We had an AI build two storefronts. Its self-review passed. Its browser QA caught 4 bugs.

Description and use

<p>We had an AI agent build two fictional storefronts one afternoon: a Warsaw watch manufacture and a tiny type foundry. It wrote the pages, ran them in a browser, declared its own work clean. Then we did what most AI-generated work never gets — we graded it.</p> <p>We call it an acceptance gate: a fixed pass/fail checklist executed against the live page at three breakpoints. Not "make it look premium" — a rubric the model can't charm.</p> <p>It caught four things the same model had already signed off on:</p> <ol> <li> <strong>A hero where a table belonged.</strong> The watch page opened with

Problem

<p>We had an AI agent build two fictional storefronts one afternoon: a Warsaw watch manufacture and a tiny type foundry. It wrote the pages, ran them in a browser, declared its own work clean. Then we did what most AI-generated work never gets — we graded it.</p> <p>We call it an acceptance gate: a fixed pass/fail checklist executed against the live page at three breakpoints. Not "make it look premium" — a rubric the model can't charm.</p> <p>It caught four things the same model had already signed off on:</p> <ol> <li> <strong>A hero where a table belonged.</strong> The watch page opened with

Outcome

<p>We had an AI agent build two fictional storefronts one afternoon: a Warsaw watch manufacture and a tiny type foundry. It wrote the pages, ran them in a browser, declared its own work clean. Then we did what most AI-generated work never gets — we graded it.</p> <p>We call it an acceptance gate: a fixed pass/fail checklist executed against the live page at three breakpoints. Not "make it look premium" — a rubric the model can't charm.</p> <p>It caught four things the same model had already signed off on:</p> <ol> <li> <strong>A hero where a table belonged.</strong> The watch page opened with

Evidence

We had an AI build two storefronts. Its self-review passed. Its browser QA caught 4 bugs.

Source