface: issues (priority, labels, assignee, linked PRs), projects, cycles, milestones, initiatives, project updates and health, documents, roadmaps and releases.
Does Linear connect to ChatGPT and Cursor too?
Yes — it is standard MCP, so any client that takes connectors can add it. The searches 'linear mcp' and 'connect linear to chatgpt' land on the same server this page describes.
What would I ratify in Unl for delivery work?
Rules you already enforce informally: scope ceilings per cycle, a definition of launch-ready, staleness limits on in-progress issues. Ratified with their reasons, they stop living in your head and start framing every board read.
Can I just use Linear's own MCP server on its own?
Yes — for the raw read it is an excellent surface, and this page's steps need nothing else. Issues, cycles, projects and health all come back accurately over its OAuth flow. What it cannot answer is whether a cycle is in trouble, because in trouble is a line you drew — a scope ceiling, a staleness limit — and the board has never been told where you drew it or why. That is the layer Unl adds on top of the same read, not a replacement for it.
Does Unl change or store my Linear data?
Not stored: what Unl reads is quarantined for the turn and never kept — no query text and no read content reaches our telemetry. Not changed by a read, either — a read only reads. Where a server exposes a tool that writes, Unl can call it, because the socket exposes a server's full surface rather than a chosen subset; but every write carries your explicit in-turn gesture, verified at the choke point, and a tool that does not declare itself read-only is treated as a write and gated by default.
What this is
Think inside your AI world — you stay in command
Unlimitless (Unl to friends) holds what you've settled, reads what your tools are showing, and catches what's changed out in the world — and hands your AI whatever bears on the work, the moment it's needed, without you asking. The right thing, in front of the model, unprompted, with you in command of the call. So you keep moving toward what you set out to build, on top of everything you've already decided.
It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector. Quick to connect, in a couple of steps.
Unlimitless is open now to invited Alpha. Apply for the Beta waitlist to come in ahead of the full launch:
Alpha is invite-only · free at launch.