Size the loan to the binding constraint
Rung 8 of 10 · Cognitive verifier
1 · Learn the move · Cognitive verifier
A loan is only as big as its tightest limit lets it be. Max LTV, minimum DSCR, minimum debt yield. The trap is letting the chat 'do the math': the mortgage constant off a rate and amortization, then three constrained loan amounts, is exactly where an LLM confidently fabricates a number. The cognitive-verifier move flips that. The AI states which constraint it EXPECTS to bind and why, sets up all three formulas with your pasted inputs, and hands YOU the arithmetic to run in a real calculator. It never asserts a final loan dollar figure as computed. That's not caution, it's the only defensible way to size a deal you'll put in front of credit committee.
What's the max loan on this deal? [PASTE: NOI, purchase price, max LTV, min DSCR, min debt yield, rate, amortization]
2 · Your turn. You write the prompt
A bridge originator has a deal at $6.4M with $410K NOI and a 6.75% rate. The credit box is 70% LTV, 1.25x DSCR, 9% debt yield. The analyst needs the maximum loan under each test and, more importantly, which test actually caps the deal. Without trusting a chatbot's mortgage-constant arithmetic.
Remember: the AI sees only your prompt, not this page. If the situation isn't in your prompt, it doesn't exist.
Optional. These shape the output when you run your prompt below, not your score.