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 meta refresh redirects. Check native semantics first, then inspect keyboard behavior, focus flow, accessible names, and screen-reader output where relevant.
.claude/skills/thedaviddias-meta-refresh/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -15% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 21% | 0% |
Unexpected page refreshes or redirects can be extremely confusing for users with cognitive disabilities and can cause screen readers to restart reading from the beginning.
<meta http-equiv="refresh"> for page redirectsCheck the HTML for any meta tags that use http-equiv="refresh" for automatic page redirection.
Replace meta refresh redirects with server-side redirects or clear, user-initiated links.
Explain why automatic page refreshes can be a major barrier for accessibility and user experience.
Review the rendered markup and interactive states that affect Avoid meta refresh redirects. 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/meta-refresh
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 15,499 | 13,023 | -16% | 1 | 1 | 0% | 2,816 | 2,705 | -4% | 0 | 0 | — |
case-02 | pass→pass | 7,388 | 4,945 | -33% | 1 | 1 | 0% | 1,425 | 1,216 | -15% | 0 | 0 | — |
case-03 | pass→pass | 11,245 | 12,007 | +7% | 1 | 1 | 0% | 2,198 | 2,650 | +21% | 0 | 0 | — |
case-04 | pass→pass | 4,435 | 4,419 | -0% | 1 | 1 | 0% | 822 | 969 | +18% | 0 | 0 | — |
case-05 | pass→pass | 12,227 | 12,262 | +0% | 1 | 1 | 0% | 2,259 | 2,517 | +11% | 0 | 0 | — |
case-06 | pass→pass | 11,773 | 11,570 | -2% | 1 | 1 | 0% | 2,165 | 2,272 | +5% | 0 | 0 | — |
case-07 | pass→pass | 12,313 | 10,248 | -17% | 1 | 1 | 0% | 2,064 | 2,043 | -1% | 0 | 0 | — |
case-08 | pass→pass | 13,116 | 9,265 | -29% | 1 | 1 | 0% | 2,427 | 2,082 | -14% | 0 | 0 | — |
case-15 | pass→pass | 13,234 | 9,609 | -27% | 1 | 1 | 0% | 2,347 | 1,959 | -17% | 0 | 0 | — |
case-09 | pass→pass | 9,468 | 5,684 | -40% | 1 | 1 | 0% | 1,649 | 1,231 | -25% | 0 | 0 | — |
case-10 | fail→pass | 8,921 | 11,070 | +24% | 1 | 1 | 0% | 1,581 | 2,243 | +42% | 0 | 0 | — |
case-11 | fail→pass | 10,758 | 10,527 | -2% | 1 | 1 | 0% | 2,128 | 2,274 | +7% | 0 | 0 | — |
case-12 | pass→pass | 11,991 | 8,219 | -31% | 1 | 1 | 0% | 2,177 | 1,734 | -20% | 0 | 0 | — |
case-13 | pass→pass | 11,838 | 7,735 | -35% | 1 | 1 | 0% | 1,961 | 1,648 | -16% | 0 | 0 | — |
case-14 | pass→pass | 15,575 | 12,208 | -22% | 1 | 1 | 0% | 2,606 | 2,519 | -3% | 0 | 0 | — |
case-16 | pass→pass | 9,647 | 12,163 | +26% | 1 | 1 | 0% | 1,703 | 2,593 | +52% | 0 | 0 | — |
case-17 | pass→pass | 12,967 | 12,601 | -3% | 1 | 1 | 0% | 2,291 | 2,554 | +11% | 0 | 0 | — |
case-18 | pass→pass | 7,951 | 5,637 | -29% | 1 | 1 | 0% | 1,518 | 1,277 | -16% | 0 | 0 | — |
case-19 | pass→pass | 11,537 | 9,139 | -21% | 1 | 1 | 0% | 1,953 | 1,773 | -9% | 0 | 0 | — |
case-20 | pass→pass | 16,582 | 14,939 | -10% | 1 | 1 | 0% | 3,076 | 3,379 | +10% | 0 | 0 | — |
case-21 | pass→pass | 10,529 | 12,191 | +16% | 1 | 1 | 0% | 2,045 | 2,526 | +24% | 0 | 0 | — |
case-22 | pass→pass | 6,935 | 7,624 | +10% | 1 | 1 | 0% | 1,288 | 1,579 | +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. 22 cases were attempted. The headline lift of +14 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.