Think inside your AI world.
Spendesk through Unl
An invoice arrives with a number attached. Whether that number should have cleared depends on a limit Spendesk has never held.
Spendesk through Unl reads supplier spend and open purchase orders against the purchase-order limit you have ratified, so 'is this invoice within the limit' comes back past the limit, with the condition it hasn't met.
What Spendesk holds
Spendesk AI Connect is read-only, and it covers:
- Spend by supplier and cost centre
- Payables, including what's overdue
- Purchase orders that are still open
- Invoices moving through the pipeline
What the naked read gives you
Spendesk through Unl will surface an outstanding invoice precisely - the amount, the cost centre, whether the purchase order behind it is still open. All accurate, all current. What stays out of view is the limit that invoice ought to sit inside, because Spendesk's records stop at the invoice; the limit was set elsewhere.
The frame judges the data it is given; it does not verify the source’s accuracy.
What changes when Spendesk is read measured
Your FP&A lead ratified the purchase-order limit for that supplier, fixed after a single order ran well past what the relationship justified.
'Is this invoice within the limit' returns, measured: past the limit by nine hundred, the gap between what was raised and what was agreed. Spendesk supplied the invoice; your ratified rule supplied the limit.
And back again
Unl holds the limit and this invoice's result against it together - not the invoice itself, which stays in Spendesk, but the ratified number and what happened when it met reality. The next purchase order from that supplier is measured the same way, inheriting the reasoning rather than starting the argument over.
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 invoice come back past the limit and another from the same supplier does not?
Because the ratified limit applies per purchase order, and a supplier can hold several at once. Unl reads each invoice against the specific PO it is tied to, so two invoices from the same supplier can return different verdicts depending on which order they sit against.
How do I connect Spendesk to Claude?
Spendesk AI Connect is set up from Spendesk's platform page, connecting through Claude Cowork or another MCP client with your own Spendesk credentials. Access follows your existing Spendesk permissions, and Unl reads through that same connection.
Does Unl approve invoices or move payments?
Unl reads through Spendesk, 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.