Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when reviewing rendered HTML, interactive components, or design-system patterns related to Fix empty and broken links. Check native semantics first, then inspect keyboard behavior, focus flow, accessible names, and screen-reader output where relevant.
.claude/skills/thedaviddias-empty-links/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flashlowest | 97% | 53 |
| gemini-3.1-pro-preview | 100% | 1 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 13% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 24% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 19% | 0% |
Screen readers announce empty links as just 'link' with no context—users have no idea where they lead or what they do, making navigation impossible.
Scan for links with no text content (empty anchor tags, icon-only links without labels). Verify all links have accessible names via text content, aria-label, or aria-labelledby. Check for broken link destinations.
Add descriptive text to empty links. For icon-only links, add aria-label describing the destination. Use visually-hidden text if visible labels are not desired. Remove or fix broken links.
Explain how screen readers announce empty links as just 'link' with no context, leaving users unable to understand where the link leads or what action it performs.
Review the rendered markup and interactive states that affect Fix empty and broken links. Flag exact elements, roles, labels, focus behavior, or keyboard interactions that violate the rule, and note how to verify the fix with browser accessibility tooling or assistive tech.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/accessibility/empty-links
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 8,427 | 7,672 | -9% | 1 | 1 | 0% | 1,743 | 1,962 | +13% | 0 | 0 | — |
case-01 | fail→pass | 10,351 | 8,668 | -16% | 1 | 1 | 0% | 2,341 | 2,257 | -4% | 0 | 0 | — |
case-02 | pass→pass | 11,295 | 10,883 | -4% | 1 | 1 | 0% | 2,094 | 2,607 | +24% | 0 | 0 | — |
case-03 | pass→pass | 6,222 | 6,252 | +0% | 1 | 1 | 0% | 1,332 | 1,580 | +19% | 0 | 0 | — |
case-04 | pass→pass | 8,885 | 11,059 | +24% | 1 | 1 | 0% | 1,672 | 2,460 | +47% | 0 | 0 | — |
case-05 | pass→pass | 8,356 | 7,775 | -7% | 1 | 1 | 0% | 1,641 | 1,829 | +11% | 0 | 0 | — |
case-06 | pass→pass | 8,221 | 5,337 | -35% | 1 | 1 | 0% | 1,637 | 1,454 | -11% | 0 | 0 | — |
case-07 | pass→pass | 8,693 | 6,434 | -26% | 1 | 1 | 0% | 1,813 | 1,503 | -17% | 0 | 0 | — |
case-08 | pass→pass | 10,099 | 8,253 | -18% | 1 | 1 | 0% | 2,082 | 1,943 | -7% | 0 | 0 | — |
case-09 | pass→pass | 9,202 | 8,595 | -7% | 1 | 1 | 0% | 1,892 | 2,016 | +7% | 0 | 0 | — |
case-10 | pass→pass | 10,349 | 9,799 | -5% | 1 | 1 | 0% | 1,825 | 2,184 | +20% | 0 | 0 | — |
case-11 | fail→pass | 9,043 | 5,704 | -37% | 1 | 1 | 0% | 1,925 | 1,459 | -24% | 0 | 0 | — |
case-12 | pass→pass | 8,544 | 8,275 | -3% | 1 | 1 | 0% | 1,826 | 1,933 | +6% | 0 | 0 | — |
case-13 | pass→pass | 5,578 | 4,642 | -17% | 1 | 1 | 0% | 1,178 | 1,245 | +6% | 0 | 0 | — |
case-14 | pass→pass | 6,273 | 5,663 | -10% | 1 | 1 | 0% | 1,353 | 1,304 | -4% | 0 | 0 | — |
case-15 | pass→pass | 13,678 | 8,472 | -38% | 1 | 1 | 0% | 2,320 | 1,883 | -19% | 0 | 0 | — |
case-16 | pass→pass | 11,122 | 11,951 | +7% | 1 | 1 | 0% | 2,231 | 2,695 | +21% | 0 | 0 | — |
case-17 | pass→pass | 4,709 | 5,042 | +7% | 1 | 1 | 0% | 893 | 1,209 | +35% | 0 | 0 | — |
case-18 | pass→pass | 7,610 | 6,206 | -18% | 1 | 1 | 0% | 1,364 | 1,537 | +13% | 0 | 0 | — |
case-19 | pass→pass | 7,644 | 5,463 | -29% | 1 | 1 | 0% | 1,324 | 1,354 | +2% | 0 | 0 | — |
case-21 | pass→pass | 7,674 | 4,245 | -45% | 1 | 1 | 0% | 1,499 | 1,138 | -24% | 0 | 0 | — |
case-22 | pass→pass | 7,576 | 8,191 | +8% | 1 | 1 | 0% | 1,667 | 2,045 | +23% | 0 | 0 | — |
case-23 | pass→pass | 7,917 | 4,823 | -39% | 1 | 1 | 0% | 1,374 | 1,270 | -8% | 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. The headline lift of +9 percentage points is the difference between those two pass rates over the 23 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.