Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Turn a reference image, screenshot, or mockup into token-driven, accessible code — infer the design system from the reference (palette, type scale, spacing, radius, layout archetype), map it to the 3-tier tokens, rebuild it, then verify with the kit's gates. Use when the user provides a design/screenshot and wants matching UI code.
.claude/skills/plugin87-image-to-code/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -31% | 0% |
Reconstruct a design from a visual reference as a real design system, not a one-off copy. Match the system (color/type/spacing language), never lift copyrighted imagery or brand assets.
taste/design-taste.md → Variance Mandate).taste/aesthetic-systems.md / python3 scripts/design_systems.py search <term> and adopt that recipe to stabilize decisions.design-tokens skill); generate a single theme.css. Verify every color pair with scripts/contrast.py / scripts/validate_contrast.py (light + dark) — a sampled brand color that fails AA gets adjusted; taste never overrides POUR.frameworks/adapter-protocol.md + components/*: one shared primitive layer, all 8 states, a11y wired, no emoji (lucide), single theme. Apply taste (design-taste.md) so it doesn't regress to generic.node scripts/measure_render.mjs, lint_hardcodes.py, taste_audit.mjs, and npm run verify.npm run verify is 100% (tokens resolve, contrast AA light+dark, no hardcodes/emoji, real-render WCAG).> Honest limit: this matches the design system, not a pixel-perfect copy. Do not reproduce the reference's photographs, logos, or copyrighted copy — substitute your own or generic placeholders.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-14 | pass→pass | 15,017 | 12,781 | -15% | 1 | 1 | 0% | 2,302 | 2,526 | +10% | 0 | 0 | — |
case-01 | fail→fail | 29,467 | 30,109 | +2% | 1 | 1 | 0% | 6,211 | 6,807 | +10% | 0 | 0 | — |
case-15 | pass→pass | 9,427 | 8,050 | -15% | 1 | 1 | 0% | 1,381 | 1,769 | +28% | 0 | 0 | — |
case-02 | fail→fail | 29,950 | 6,279 | -79% | 1 | 1 | 0% | 6,204 | 841 | -86% | 0 | 0 | — |
case-03 | fail→fail | 25,863 | 27,441 | +6% | 1 | 1 | 0% | 6,206 | 6,802 | +10% | 0 | 0 | — |
case-04 | fail→fail | 9,642 | 8,270 | -14% | 1 | 1 | 0% | 1,627 | 1,832 | +13% | 0 | 0 | — |
case-05 | pass→pass | 11,452 | 4,450 | -61% | 1 | 1 | 0% | 1,751 | 1,259 | -28% | 0 | 0 | — |
case-16 | pass→pass | 11,556 | 7,409 | -36% | 1 | 1 | 0% | 1,843 | 1,853 | +1% | 0 | 0 | — |
case-06 | fail→pass | 13,564 | 8,243 | -39% | 1 | 1 | 0% | 2,090 | 1,855 | -11% | 0 | 0 | — |
case-07 | fail→pass | 12,834 | 8,368 | -35% | 1 | 1 | 0% | 1,913 | 1,960 | +2% | 0 | 0 | — |
case-08 | fail→pass | 11,796 | 5,087 | -57% | 1 | 1 | 0% | 1,876 | 1,327 | -29% | 0 | 0 | — |
case-09 | pass→pass | 10,394 | 3,464 | -67% | 1 | 1 | 0% | 1,624 | 1,084 | -33% | 0 | 0 | — |
case-10 | pass→pass | 11,630 | 5,250 | -55% | 1 | 1 | 0% | 1,801 | 1,272 | -29% | 0 | 0 | — |
case-11 | pass→pass | 20,337 | 10,378 | -49% | 1 | 1 | 0% | 3,213 | 2,321 | -28% | 0 | 0 | — |
case-12 | pass→pass | 11,184 | 9,129 | -18% | 1 | 1 | 0% | 1,766 | 1,884 | +7% | 0 | 0 | — |
case-13 | pass→pass | 13,821 | 11,087 | -20% | 1 | 1 | 0% | 2,127 | 2,424 | +14% | 0 | 0 | — |
case-17 | pass→pass | 10,912 | 3,807 | -65% | 1 | 1 | 0% | 1,686 | 1,169 | -31% | 0 | 0 | — |
case-18 | fail→fail | 15,295 | 5,868 | -62% | 1 | 1 | 0% | 2,250 | 1,503 | -33% | 0 | 0 | — |
case-19 | fail→pass | 14,449 | 8,138 | -44% | 1 | 1 | 0% | 2,370 | 1,929 | -19% | 0 | 0 | — |
case-20 | fail→pass | 8,069 | 2,387 | -70% | 1 | 1 | 0% | 1,351 | 928 | -31% | 0 | 0 | — |
case-21 | pass→pass | 31,859 | 31,261 | -2% | 1 | 1 | 0% | 6,173 | 6,769 | +10% | 0 | 0 | — |
case-22 | pass→pass | 14,161 | 14,850 | +5% | 1 | 1 | 0% | 2,688 | 3,472 | +29% | 0 | 0 | — |
case-23 | pass→pass | 18,589 | 20,009 | +8% | 1 | 1 | 0% | 3,940 | 4,495 | +14% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +22 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.