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 videos accessible with captions. Validate the final browser-facing markup, not just the source framework abstraction.
.claude/skills/thedaviddias-video-accessibility/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-11 | ✓→✓ | ＝ Same ✓ | -13% | 0% |
| case-10 | ✓→✓ | ＝ Same ✓ | -24% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -25% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 13% | 0% |
Videos without captions exclude deaf and hard-of-hearing users, while videos without audio descriptions exclude blind users from understanding visual content.
Verify videos have accurate closed captions (not auto-generated), audio descriptions for visual content, transcripts, keyboard-accessible pause controls, and no autoplay. Confirm spacebar pauses video without scrolling the page.
Add track elements for captions and audio descriptions, provide text transcripts, implement global pause controls, disable autoplay, and ensure media controls are keyboard accessible with proper focus management.
Explain how captions help deaf and hard-of-hearing users, audio descriptions help blind users understand visual content, and pause controls help users with cognitive disabilities who find autoplay disorienting.
Review templates, server-rendered HTML, and shared components that output markup related to Make videos accessible with captions. 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/video-accessibility
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | pass→pass | 10,087 | 7,349 | -27% | 1 | 1 | 0% | 1,845 | 1,609 | -13% | 0 | 0 | — |
case-05 | fail→pass | 12,545 | 7,779 | -38% | 1 | 1 | 0% | 2,008 | 1,608 | -20% | 0 | 0 | — |
case-10 | pass→pass | 11,527 | 8,612 | -25% | 1 | 1 | 0% | 2,376 | 1,811 | -24% | 0 | 0 | — |
case-01 | fail→fail | 12,755 | 9,362 | -27% | 1 | 1 | 0% | 2,583 | 1,978 | -23% | 0 | 0 | — |
case-02 | fail→fail | 15,007 | 8,041 | -46% | 1 | 1 | 0% | 2,425 | 1,540 | -36% | 0 | 0 | — |
case-03 | pass→pass | 12,079 | 7,961 | -34% | 1 | 1 | 0% | 2,071 | 1,551 | -25% | 0 | 0 | — |
case-04 | pass→pass | 10,497 | 10,453 | -0% | 1 | 1 | 0% | 2,012 | 2,271 | +13% | 0 | 0 | — |
case-06 | pass→pass | 10,518 | 7,913 | -25% | 1 | 1 | 0% | 1,780 | 1,614 | -9% | 0 | 0 | — |
case-07 | fail→fail | 9,799 | 7,045 | -28% | 1 | 1 | 0% | 1,614 | 1,449 | -10% | 0 | 0 | — |
case-08 | pass→pass | 8,776 | 5,909 | -33% | 1 | 1 | 0% | 1,706 | 1,513 | -11% | 0 | 0 | — |
case-09 | pass→pass | 14,300 | 10,429 | -27% | 1 | 1 | 0% | 2,201 | 2,058 | -6% | 0 | 0 | — |
case-12 | pass→pass | 17,572 | 13,150 | -25% | 1 | 1 | 0% | 3,527 | 2,884 | -18% | 0 | 0 | — |
case-13 | pass→pass | 7,999 | 4,401 | -45% | 1 | 1 | 0% | 1,390 | 1,012 | -27% | 0 | 0 | — |
case-14 | pass→pass | 13,435 | 9,047 | -33% | 1 | 1 | 0% | 2,297 | 1,784 | -22% | 0 | 0 | — |
case-15 | pass→pass | 13,105 | 7,578 | -42% | 1 | 1 | 0% | 2,004 | 1,358 | -32% | 0 | 0 | — |
case-16 | pass→pass | 15,349 | 10,206 | -34% | 1 | 1 | 0% | 2,745 | 2,023 | -26% | 0 | 0 | — |
case-17 | pass→pass | 10,947 | 9,338 | -15% | 1 | 1 | 0% | 2,083 | 2,093 | +0% | 0 | 0 | — |
case-18 | pass→pass | 13,457 | 6,718 | -50% | 1 | 1 | 0% | 2,136 | 1,440 | -33% | 0 | 0 | — |
case-19 | pass→pass | 13,858 | 7,233 | -48% | 1 | 1 | 0% | 2,244 | 1,458 | -35% | 0 | 0 | — |
case-20 | pass→pass | 15,656 | 11,558 | -26% | 1 | 1 | 0% | 2,703 | 2,225 | -18% | 0 | 0 | — |
case-21 | pass→pass | 15,492 | 9,974 | -36% | 1 | 1 | 0% | 2,823 | 2,103 | -26% | 0 | 0 | — |
case-22 | pass→pass | 13,040 | 8,373 | -36% | 1 | 1 | 0% | 2,385 | 1,804 | -24% | 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. 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.