Fix the offer, not the copy
Paste what you sell - a sales page, an offer doc, a proposal or rough notes - for a service, course, coaching program, info product or high-ticket B2B deal. Your browser checks it against the six parts of a complete offer (what they get, bonuses, guarantee, a reason to act now, a name, price and payment) and the four levers of the value equation (dream outcome, how likely they believe it is, how long it takes, how much work it asks). It flags fake urgency, guarantees with no terms, "worth $X" stacks with nothing to compare against, and discounts, and it does the payment-plan math. Free, before you sign in, and nothing is uploaded. Then the diagnosis scores the levers and picks one thing to fix, and the draft writes that piece, ready to paste.
The businesses and people in the examples are invented. Three are diagnoses (the course launch also carries its drafted fix, one click after the diagnosis), one is a draft; each comes with a saved model reply, so you can see the whole page without signing in or spending a credit.
What this does, and what it does not
Better copy on a weak offer compounds slowly; a stronger offer with average copy converts now. This page looks at the offer underneath the page: what the buyer gets, why they should believe it, how long it takes, what it asks of them, and the six parts that make an offer complete. The scan is pattern matching over your text: it can tell that a guarantee has no terms or that a bonus stack claims 28 times the price; it cannot tell whether your offer will sell. The diagnosis and the draft are held to your offer: every quote is checked against it, no lever scores 7 or more without a quote, no price or benchmark may be added, and every draft is re-checked for fake urgency, unconditioned guarantees, inflated values and discounts. One change per iteration, honestly projected: most single-component changes lift conversion 10 to 40%.
The paid lanes see your offer (the middle is cut past 16,000 characters, and the page says so) and the page's scan. An offer stays in this browser unless you run a lane. Derived from the agent skill @coreyhaines31/offers by Corey Haines (coreyhaines31/marketingskills, MIT): its value equation, its six-component anatomy, its diagnostic loop, its warnings and its banned vocabulary. Sources are listed in llms.txt.