Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Landing page quality assessment for paid advertising campaigns. Evaluates message match, page speed, mobile experience, trust signals, form optimization, and conversion rate potential. Triggers on: "landing page", "post-click experience", "landing page audit", "conversion rate", "landing page optimization", "LP audit"
.claude/skills/miosa-osa-ads-landing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 50% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 84% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 96% | 0% |
> Evaluates landing pages specifically for paid ad campaign performance.
bash/ads landing /ads landing --url https://example.com/landing
The #1 landing page issue in ad campaigns — does the page match the ad?
| Level | Description | Score | |-------|-------------|-------| | Exact match | Headline, offer, CTA all align perfectly | 100% | | Partial match | Headline matches but offer/CTA differs | 60% | | Weak match | Generic page, loosely related to ad | 30% | | Mismatch | Page content doesn't reflect ad promise | 0% |
For every 1s delay, CVR drops ~7%.
| Metric | Pass | Warning | Fail | |--------|------|---------|------| | LCP | <2.5s | 2.5-4.0s | >4.0s | | FID/INP | <100ms | 100-200ms | >200ms | | CLS | <0.1 | 0.1-0.25 | >0.25 | | Time to Interactive | <3.0s | 3.0-5.0s | >5.0s | | Page weight | <2MB | 2-5MB | >5MB |
75%+ of ad clicks come from mobile.
| Fields | Expected CVR Impact | Use Case | |--------|-------------------|----------| | 1-3 fields | Highest CVR | Top-of-funnel, free offer | | 4-5 fields | Moderate CVR | Mid-funnel, qualified leads | | 6-8 fields | Lower CVR | Bottom-funnel, sales-ready | | 9+ fields | Lowest CVR | Only for high-value offers |
| Platform | Key Requirement | Notes | |----------|----------------|-------| | Google | QS component: landing page experience | Directly affects ad rank and CPC | | Meta | Page load speed critical | Slow pages = Meta penalizes delivery | | LinkedIn | Professional, B2B appropriate | Match LinkedIn's professional context | | TikTok | Mobile-first mandatory | 95%+ TikTok traffic is mobile | | Microsoft | Desktop-optimized matters more | Higher desktop % than other platforms |
LANDING-PAGE-REPORT.md — Per-page assessment with scores| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 20,201 | 18,637 | -8% | 1 | 1 | 0% | 2,633 | 3,784 | +44% | 0 | 0 | — |
case-01 | fail→fail | 16,028 | 43,342 | +170% | 1 | 1 | 0% | 2,513 | 4,882 | +94% | 0 | 0 | — |
case-02 | fail→fail | 20,342 | 24,521 | +21% | 1 | 1 | 0% | 3,114 | 4,975 | +60% | 0 | 0 | — |
case-03 | fail→pass | 22,251 | 22,402 | +1% | 1 | 1 | 0% | 3,628 | 4,817 | +33% | 0 | 0 | — |
case-04 | pass→pass | 7,833 | 13,661 | +74% | 1 | 1 | 0% | 1,400 | 3,375 | +141% | 0 | 0 | — |
case-05 | pass→pass | 18,056 | 22,312 | +24% | 1 | 1 | 0% | 2,756 | 3,841 | +39% | 0 | 0 | — |
case-06 | pass→pass | 16,879 | 16,839 | -0% | 1 | 1 | 0% | 2,690 | 3,824 | +42% | 0 | 0 | — |
case-07 | fail→pass | 9,289 | 7,058 | -24% | 1 | 1 | 0% | 1,576 | 2,087 | +32% | 0 | 0 | — |
case-08 | fail→pass | 10,484 | 8,737 | -17% | 1 | 1 | 0% | 1,675 | 2,509 | +50% | 0 | 0 | — |
case-09 | fail→pass | 7,998 | 6,979 | -13% | 1 | 1 | 0% | 1,029 | 1,892 | +84% | 0 | 0 | — |
case-10 | pass→pass | 9,701 | 6,446 | -34% | 1 | 1 | 0% | 1,612 | 2,004 | +24% | 0 | 0 | — |
case-11 | fail→pass | 6,433 | 5,628 | -13% | 1 | 1 | 0% | 874 | 1,711 | +96% | 0 | 0 | — |
case-12 | pass→pass | 11,918 | 7,470 | -37% | 1 | 1 | 0% | 1,805 | 2,134 | +18% | 0 | 0 | — |
case-13 | pass→pass | 14,008 | 12,067 | -14% | 1 | 1 | 0% | 2,259 | 3,005 | +33% | 0 | 0 | — |
case-14 | pass→pass | 10,831 | 9,313 | -14% | 1 | 1 | 0% | 1,756 | 2,513 | +43% | 0 | 0 | — |
case-15 | pass→pass | 17,449 | 10,783 | -38% | 1 | 1 | 0% | 2,265 | 2,681 | +18% | 0 | 0 | — |
case-16 | pass→fail | 17,251 | 15,125 | -12% | 1 | 1 | 0% | 2,676 | 2,889 | +8% | 0 | 0 | — |
case-17 | pass→pass | 10,900 | 5,195 | -52% | 1 | 1 | 0% | 1,667 | 1,710 | +3% | 0 | 0 | — |
case-18 | pass→pass | 16,241 | 13,484 | -17% | 1 | 1 | 0% | 2,091 | 3,051 | +46% | 0 | 0 | — |
case-19 | fail→pass | 10,614 | 14,110 | +33% | 1 | 1 | 0% | 1,605 | 3,265 | +103% | 0 | 0 | — |
case-21 | pass→pass | 12,992 | 7,845 | -40% | 1 | 1 | 0% | 2,010 | 2,388 | +19% | 0 | 0 | — |
case-22 | fail→pass | 15,173 | 6,422 | -58% | 1 | 1 | 0% | 1,995 | 2,097 | +5% | 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 +27 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.