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 Use descriptive link text. Check native semantics first, then inspect keyboard behavior, focus flow, accessible names, and screen-reader output where relevant.
.claude/skills/thedaviddias-link-text/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-20 | ✓→✗ | ▼ Worse | 3% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 1% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -15% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 23% | 0% |
Screen reader users navigate by listing all links—'click here' repeated 10 times tells them nothing, while descriptive links let them jump directly to relevant content.
Review all links and verify text describes the destination. Avoid generic phrases like 'click here', 'read more', 'learn more' without additional context. Check that screen readers can understand link purpose out of context.
Replace generic link text with descriptive phrases. Use aria-label or aria-labelledby when visual design requires short text but context is needed. Ensure each link's purpose is clear from its text alone.
Explain how screen reader users often navigate by listing all links on a page, and why 'click here' repeated multiple times provides no useful information about where each link leads.
Review the rendered markup and interactive states that affect Use descriptive link text. 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/link-text
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,008 | 13,332 | -33% | 1 | 1 | 0% | 3,301 | 2,784 | -16% | 0 | 0 | — |
case-02 | fail→fail | 13,834 | 12,783 | -8% | 1 | 1 | 0% | 2,345 | 2,320 | -1% | 0 | 0 | — |
case-03 | pass→pass | 11,043 | 9,287 | -16% | 1 | 1 | 0% | 1,974 | 1,991 | +1% | 0 | 0 | — |
case-04 | pass→pass | 14,919 | 10,835 | -27% | 1 | 1 | 0% | 2,331 | 1,986 | -15% | 0 | 0 | — |
case-05 | pass→pass | 11,477 | 10,779 | -6% | 1 | 1 | 0% | 1,679 | 2,064 | +23% | 0 | 0 | — |
case-06 | pass→pass | 6,816 | 4,091 | -40% | 1 | 1 | 0% | 928 | 882 | -5% | 0 | 0 | — |
case-07 | pass→pass | 14,749 | 11,828 | -20% | 1 | 1 | 0% | 1,952 | 1,969 | +1% | 0 | 0 | — |
case-08 | pass→pass | 9,924 | 6,600 | -33% | 1 | 1 | 0% | 1,624 | 1,387 | -15% | 0 | 0 | — |
case-09 | pass→pass | 9,484 | 5,343 | -44% | 1 | 1 | 0% | 1,545 | 1,141 | -26% | 0 | 0 | — |
case-10 | pass→pass | 13,494 | 8,743 | -35% | 1 | 1 | 0% | 2,062 | 1,655 | -20% | 0 | 0 | — |
case-11 | pass→pass | 11,424 | 8,224 | -28% | 1 | 1 | 0% | 1,799 | 1,614 | -10% | 0 | 0 | — |
case-12 | pass→pass | 19,487 | 13,266 | -32% | 1 | 1 | 0% | 2,882 | 2,352 | -18% | 0 | 0 | — |
case-13 | pass→pass | 13,188 | 13,878 | +5% | 1 | 1 | 0% | 1,797 | 2,078 | +16% | 0 | 0 | — |
case-14 | pass→pass | 10,058 | 9,339 | -7% | 1 | 1 | 0% | 1,431 | 1,774 | +24% | 0 | 0 | — |
case-15 | pass→pass | 13,496 | 10,466 | -22% | 1 | 1 | 0% | 2,205 | 2,023 | -8% | 0 | 0 | — |
case-16 | pass→pass | 15,980 | 13,839 | -13% | 1 | 1 | 0% | 2,284 | 2,447 | +7% | 0 | 0 | — |
case-17 | pass→pass | 7,935 | 7,550 | -5% | 1 | 1 | 0% | 1,181 | 1,458 | +23% | 0 | 0 | — |
case-18 | pass→pass | 11,165 | 8,601 | -23% | 1 | 1 | 0% | 1,778 | 1,749 | -2% | 0 | 0 | — |
case-19 | pass→pass | 10,603 | 11,658 | +10% | 1 | 1 | 0% | 1,894 | 2,161 | +14% | 0 | 0 | — |
case-20 | pass→fail | 13,646 | 11,818 | -13% | 1 | 1 | 0% | 2,144 | 2,202 | +3% | 0 | 0 | — |
case-21 | pass→pass | 10,882 | 11,218 | +3% | 1 | 1 | 0% | 1,830 | 2,355 | +29% | 0 | 0 | — |
case-22 | pass→pass | 9,074 | 7,543 | -17% | 1 | 1 | 0% | 1,497 | 1,502 | +0% | 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 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.