Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when reviewing templates, rendered HTML, or shared components related to Use semantic HTML elements. Validate the final browser-facing markup, not just the source framework abstraction.
.claude/skills/thedaviddias-html5-semantic-elements/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -24% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -4% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -29% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -7% | 0% |
Screen readers use semantic elements to navigate pages—users can jump directly to <main> content or <nav>. Search engines also use semantics to understand content hierarchy.
Review this HTML structure to ensure it uses appropriate semantic elements like header, main, section, article, aside, and footer instead of generic div elements.
Replace generic div elements with semantic HTML5 elements that accurately describe the content structure and purpose.
Explain how semantic HTML elements improve accessibility, SEO, and code maintainability by providing meaningful structure to web content.
Review templates, server-rendered HTML, and shared components that output markup related to Use semantic HTML elements. Flag exact elements, attributes, and routes where the rendered HTML violates the rule.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/html/html5-semantic-elements
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 7,120 | 4,201 | -41% | 1 | 1 | 0% | 1,574 | 1,199 | -24% | 0 | 0 | — |
case-02 | pass→pass | 10,835 | 9,194 | -15% | 1 | 1 | 0% | 2,054 | 1,963 | -4% | 0 | 0 | — |
case-03 | pass→pass | 9,901 | 5,386 | -46% | 1 | 1 | 0% | 1,863 | 1,331 | -29% | 0 | 0 | — |
case-04 | pass→pass | 8,177 | 5,871 | -28% | 1 | 1 | 0% | 1,602 | 1,486 | -7% | 0 | 0 | — |
case-05 | pass→pass | 5,726 | 4,278 | -25% | 1 | 1 | 0% | 1,060 | 1,065 | +0% | 0 | 0 | — |
case-06 | pass→pass | 3,191 | 2,959 | -7% | 1 | 1 | 0% | 594 | 872 | +47% | 0 | 0 | — |
case-07 | pass→pass | 6,883 | 5,158 | -25% | 1 | 1 | 0% | 1,404 | 1,313 | -6% | 0 | 0 | — |
case-08 | pass→pass | 7,690 | 5,478 | -29% | 1 | 1 | 0% | 1,624 | 1,368 | -16% | 0 | 0 | — |
case-09 | pass→pass | 12,264 | 9,337 | -24% | 1 | 1 | 0% | 2,439 | 2,157 | -12% | 0 | 0 | — |
case-10 | pass→pass | 12,619 | 12,129 | -4% | 1 | 1 | 0% | 2,236 | 2,303 | +3% | 0 | 0 | — |
case-11 | pass→pass | 6,508 | 5,439 | -16% | 1 | 1 | 0% | 1,295 | 1,305 | +1% | 0 | 0 | — |
case-12 | pass→pass | 7,304 | 5,011 | -31% | 1 | 1 | 0% | 1,438 | 1,267 | -12% | 0 | 0 | — |
case-21 | fail→pass | 8,641 | 5,274 | -39% | 1 | 1 | 0% | 1,631 | 1,309 | -20% | 0 | 0 | — |
case-13 | pass→pass | 4,258 | 3,576 | -16% | 1 | 1 | 0% | 1,008 | 1,003 | -0% | 0 | 0 | — |
case-14 | pass→pass | 6,625 | 4,141 | -37% | 1 | 1 | 0% | 1,262 | 1,195 | -5% | 0 | 0 | — |
case-15 | pass→pass | 12,710 | 7,455 | -41% | 1 | 1 | 0% | 2,080 | 1,654 | -20% | 0 | 0 | — |
case-16 | pass→pass | 5,876 | 3,696 | -37% | 1 | 1 | 0% | 1,087 | 1,031 | -5% | 0 | 0 | — |
case-17 | pass→pass | 6,219 | 3,293 | -47% | 1 | 1 | 0% | 982 | 927 | -6% | 0 | 0 | — |
case-18 | pass→pass | 8,380 | 5,838 | -30% | 1 | 1 | 0% | 1,742 | 1,494 | -14% | 0 | 0 | — |
case-19 | pass→pass | 8,440 | 6,477 | -23% | 1 | 1 | 0% | 1,546 | 1,431 | -7% | 0 | 0 | — |
case-20 | pass→pass | 5,881 | 4,269 | -27% | 1 | 1 | 0% | 1,280 | 1,181 | -8% | 0 | 0 | — |
case-22 | pass→pass | 6,215 | 3,772 | -39% | 1 | 1 | 0% | 1,414 | 940 | -34% | 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 +5 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.