Think inside your AI world.
Expensify through Unl
A receipt is a fact. Whether it belongs under the cap your organisation ratified is a judgement Expensify was never built to make.
Expensify through Unl reads expenses, reports and receipts against the per-head spend cap you have ratified, so 'is this claim under the cap' comes back over the cap, with the condition it hasn't met.
What Expensify holds
Expensify's MCP server exposes a single read-only Search tool, scoped to your existing permissions, that reaches:
- Expenses, queried in real time
- Reports, in whatever state they're sitting
- Invoices, tied to a vendor or a period
- Receipts, the images and the data attached to them
What the naked read gives you
Search a client dinner through Expensify and Unl returns the receipt, the amount, the report it belongs to, in real time. Every detail checks out. Missing is any sense of whether that amount belongs under a cap, since Expensify's job stops at the receipt - the cap itself was decided somewhere it never had access to.
The frame judges the data it is given; it does not verify the source’s accuracy.
What changes when Expensify is read measured
Your finance manager ratified the per-head cap on client entertainment, set after two quarters of dinners that crept past what the budget could absorb.
'Is this claim under the cap' returns, measured: over the cap by eighteen pounds, the amount the claim exceeds it by. Expensify supplied the receipt; your ratified rule supplied the cap.
And back again
The cap, the reason finance drew the line where they did, and how this particular dinner measured against it become one entry inside Unl rather than three separate facts. Anyone reviewing spend later inherits the ratified cap and the why behind it, not just a pass or fail on a receipt.
The answer comes back measured against what you already decided, and why.
The lane is live and open to this tool today: 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.
Questions people ask
Why does one dinner receipt come back over the cap and a similar one does not?
Because the cap is ratified per category and sometimes per team, not as one figure for every claim. Unl reads each receipt against the cap that actually applies to it, so two dinners of a similar size can return different verdicts depending on which budget line they sit under.
How do I connect Expensify to Claude?
Expensify's MCP server is hosted at expensify.com/mcp. Authorise it from Claude with a one-time OAuth 2.1 login to your Expensify account, no API key or export needed. Unl then reads through the same read-only Search tool Expensify exposes.
Does Unl approve expenses or move money?
Unl reads through Expensify, and can write back on your explicit gesture — it never acts as a side effect of a read.
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.