Think inside your AI world.
How do I connect Linear to ChatGPT?
Linear's official hosted server authenticates over OAuth 2.1 and serves any MCP client — ChatGPT gets the same workspace read Claude does.
Direct answer: connect Linear's official hosted MCP server to ChatGPT. In ChatGPT, add it from the connector settings (developer mode); the tool's own docs carry the current path. Authorise over OAuth 2.1 and ChatGPT reads issues with their priorities, labels, assignees and linked PRs; projects, cycles and milestones; initiatives, updates and health; documents, roadmaps and releases.
The workspace, from either assistant
OpenAI's own prompt library points users at exactly these questions — ticket progress, PR status — which tells you where the demand sits. Connected, the answers come from the board's live state: what moved, what stalled, what the milestone actually contains.
Triage is a promise with a clock
Worked case: your team ratified a P1 rule — anything marked urgent gets triaged within twenty-four hours, priced by a P1 that once sat mislabelled over a weekend and cost a customer. Linear timestamps everything; the twenty-four-hour promise is yours, and clocks don't run on promises they haven't been told about.
Through Unl the issue read arrives with the clock running: one P1 at nineteen hours untriaged — five hours from breaching the rule the lost-customer weekend wrote. The escalation happens inside the read, on whichever surface asked.
Linear to ChatGPT is the official hosted server in another client; through Unl the board is read under the delivery promises you ratified — the P1 clock counts down in the answer itself.
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
Is the ChatGPT read surface the same as Claude's?
Identical — issues, projects, cycles, milestones, initiatives, updates, documents, roadmaps, releases. One server over OAuth 2.1, many clients.
Why do assistants keep suggesting ticket-progress questions?
Because the demand is real: platform prompt libraries teach users to ask about PRs and tickets. The naked read answers them; a ratified rule set makes the answers verdict-shaped.
Does Unl care which assistant I ask from?
No. It is fullest on Claude and Claude Code, where it is tuned, and reaches any MCP surface — the ratified promise applies to every read wherever you work.
What if a P1 gets downgraded while its clock is running?
Then the clock stops meaning what it meant, and the honest record is that the priority changed rather than that the promise was kept. Keep the downgrade as its own decision with its reason. Otherwise your delivery record improves every time you reclassify something, which is the one way to hit every target and learn nothing.
What this is
Think inside your AI world — you stay in command
Save the thoughts, decisions and targets worth keeping, each with its reasoning, carried into every AI session the moment they matter. A new unit of exchange between you and your AI: the Settled Why with standing that travels. Unprompted.
Your whole AI world. What you decided at the epicentre. It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector — quick to connect, in a couple of steps.
MCP native·Human settled·Model agnostic·Your data
Measured Context
Connect Unl to bring the right information into the moment.
Your sources, read against the criteria you set.
The full product, open. Free at launch.