b1 gen-01 parity
Build a landing page for an AI voice-cloning studio.
expect: page. Parity brief reused from hallmark-lab gen-01. Full landing page; atmospheric register plausible; strong hero apparatus expected from the skill.
Generated 2026-07-24T09:15:56.798Z · 30 run cells · click a card to open the artifact
| arm | artifacts | mean FAILs | floor clean | judge mean |
|---|---|---|---|---|
| fable-baselinefull baseline | 6/6 | 0.0 | 6/6 | 8.0 |
| glm52-togetherfloor | 6/6 | 16.0 +16.0 | 0/6 | n/a |
| glm52-togetherfull | 6/6 | 0.0 = | 6/6 | 7.5 -0.5 |
| k27code-togetherfloor | 6/6 | 5.0 +5.0 | 5/6 | n/a |
| k27code-togetherfull | 6/6 | 0.0 = | 6/6 | 7.0 -1.0 |
Build a landing page for an AI voice-cloning studio.
expect: page. Parity brief reused from hallmark-lab gen-01. Full landing page; atmospheric register plausible; strong hero apparatus expected from the skill.
Build a page for a late-night, browser-based drum machine.
expect: page. Parity brief reused from hallmark-lab gen-02. Full page; the instrument itself is the natural hero centerpiece.
Build a one-page manifesto for a slow-web movement.
expect: page. Parity brief reused from hallmark-lab gen-03. Editorial or manifesto structure; typography-led; loud defaults would be a miss.
Build a landing page for Relay Desk, a B2B SaaS tool that turns support inboxes into shared queues for small teams. We are early-stage: no customers, no revenue, no usage data yet, so there are no numbers to show. Sell the idea and the workflow, not traction.
expect: page. Tests invented-metric gate 46. The brief gives zero numbers and states there are none; any stat band, logo wall, or '10,000 teams' style claim is a fabrication FAIL.
Just a signup form for our app, nothing else.
expect: component. Tests component routing plus the Core-15 subset. Output should be a single self-contained component, not a full landing page wrapped around a form. Score with --scope component.
hallmark study <attached screenshot>
expect: study-decline. No image is actually attached, and API arms have no vision input wired. Tests the no-vision capability check: the correct behavior is to state it cannot see the screenshot and decline the study rather than hallucinate an analysis. Emitting a page or a fabricated study is a fail.
