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 Avoid redundant image alternative text. Check native semantics first, then inspect keyboard behavior, focus flow, accessible names, and screen-reader output where relevant.
.claude/skills/thedaviddias-image-redundant-alt/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-18 | ✓→✓ | ＝ Same ✓ | -4% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 42% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -2% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -10% | 0% |
Redundant alt text increases the verbosity for screen reader users without adding value, leading to a poorer user experience and slower navigation.
Scan image alt attributes for redundant phrases like 'image of', 'photo of', or text that duplicates surrounding content.
Remove redundant phrases from the alt attribute and ensure the text provides unique, descriptive information.
Explain why screen readers already announce 'graphic' or 'image', making phrases like 'image of' unnecessary and repetitive.
Review the rendered markup and interactive states that affect Avoid redundant image alternative 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/image-redundant-alt
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | pass→pass | 8,422 | 7,015 | -17% | 1 | 1 | 0% | 1,441 | 1,386 | -4% | 0 | 0 | — |
case-01 | pass→pass | 3,231 | 3,279 | +1% | 1 | 1 | 0% | 631 | 895 | +42% | 0 | 0 | — |
case-02 | pass→pass | 7,875 | 6,495 | -18% | 1 | 1 | 0% | 1,425 | 1,390 | -2% | 0 | 0 | — |
case-03 | pass→pass | 6,288 | 4,054 | -36% | 1 | 1 | 0% | 1,079 | 970 | -10% | 0 | 0 | — |
case-04 | pass→pass | 3,416 | 3,658 | +7% | 1 | 1 | 0% | 578 | 957 | +66% | 0 | 0 | — |
case-05 | pass→pass | 6,083 | 3,358 | -45% | 1 | 1 | 0% | 1,013 | 866 | -15% | 0 | 0 | — |
case-06 | pass→pass | 10,033 | 7,997 | -20% | 1 | 1 | 0% | 1,767 | 1,707 | -3% | 0 | 0 | — |
case-07 | pass→pass | 2,652 | 2,985 | +13% | 1 | 1 | 0% | 474 | 748 | +58% | 0 | 0 | — |
case-08 | pass→pass | 5,911 | 5,428 | -8% | 1 | 1 | 0% | 1,084 | 1,215 | +12% | 0 | 0 | — |
case-09 | pass→pass | 3,410 | 2,721 | -20% | 1 | 1 | 0% | 561 | 748 | +33% | 0 | 0 | — |
case-10 | pass→pass | 6,980 | 6,129 | -12% | 1 | 1 | 0% | 1,304 | 1,428 | +10% | 0 | 0 | — |
case-11 | pass→pass | 7,667 | 5,431 | -29% | 1 | 1 | 0% | 1,329 | 1,180 | -11% | 0 | 0 | — |
case-12 | pass→pass | 5,555 | 4,402 | -21% | 1 | 1 | 0% | 1,032 | 1,046 | +1% | 0 | 0 | — |
case-13 | pass→pass | 3,887 | 4,718 | +21% | 1 | 1 | 0% | 743 | 1,108 | +49% | 0 | 0 | — |
case-14 | pass→pass | 2,733 | 2,598 | -5% | 1 | 1 | 0% | 466 | 686 | +47% | 0 | 0 | — |
case-15 | pass→pass | 14,616 | 10,940 | -25% | 1 | 1 | 0% | 2,413 | 2,037 | -16% | 0 | 0 | — |
case-16 | pass→pass | 3,285 | 4,341 | +32% | 1 | 1 | 0% | 620 | 1,079 | +74% | 0 | 0 | — |
case-17 | pass→pass | 2,827 | 2,537 | -10% | 1 | 1 | 0% | 533 | 707 | +33% | 0 | 0 | — |
case-19 | pass→pass | 9,723 | 4,970 | -49% | 1 | 1 | 0% | 1,684 | 1,103 | -35% | 0 | 0 | — |
case-20 | pass→pass | 8,263 | 6,093 | -26% | 1 | 1 | 0% | 1,427 | 1,380 | -3% | 0 | 0 | — |
case-21 | fail→pass | 11,363 | 8,667 | -24% | 1 | 1 | 0% | 2,107 | 1,874 | -11% | 0 | 0 | — |
case-22 | fail→fail | 12,450 | 9,175 | -26% | 1 | 1 | 0% | 2,387 | 2,018 | -15% | 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.