Think inside your AI world.
How do I connect Attio to ChatGPT?
Attio's structured relationship graph serves any MCP client — ChatGPT searches and reads it like the database it secretly is.
Direct answer: connect Attio's MCP server to ChatGPT. In ChatGPT, add it from the connector settings (developer mode); the tool's own docs carry the current path. Authorised, ChatGPT searches and reads your workspace — people, companies, interactions — and manages records where you allow it.
The graph, asked directly
Who went quiet after the second call, which companies touched us twice this quarter, what actually happened on the account — retrieved from the graph in conversation, wherever the question occurs to you.
Intros are debts until closed
Worked case: you ratified an intro-debt rule — every introduction you accept gets a close-the-loop note to the introducer within a fortnight, because a forgotten thank-you once cooled your best referrer for a year. Attio records the intro and the silence equally; the fortnight is a debt schedule only you hold.
Through Unl the relationship read arrives with the ledger open: two intros past the fortnight unclosed — debts, by the rule the cooled referrer wrote. Relationship capital stops leaking through the crack between memory and record.
Attio to ChatGPT reads the relationship graph; through Unl it reads the graph against the relationship debts you ratified — the unclosed loop surfaces before it costs a referrer.
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 does ChatGPT get from Attio?
Workspace search and reads over people, companies and interactions, plus record management where authorised — the same MCP surface every client shares.
Does the measured layer need Attio-specific setup?
No — rules live in Unl and apply to whatever read touches them. Ratify the fortnight once; it governs every workspace read from any client.
What other relationship rules earn their keep?
Re-approach windows, response-time norms for key accounts, a qualification bar for partner time. Small rules, real incidents, served at read time.
What if a relationship is quiet by agreement rather than neglect?
Then it is not a debt and the rule should not call it one. Mark the agreement where the rule can see it, or you will be reminded to chase someone who asked not to be chased. A standard is only as good as the exceptions somebody bothered to record.
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.