What will sink this before you build it?
Paste the brief for what you are about to build - an MVP, a SaaS product, a feature, a landing page, an internal tool or an agent workflow. Your browser checks it against seven risks: demand, positioning, monetization, retention, trust, distribution and feature adoption. It shows which ones your brief backs with evidence (line by line), which it only names, and which it never mentions; it flags the sentences that sound like evidence but are not ("everyone needs this", "huge market", "no competitors"); and it sets an evidence floor. Free, before you sign in, and nothing is uploaded. Then the review names the one assumption most likely to break the project, the smallest evidence to get first and what to delay, and the test plan turns that assumption into a test with numbers that decide it.
The companies and people in the examples are invented. Three are reviews (the invoice one also carries its test plan, one click after the review), one is a test plan; 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
The goal is not to block building. It is to find the highest-risk assumption, the smallest validation step, and the build scope that should wait until the evidence improves. The scan is pattern matching over your text: it can tell that a brief names a price but never says anyone paid, or that it plans nine features and counts zero conversations; it cannot tell whether the idea is good. The review and the test plan are held to the brief: every quote is checked against it, a risk can only be rated low with a quote, claims are never counted as evidence, and no market size, benchmark or percentage may be added that your brief does not contain. Product risk is kept apart from engineering difficulty.
The paid lanes see your brief (the middle is cut past 16,000 characters, and the page says so) and the page's scan. A brief stays in this browser unless you run a lane. Derived from the agent skill @wshobson/before-you-build by Seth Hobson (wshobson/agents, MIT): its seven-risk checklist, its deeper questions and its five-part output. Sources are listed in llms.txt.