Install any skill in seconds. Free to start, no credit card required.
Get Started Free →三张方形卡片轮播, 标题串联, 品牌 mark + 编号
.claude/skills/nexu-io-social-carousel/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 128% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -2% | 0% |
【模板: 三联社交轮播 / Social Carousel】 【意图】3 张 1080×1080 方形卡片, headline 跨张串联。 【布局】
【设计细节】
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→pass | 10,309 | 8,350 | -19% | 1 | 1 | 0% | 1,706 | 1,513 | -11% | 0 | 0 | — |
case-01 | fail→pass | 14,943 | 12,144 | -19% | 1 | 1 | 0% | 2,537 | 2,444 | -4% | 0 | 0 | — |
case-02 | fail→pass | 14,350 | 24,274 | +69% | 1 | 1 | 0% | 2,430 | 5,535 | +128% | 0 | 0 | — |
case-03 | fail→fail | 17,451 | 10,872 | -38% | 1 | 1 | 0% | 2,783 | 2,045 | -27% | 0 | 0 | — |
case-04 | fail→fail | 12,436 | 13,774 | +11% | 1 | 1 | 0% | 2,249 | 2,672 | +19% | 0 | 0 | — |
case-05 | fail→pass | 9,104 | 6,767 | -26% | 1 | 1 | 0% | 1,636 | 1,337 | -18% | 0 | 0 | — |
case-06 | fail→pass | 10,148 | 8,497 | -16% | 1 | 1 | 0% | 1,697 | 1,664 | -2% | 0 | 0 | — |
case-08 | pass→pass | 11,002 | 6,382 | -42% | 1 | 1 | 0% | 1,908 | 1,307 | -31% | 0 | 0 | — |
case-09 | pass→pass | 10,746 | 7,720 | -28% | 1 | 1 | 0% | 1,813 | 1,507 | -17% | 0 | 0 | — |
case-10 | pass→pass | 14,713 | 15,452 | +5% | 1 | 1 | 0% | 2,351 | 3,168 | +35% | 0 | 0 | — |
case-11 | pass→pass | 10,846 | 10,336 | -5% | 1 | 1 | 0% | 1,850 | 1,816 | -2% | 0 | 0 | — |
case-12 | fail→pass | 6,446 | 8,388 | +30% | 1 | 1 | 0% | 1,090 | 1,707 | +57% | 0 | 0 | — |
case-13 | pass→pass | 10,398 | 9,646 | -7% | 1 | 1 | 0% | 1,869 | 1,910 | +2% | 0 | 0 | — |
case-14 | pass→pass | 9,474 | 5,471 | -42% | 1 | 1 | 0% | 1,515 | 1,204 | -21% | 0 | 0 | — |
case-15 | pass→pass | 8,016 | 6,537 | -18% | 1 | 1 | 0% | 1,445 | 1,292 | -11% | 0 | 0 | — |
case-16 | pass→pass | 10,834 | 11,025 | +2% | 1 | 1 | 0% | 1,946 | 2,047 | +5% | 0 | 0 | — |
case-17 | pass→pass | 8,574 | 9,882 | +15% | 1 | 1 | 0% | 1,620 | 2,011 | +24% | 0 | 0 | — |
case-18 | pass→pass | 16,437 | 10,485 | -36% | 1 | 1 | 0% | 2,923 | 2,076 | -29% | 0 | 0 | — |
case-19 | pass→pass | 10,130 | 9,015 | -11% | 1 | 1 | 0% | 1,596 | 1,784 | +12% | 0 | 0 | — |
case-20 | fail→fail | 14,384 | 11,946 | -17% | 1 | 1 | 0% | 2,365 | 2,285 | -3% | 0 | 0 | — |
case-21 | pass→pass | 7,166 | 5,681 | -21% | 1 | 1 | 0% | 1,184 | 1,094 | -8% | 0 | 0 | — |
case-22 | fail→fail | 12,566 | 9,900 | -21% | 1 | 1 | 0% | 2,025 | 1,753 | -13% | 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.
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.