Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when reviewing stylesheets, component styles, and responsive behavior related to Optimize web font formats. Check the rendered layout across breakpoints and interaction states before proposing a fix.
.claude/skills/thedaviddias-webfont-format/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 7 |
| gemini-3.1-pro-preview | 100% | 1 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
Modern font formats reduce file sizes by 30-50%—directly improving page load speed and Core Web Vitals scores.
Verify that web fonts use modern formats (WOFF2, WOFF) with proper fallbacks and are optimized for performance with font-display and preloading strategies.
Convert fonts to WOFF2/WOFF formats, implement proper font loading strategies, and optimize font delivery with compression and caching.
Explain how modern web font formats reduce file sizes, improve loading performance, and provide better browser support compared to legacy formats.
Review stylesheets, component styles, and responsive states related to Optimize web font formats. Flag exact selectors, declarations, or breakpoints that violate the rule in the rendered UI.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/css/webfont-format
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 13,428 | 9,054 | -33% | 1 | 1 | 0% | 2,854 | 2,178 | -24% | 0 | 0 | — |
case-02 | pass→pass | 7,026 | 6,453 | -8% | 1 | 1 | 0% | 1,296 | 1,355 | +5% | 0 | 0 | — |
case-16 | pass→pass | 15,906 | 12,308 | -23% | 1 | 1 | 0% | 2,888 | 2,439 | -16% | 0 | 0 | — |
case-03 | pass→pass | 5,759 | 12,915 | +124% | 1 | 1 | 0% | 1,071 | 1,211 | +13% | 0 | 0 | — |
case-04 | pass→pass | 4,862 | 3,853 | -21% | 1 | 1 | 0% | 832 | 999 | +20% | 0 | 0 | — |
case-05 | pass→pass | 3,859 | 4,034 | +5% | 1 | 1 | 0% | 693 | 1,035 | +49% | 0 | 0 | — |
case-06 | pass→pass | 4,819 | 6,873 | +43% | 1 | 1 | 0% | 1,054 | 1,662 | +58% | 0 | 0 | — |
case-07 | pass→pass | 14,037 | 11,316 | -19% | 1 | 1 | 0% | 2,546 | 2,403 | -6% | 0 | 0 | — |
case-08 | pass→pass | 11,099 | 6,880 | -38% | 1 | 1 | 0% | 1,894 | 1,544 | -18% | 0 | 0 | — |
case-09 | pass→pass | 4,677 | 3,943 | -16% | 1 | 1 | 0% | 878 | 933 | +6% | 0 | 0 | — |
case-10 | pass→pass | 3,464 | 3,666 | +6% | 1 | 1 | 0% | 671 | 765 | +14% | 0 | 0 | — |
case-11 | pass→pass | 12,033 | 7,711 | -36% | 1 | 1 | 0% | 2,149 | 1,591 | -26% | 0 | 0 | — |
case-12 | fail→pass | 12,145 | 4,818 | -60% | 1 | 1 | 0% | 2,246 | 1,234 | -45% | 0 | 0 | — |
case-13 | fail→pass | 8,073 | 5,073 | -37% | 1 | 1 | 0% | 1,521 | 1,360 | -11% | 0 | 0 | — |
case-14 | pass→pass | 9,483 | 7,303 | -23% | 1 | 1 | 0% | 1,668 | 1,729 | +4% | 0 | 0 | — |
case-15 | pass→pass | 12,911 | 10,484 | -19% | 1 | 1 | 0% | 2,619 | 2,259 | -14% | 0 | 0 | — |
case-17 | fail→pass | 13,866 | 10,155 | -27% | 1 | 1 | 0% | 2,404 | 2,145 | -11% | 0 | 0 | — |
case-18 | pass→pass | 10,729 | 5,468 | -49% | 1 | 1 | 0% | 1,788 | 1,340 | -25% | 0 | 0 | — |
case-19 | pass→pass | 3,242 | 3,073 | -5% | 1 | 1 | 0% | 578 | 855 | +48% | 0 | 0 | — |
case-20 | pass→pass | 12,116 | 7,606 | -37% | 1 | 1 | 0% | 2,135 | 1,683 | -21% | 0 | 0 | — |
case-21 | pass→pass | 10,131 | 6,737 | -34% | 1 | 1 | 0% | 1,776 | 1,529 | -14% | 0 | 0 | — |
case-22 | pass→pass | 13,332 | 12,199 | -8% | 1 | 1 | 0% | 2,165 | 2,274 | +5% | 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. 22 cases were attempted. The headline lift of +18 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.