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 Use correct list structure. Check native semantics first, then inspect keyboard behavior, focus flow, accessible names, and screen-reader output where relevant.
.claude/skills/thedaviddias-list-structure/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -6% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 39% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 27% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -3% | 0% |
Screen readers rely on a specific parent-child relationship for lists to correctly announce the number of items and the user's current position within the list.
<ul>) and ordered (<ol>) lists must only contain <li> elements<div> or <p> directly inside a list<li> elementCheck the structure of all lists to ensure only <li> elements are direct children of <ul> or <ol>.
Remove or move any non-li elements from within <ul> and <ol> tags.
Explain why assistive technology needs a consistent list structure to provide accurate navigation information.
Review the rendered markup and interactive states that affect Use correct list structure. 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/list-structure
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 11,331 | 9,217 | -19% | 1 | 1 | 0% | 1,908 | 1,789 | -6% | 0 | 0 | — |
case-02 | pass→pass | 6,694 | 7,096 | +6% | 1 | 1 | 0% | 1,061 | 1,476 | +39% | 0 | 0 | — |
case-03 | pass→pass | 7,666 | 8,059 | +5% | 1 | 1 | 0% | 1,190 | 1,510 | +27% | 0 | 0 | — |
case-04 | pass→pass | 10,358 | 7,887 | -24% | 1 | 1 | 0% | 1,483 | 1,443 | -3% | 0 | 0 | — |
case-05 | pass→pass | 7,194 | 7,161 | -0% | 1 | 1 | 0% | 1,091 | 1,367 | +25% | 0 | 0 | — |
case-06 | pass→pass | 4,074 | 6,070 | +49% | 1 | 1 | 0% | 527 | 1,427 | +171% | 0 | 0 | — |
case-07 | pass→pass | 7,679 | 9,522 | +24% | 1 | 1 | 0% | 1,127 | 1,741 | +54% | 0 | 0 | — |
case-08 | pass→pass | 19,107 | 16,294 | -15% | 1 | 1 | 0% | 2,593 | 2,575 | -1% | 0 | 0 | — |
case-09 | pass→pass | 3,945 | 6,316 | +60% | 1 | 1 | 0% | 626 | 1,244 | +99% | 0 | 0 | — |
case-10 | pass→pass | 9,026 | 9,909 | +10% | 1 | 1 | 0% | 1,345 | 1,625 | +21% | 0 | 0 | — |
case-11 | pass→pass | 4,481 | 6,497 | +45% | 1 | 1 | 0% | 682 | 1,207 | +77% | 0 | 0 | — |
case-12 | pass→pass | 3,443 | 7,684 | +123% | 1 | 1 | 0% | 631 | 1,367 | +117% | 0 | 0 | — |
case-13 | pass→pass | 4,653 | 7,692 | +65% | 1 | 1 | 0% | 607 | 1,493 | +146% | 0 | 0 | — |
case-14 | pass→pass | 3,295 | 6,932 | +110% | 1 | 1 | 0% | 613 | 1,187 | +94% | 0 | 0 | — |
case-15 | pass→pass | 8,681 | 10,684 | +23% | 1 | 1 | 0% | 1,288 | 1,865 | +45% | 0 | 0 | — |
case-16 | pass→pass | 18,145 | 16,043 | -12% | 1 | 1 | 0% | 2,703 | 2,535 | -6% | 0 | 0 | — |
case-17 | pass→pass | 9,121 | 10,829 | +19% | 1 | 1 | 0% | 1,485 | 1,873 | +26% | 0 | 0 | — |
case-18 | pass→pass | 10,313 | 6,848 | -34% | 1 | 1 | 0% | 1,433 | 1,275 | -11% | 0 | 0 | — |
case-19 | fail→pass | 10,693 | 10,457 | -2% | 1 | 1 | 0% | 1,556 | 1,913 | +23% | 0 | 0 | — |
case-20 | pass→pass | 9,621 | 7,134 | -26% | 1 | 1 | 0% | 1,583 | 1,468 | -7% | 0 | 0 | — |
case-21 | pass→pass | 12,592 | 12,782 | +2% | 1 | 1 | 0% | 2,081 | 2,345 | +13% | 0 | 0 | — |
case-22 | pass→pass | 9,454 | 10,248 | +8% | 1 | 1 | 0% | 1,485 | 1,832 | +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 +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.