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 scrolljacking and custom scroll behavior. Check native semantics first, then inspect keyboard behavior, focus flow, accessible names, and screen-reader output where relevant.
.claude/skills/thedaviddias-scrolljacking/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | -20% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | -22% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 8% | 0% |
Scrolljacking breaks user expectations, interferes with assistive technologies, and creates unpredictable experiences that frustrate users with motor impairments or cognitive disabilities.
Test scrolling behavior and verify it feels native. Check that scroll speed is not modified, scroll direction is not inverted, and scroll events are not hijacked for animations or navigation without user consent.
Remove JavaScript that overrides default scroll behavior. If custom scrolling is essential, respect prefers-reduced-motion and provide a way to disable custom scroll effects.
Explain how scrolljacking disrupts expected navigation patterns, interferes with assistive technologies, and creates unpredictable experiences for users with motor impairments or cognitive disabilities.
Review the rendered markup and interactive states that affect Avoid scrolljacking and custom scroll behavior. 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/scrolljacking
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | pass→pass | 13,172 | 8,727 | -34% | 1 | 1 | 0% | 2,237 | 1,789 | -20% | 0 | 0 | — |
case-14 | fail→pass | 19,363 | 18,692 | -3% | 1 | 1 | 0% | 3,422 | 3,517 | +3% | 0 | 0 | — |
case-07 | pass→pass | 17,839 | 12,004 | -33% | 1 | 1 | 0% | 2,977 | 2,321 | -22% | 0 | 0 | — |
case-05 | pass→pass | 15,809 | 13,678 | -13% | 1 | 1 | 0% | 2,632 | 2,834 | +8% | 0 | 0 | — |
case-01 | pass→pass | 9,858 | 8,018 | -19% | 1 | 1 | 0% | 1,735 | 1,673 | -4% | 0 | 0 | — |
case-02 | pass→pass | 16,832 | 14,651 | -13% | 1 | 1 | 0% | 2,868 | 2,961 | +3% | 0 | 0 | — |
case-03 | pass→pass | 15,087 | 12,884 | -15% | 1 | 1 | 0% | 2,359 | 2,394 | +1% | 0 | 0 | — |
case-04 | pass→pass | 13,361 | 12,857 | -4% | 1 | 1 | 0% | 2,441 | 2,716 | +11% | 0 | 0 | — |
case-08 | pass→pass | 17,484 | 9,336 | -47% | 1 | 1 | 0% | 2,699 | 1,812 | -33% | 0 | 0 | — |
case-09 | pass→pass | 17,557 | 13,369 | -24% | 1 | 1 | 0% | 2,916 | 2,522 | -14% | 0 | 0 | — |
case-10 | pass→pass | 15,766 | 10,366 | -34% | 1 | 1 | 0% | 2,615 | 2,093 | -20% | 0 | 0 | — |
case-11 | pass→pass | 13,036 | 9,019 | -31% | 1 | 1 | 0% | 2,464 | 1,996 | -19% | 0 | 0 | — |
case-12 | pass→pass | 16,243 | 13,983 | -14% | 1 | 1 | 0% | 2,573 | 2,397 | -7% | 0 | 0 | — |
case-13 | pass→pass | 7,965 | 5,820 | -27% | 1 | 1 | 0% | 1,257 | 1,215 | -3% | 0 | 0 | — |
case-15 | pass→pass | 16,044 | 13,817 | -14% | 1 | 1 | 0% | 2,327 | 2,382 | +2% | 0 | 0 | — |
case-16 | pass→pass | 10,863 | 11,840 | +9% | 1 | 1 | 0% | 1,904 | 2,278 | +20% | 0 | 0 | — |
case-17 | pass→pass | 15,252 | 10,715 | -30% | 1 | 1 | 0% | 2,493 | 2,098 | -16% | 0 | 0 | — |
case-18 | pass→pass | 13,698 | 11,507 | -16% | 1 | 1 | 0% | 2,082 | 2,244 | +8% | 0 | 0 | — |
case-19 | fail→fail | 17,316 | 12,878 | -26% | 1 | 1 | 0% | 2,834 | 2,375 | -16% | 0 | 0 | — |
case-20 | fail→pass | 16,644 | 9,922 | -40% | 1 | 1 | 0% | 2,870 | 1,869 | -35% | 0 | 0 | — |
case-21 | pass→pass | 11,033 | 6,681 | -39% | 1 | 1 | 0% | 2,047 | 1,375 | -33% | 0 | 0 | — |
case-22 | pass→pass | 14,219 | 7,268 | -49% | 1 | 1 | 0% | 2,397 | 1,507 | -37% | 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 +9 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.