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 Make pagination accessible. Validate the final browser-facing markup, not just the source framework abstraction.
.claude/skills/thedaviddias-pagination-accessibility/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -16% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -24% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -4% | 0% |
Without proper markup, screen reader users cannot understand which page they're on or navigate efficiently through paginated content.
Verify pagination uses nav element with aria-label, indicates current page with aria-current, and is keyboard navigable.
Implement pagination with nav role='navigation', aria-label, aria-current='page' for active page, and focusable controls.
Explain how accessible pagination helps screen reader users understand their position and navigate through paged content.
Review templates, server-rendered HTML, and shared components that output markup related to Make pagination accessible. 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/pagination-accessibility
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 10,714 | 7,713 | -28% | 1 | 1 | 0% | 2,150 | 1,807 | -16% | 0 | 0 | — |
case-02 | pass→pass | 7,056 | 4,324 | -39% | 1 | 1 | 0% | 1,428 | 1,090 | -24% | 0 | 0 | — |
case-03 | pass→pass | 4,898 | 3,662 | -25% | 1 | 1 | 0% | 1,036 | 997 | -4% | 0 | 0 | — |
case-04 | pass→pass | 8,628 | 4,893 | -43% | 1 | 1 | 0% | 1,694 | 1,220 | -28% | 0 | 0 | — |
case-05 | pass→pass | 7,670 | 4,548 | -41% | 1 | 1 | 0% | 1,407 | 1,195 | -15% | 0 | 0 | — |
case-06 | pass→pass | 3,778 | 3,813 | +1% | 1 | 1 | 0% | 727 | 947 | +30% | 0 | 0 | — |
case-07 | pass→pass | 2,633 | 2,176 | -17% | 1 | 1 | 0% | 487 | 667 | +37% | 0 | 0 | — |
case-08 | pass→pass | 2,848 | 2,595 | -9% | 1 | 1 | 0% | 541 | 705 | +30% | 0 | 0 | — |
case-09 | pass→pass | 4,517 | 3,723 | -18% | 1 | 1 | 0% | 953 | 1,007 | +6% | 0 | 0 | — |
case-10 | pass→pass | 6,802 | 4,677 | -31% | 1 | 1 | 0% | 1,336 | 1,205 | -10% | 0 | 0 | — |
case-19 | pass→pass | 4,559 | 2,899 | -36% | 1 | 1 | 0% | 809 | 766 | -5% | 0 | 0 | — |
case-11 | pass→pass | 5,080 | 5,333 | +5% | 1 | 1 | 0% | 898 | 1,307 | +46% | 0 | 0 | — |
case-12 | pass→pass | 8,080 | 4,327 | -46% | 1 | 1 | 0% | 1,582 | 1,072 | -32% | 0 | 0 | — |
case-13 | pass→pass | 7,900 | 7,267 | -8% | 1 | 1 | 0% | 1,652 | 1,613 | -2% | 0 | 0 | — |
case-14 | pass→pass | 6,764 | 4,727 | -30% | 1 | 1 | 0% | 1,486 | 1,262 | -15% | 0 | 0 | — |
case-15 | pass→pass | 9,490 | 7,022 | -26% | 1 | 1 | 0% | 1,747 | 1,582 | -9% | 0 | 0 | — |
case-16 | fail→pass | 6,396 | 5,095 | -20% | 1 | 1 | 0% | 1,222 | 1,183 | -3% | 0 | 0 | — |
case-17 | pass→pass | 9,572 | 9,215 | -4% | 1 | 1 | 0% | 1,757 | 1,938 | +10% | 0 | 0 | — |
case-18 | pass→pass | 7,119 | 3,982 | -44% | 1 | 1 | 0% | 1,390 | 1,072 | -23% | 0 | 0 | — |
case-20 | pass→pass | 7,898 | 5,457 | -31% | 1 | 1 | 0% | 1,691 | 1,385 | -18% | 0 | 0 | — |
case-21 | pass→pass | 2,496 | 3,086 | +24% | 1 | 1 | 0% | 478 | 751 | +57% | 0 | 0 | — |
case-22 | fail→pass | 6,557 | 4,441 | -32% | 1 | 1 | 0% | 1,271 | 1,063 | -16% | 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 +9 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.