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 file uploads accessible. Validate the final browser-facing markup, not just the source framework abstraction.
.claude/skills/thedaviddias-file-upload-accessibility/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -6% | 0% |
Unlabeled file inputs and inaccessible drag zones exclude screen reader and keyboard users from completing file upload tasks.
Verify file inputs have proper labels, accept attributes for file types, and accessible feedback for upload progress and errors.
Implement file uploads with visible labels, accept attribute restrictions, drag-and-drop alternatives, and ARIA live region feedback.
Explain how accessible file uploads provide clear instructions, feedback, and work for users relying on assistive technologies.
Review templates, server-rendered HTML, and shared components that output markup related to Make file uploads 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/file-upload-accessibility
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 16,150 | 15,542 | -4% | 1 | 1 | 0% | 3,606 | 3,746 | +4% | 0 | 0 | — |
case-02 | pass→pass | 11,582 | 7,099 | -39% | 1 | 1 | 0% | 2,309 | 1,660 | -28% | 0 | 0 | — |
case-03 | pass→pass | 12,615 | 12,322 | -2% | 1 | 1 | 0% | 2,666 | 3,104 | +16% | 0 | 0 | — |
case-04 | fail→fail | 11,975 | 7,659 | -36% | 1 | 1 | 0% | 2,753 | 1,930 | -30% | 0 | 0 | — |
case-05 | pass→pass | 6,502 | 3,884 | -40% | 1 | 1 | 0% | 1,498 | 1,018 | -32% | 0 | 0 | — |
case-06 | fail→fail | 14,151 | 10,030 | -29% | 1 | 1 | 0% | 3,105 | 2,725 | -12% | 0 | 0 | — |
case-07 | fail→pass | 9,448 | 6,142 | -35% | 1 | 1 | 0% | 1,966 | 1,730 | -12% | 0 | 0 | — |
case-08 | fail→pass | 16,503 | 8,560 | -48% | 1 | 1 | 0% | 3,429 | 2,358 | -31% | 0 | 0 | — |
case-09 | pass→pass | 9,959 | 6,518 | -35% | 1 | 1 | 0% | 2,427 | 1,632 | -33% | 0 | 0 | — |
case-10 | pass→fail | 16,748 | 15,679 | -6% | 1 | 1 | 0% | 3,794 | 3,969 | +5% | 0 | 0 | — |
case-11 | fail→pass | 8,902 | 5,471 | -39% | 1 | 1 | 0% | 2,099 | 1,579 | -25% | 0 | 0 | — |
case-12 | pass→pass | 21,121 | 18,671 | -12% | 1 | 1 | 0% | 5,057 | 4,598 | -9% | 0 | 0 | — |
case-13 | fail→pass | 11,640 | 10,120 | -13% | 1 | 1 | 0% | 2,673 | 2,525 | -6% | 0 | 0 | — |
case-14 | pass→pass | 13,368 | 5,908 | -56% | 1 | 1 | 0% | 2,588 | 1,531 | -41% | 0 | 0 | — |
case-15 | fail→fail | 14,155 | 8,636 | -39% | 1 | 1 | 0% | 3,333 | 2,004 | -40% | 0 | 0 | — |
case-16 | pass→pass | 7,063 | 5,468 | -23% | 1 | 1 | 0% | 1,513 | 1,420 | -6% | 0 | 0 | — |
case-17 | fail→fail | 9,001 | 8,499 | -6% | 1 | 1 | 0% | 1,740 | 1,791 | +3% | 0 | 0 | — |
case-18 | pass→pass | 5,482 | 4,407 | -20% | 1 | 1 | 0% | 1,142 | 1,225 | +7% | 0 | 0 | — |
case-19 | fail→fail | 10,995 | 7,744 | -30% | 1 | 1 | 0% | 2,191 | 2,031 | -7% | 0 | 0 | — |
case-20 | pass→pass | 15,669 | 9,999 | -36% | 1 | 1 | 0% | 3,083 | 2,233 | -28% | 0 | 0 | — |
case-21 | fail→pass | 10,146 | 7,324 | -28% | 1 | 1 | 0% | 2,142 | 1,844 | -14% | 0 | 0 | — |
case-22 | pass→fail | 25,367 | 13,638 | -46% | 1 | 1 | 0% | 6,182 | 3,779 | -39% | 0 | 0 | — |
case-23 | fail→pass | 9,463 | 5,993 | -37% | 1 | 1 | 0% | 2,109 | 1,620 | -23% | 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 +22 percentage points is the difference between those two pass rates over the 23 comparable cases. 3 cases got worse with the skill loaded, and they are 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.