Measured convergence before: split hero 13/19, top-bar nav 19/19,
one-line notice footer 14/19, body Pretendard 17/21, 10 similarity
pairs above 0.78. Each showcase got a pre-assigned hero family, nav
form, footer form, type pair, background, accent hue, radius, motion
grammar and material from outputs/rework-2026-10-06/plan.md; direction
cards and the chosen reason are recorded in every design.md.
After: 0 similarity warnings, 19 distinct display faces, 7 hero
families, 5 nav forms, 5 footer forms, no scroll reveals, 5 showcases
with SVG filter materials. Example contact values replace bracket
placeholders under the existing fictional-brand notices. Thumbnails,
OG images, card copy and three e2e assertions updated; dead CSS in
dure-enrollment and hannam-jari removed; similarity.mjs learns the new
font families.
- Restore the step-0 interview as a mechanical gate the skill explicitly
depends on; add harness.md (per-harness question tools, limits,
fallbacks) and brief-interview.md (slots, question cards, rounds).
- Add 10 reference docs absorbed from external design skills
(accessibility, interaction-feel, elevation, color, icons, product-copy,
component-systems, critique, change-review, print-email) and extend
existing references.
- Add a review-only route and two hard-gate clauses (truncated content
reachability, three-flashes limit).
- design-gate: split tap targets into WCAG 2.5.8 and 44px contract layers,
run axe-core when available, and fix false positives found on a real
site (decorative alt="", stacked wordmark line count, url-only pages).
- lint-skill: fail if the interview gate section or its links disappear.
- Ship agents/openai.yaml and THIRD_PARTY_NOTICES.md.
