Think inside your AI world.
How do I connect Ramp to ChatGPT?
Spend data answers best to questioning, and ChatGPT over MCP questions it directly — transactions, vendors, patterns, live.
Direct answer: connect Ramp's MCP server to ChatGPT. In ChatGPT, add it from the connector settings (developer mode); the tool's own docs carry the current path. Once authorised, ChatGPT reads your spend surface — transactions, vendor adoption, purchasing patterns — your organisation's slice of the data Ramp builds across 50k+ businesses.
Questions the spend data answers
Which vendors are new this quarter, what renewed without anyone deciding, where the pattern broke — asked conversationally, answered from the cards. The vendor-level view is where assistants beat dashboards: follow-up questions cost nothing.
Every vendor needs an owner
Worked case: you ratified a new-vendor rule — nothing recurring gets a card without a named owner and a renewal date on record, after an audit found four subscriptions nobody admitted to. Ramp sees every new vendor the moment it charges. Ownership is an obligation you invented, and card data has no column for obligations.
Through Unl the vendor read arrives audited: two new recurring vendors this month, one ownerless — against your rule, the one the four orphans wrote. The subscription creep audit stops being quarterly archaeology and becomes a standing property of every spend read.
Ramp to ChatGPT reads the spend; through Unl it reads the spend against the vendor discipline you ratified — orphan subscriptions get caught at the first charge, not the fourth audit.
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 a Ramp connection?
Transactions, vendor adoption and purchasing patterns — the observational spend surface, identical to what any MCP client reads.
Does this replace Ramp's own controls?
No — Ramp's approval flows and limits keep doing their job. The conversational read adds questioning; Unl adds your ratified rules to the questioning.
What is a good first spend rule to ratify?
The new-vendor owner rule pays off immediately: it is cheap to honour, catches creep at the source, and its why — the orphan audit — makes it self-explaining forever.
What if a charge is legitimate but nobody will own it?
Then it is not an orphan by your rule, and it is still a problem. The rule catches the shape it was written for; an unowned but justified cost is a different shape and needs its own call. A second rule is usually better than widening the first until it catches everything and means 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.