Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Optimize landing pages for conversions, performance, and SEO. Use when improving landing pages, increasing conversions, or optimizing page performance.
.claude/skills/onewave-ai-landing-page-optimizer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 6% | 0% |
Audit and optimize a landing page across conversion, performance, and SEO.
references/checklists.md — above-the-fold and quick-wins checklistsreferences/page-structure.md — optimal eight-section page layoutreferences/cta-optimization.md — CTA copy, friction reduction, placementreferences/performance.md — image/CSS optimization, Core Web Vitals targets, mobilereferences/seo-and-tracking.md — meta tags, structured data, GA4 tracking, A/B testsreferences/checklists.md.references/page-structure.md; flag missing or out-of-order sections.references/cta-optimization.md — sharpen copy, reduce friction, and place CTAs at each high-intent point.references/performance.md; meet the Core Web Vitals targets and apply mobile patterns.references/seo-and-tracking.md — meta tags, structured data, and GA4 conversion events.references/checklists.md and queue prioritized A/B tests.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 34,040 | 29,132 | -14% | 1 | 1 | 0% | 5,644 | 4,950 | -12% | 0 | 0 | — |
case-02 | fail→pass | 27,845 | 24,004 | -14% | 1 | 1 | 0% | 4,444 | 4,313 | -3% | 0 | 0 | — |
case-03 | pass→pass | 17,480 | 15,304 | -12% | 1 | 1 | 0% | 2,620 | 2,786 | +6% | 0 | 0 | — |
case-04 | pass→pass | 36,470 | 41,612 | +14% | 1 | 1 | 0% | 5,778 | 6,515 | +13% | 0 | 0 | — |
case-05 | pass→pass | 15,178 | 15,187 | +0% | 1 | 1 | 0% | 2,860 | 3,043 | +6% | 0 | 0 | — |
case-06 | pass→pass | 13,700 | 15,337 | +12% | 1 | 1 | 0% | 2,369 | 3,010 | +27% | 0 | 0 | — |
case-07 | fail→fail | 13,151 | 9,567 | -27% | 1 | 1 | 0% | 2,182 | 2,009 | -8% | 0 | 0 | — |
case-08 | pass→pass | 10,821 | 12,112 | +12% | 1 | 1 | 0% | 1,796 | 2,279 | +27% | 0 | 0 | — |
case-09 | pass→pass | 14,895 | 12,049 | -19% | 1 | 1 | 0% | 2,243 | 2,239 | -0% | 0 | 0 | — |
case-10 | pass→pass | 10,960 | 11,866 | +8% | 1 | 1 | 0% | 1,753 | 2,160 | +23% | 0 | 0 | — |
case-11 | pass→pass | 14,970 | 12,895 | -14% | 1 | 1 | 0% | 2,681 | 2,701 | +1% | 0 | 0 | — |
case-12 | pass→pass | 13,421 | 14,378 | +7% | 1 | 1 | 0% | 2,198 | 2,781 | +27% | 0 | 0 | — |
case-13 | pass→pass | 11,321 | 9,171 | -19% | 1 | 1 | 0% | 2,360 | 2,147 | -9% | 0 | 0 | — |
case-14 | pass→pass | 12,513 | 8,324 | -33% | 1 | 1 | 0% | 2,479 | 2,183 | -12% | 0 | 0 | — |
case-15 | pass→pass | 13,016 | 9,599 | -26% | 1 | 1 | 0% | 2,581 | 2,157 | -16% | 0 | 0 | — |
case-16 | pass→pass | 13,518 | 11,175 | -17% | 1 | 1 | 0% | 2,337 | 2,190 | -6% | 0 | 0 | — |
case-17 | pass→pass | 13,560 | 10,793 | -20% | 1 | 1 | 0% | 2,300 | 2,206 | -4% | 0 | 0 | — |
case-18 | fail→pass | 9,861 | 8,245 | -16% | 1 | 1 | 0% | 1,649 | 1,771 | +7% | 0 | 0 | — |
case-19 | pass→pass | 11,840 | 12,438 | +5% | 1 | 1 | 0% | 2,326 | 2,804 | +21% | 0 | 0 | — |
case-20 | pass→pass | 12,594 | 9,207 | -27% | 1 | 1 | 0% | 2,173 | 1,901 | -13% | 0 | 0 | — |
case-21 | pass→pass | 11,257 | 7,762 | -31% | 1 | 1 | 0% | 2,055 | 1,839 | -11% | 0 | 0 | — |
case-22 | fail→pass | 14,265 | 8,727 | -39% | 1 | 1 | 0% | 2,285 | 1,815 | -21% | 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 +18 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.