Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Amazon Store builder — page layouts, brand story, shoppable images, traffic driving, conversion optimization
.claude/skills/nexscope-ai-amazon-storefront-design/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -26% | 0% |
Amazon Store builder — page layouts, brand story, shoppable images, traffic driving, conversion optimization
Supported platforms: Amazon (US, UK, DE, CA, JP, AU, and all marketplaces).
Built by Nexscope — your AI assistant for smarter e-commerce decisions.
bashnpx skills add nexscope/amazon-storefront-design
Help me with amazon storefront design for my e-commerce business.Step 1: Collect information from the user's message — product, platform, current situation, and goals.
Step 2: Ask one follow-up with all remaining questions using multiple-choice format. Allow shorthand answers (e.g., "1b 2c 3a").
Step 3: Research and analyze using the frameworks and methodology below.
Step 4: Deliver structured, actionable output with specific recommendations, not vague advice.
More e-commerce skills: nexscope-ai/eCommerce-Skills
Amazon-specific skills: nexscope-ai/Amazon-Skills
Built by Nexscope — your AI assistant for smarter e-commerce decisions.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 18,787 | 7,699 | -59% | 1 | 1 | 0% | 2,850 | 1,700 | -40% | 0 | 0 | — |
case-02 | fail→fail | 21,867 | 22,033 | +1% | 1 | 1 | 0% | 3,286 | 3,657 | +11% | 0 | 0 | — |
case-03 | fail→pass | 22,904 | 23,689 | +3% | 1 | 1 | 0% | 3,359 | 3,993 | +19% | 0 | 0 | — |
case-04 | pass→pass | 18,043 | 12,477 | -31% | 1 | 1 | 0% | 2,749 | 2,168 | -21% | 0 | 0 | — |
case-05 | pass→pass | 15,324 | 17,316 | +13% | 1 | 1 | 0% | 2,338 | 3,104 | +33% | 0 | 0 | — |
case-06 | pass→pass | 14,765 | 13,828 | -6% | 1 | 1 | 0% | 2,374 | 2,543 | +7% | 0 | 0 | — |
case-07 | fail→pass | 17,801 | 17,198 | -3% | 1 | 1 | 0% | 2,688 | 3,434 | +28% | 0 | 0 | — |
case-08 | fail→pass | 14,544 | 9,675 | -33% | 1 | 1 | 0% | 2,448 | 1,871 | -24% | 0 | 0 | — |
case-09 | fail→pass | 18,189 | 10,474 | -42% | 1 | 1 | 0% | 2,573 | 1,912 | -26% | 0 | 0 | — |
case-10 | fail→pass | 18,824 | 6,323 | -66% | 1 | 1 | 0% | 2,661 | 1,331 | -50% | 0 | 0 | — |
case-11 | fail→pass | 16,085 | 6,179 | -62% | 1 | 1 | 0% | 2,452 | 1,248 | -49% | 0 | 0 | — |
case-12 | pass→pass | 10,067 | 11,134 | +11% | 1 | 1 | 0% | 1,569 | 2,111 | +35% | 0 | 0 | — |
case-13 | fail→pass | 13,772 | 15,548 | +13% | 1 | 1 | 0% | 1,988 | 2,897 | +46% | 0 | 0 | — |
case-14 | fail→pass | 22,978 | 20,097 | -13% | 1 | 1 | 0% | 3,445 | 3,224 | -6% | 0 | 0 | — |
case-15 | fail→pass | 18,277 | 15,420 | -16% | 1 | 1 | 0% | 2,725 | 2,649 | -3% | 0 | 0 | — |
case-16 | fail→pass | 16,146 | 14,416 | -11% | 1 | 1 | 0% | 2,329 | 2,630 | +13% | 0 | 0 | — |
case-17 | fail→pass | 15,794 | 7,289 | -54% | 1 | 1 | 0% | 2,560 | 1,558 | -39% | 0 | 0 | — |
case-18 | fail→pass | 12,558 | 8,594 | -32% | 1 | 1 | 0% | 2,036 | 1,749 | -14% | 0 | 0 | — |
case-19 | fail→pass | 13,913 | 16,186 | +16% | 1 | 1 | 0% | 2,068 | 3,194 | +54% | 0 | 0 | — |
case-20 | fail→pass | 15,098 | 5,850 | -61% | 1 | 1 | 0% | 2,221 | 1,204 | -46% | 0 | 0 | — |
case-21 | fail→pass | 16,664 | 8,263 | -50% | 1 | 1 | 0% | 2,306 | 1,657 | -28% | 0 | 0 | — |
case-22 | pass→pass | 15,087 | 20,334 | +35% | 1 | 1 | 0% | 2,387 | 3,512 | +47% | 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 +73 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.