Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Prepare selected workspace images and information cards as WeChat Official Account cover and article assets without reading outside the active workspace.
.claude/skills/devin-axis-wechat-official-assets/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -68% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -41% | 0% |
用于把信息卡片、海报或工作区图片接入公众号图文。
upload-cover-image 或 upload-article-image 上传当前工作区内的明确文件;不得猜测或读取工作区外路径。上传图片不等于发布文章;后续仍需创建草稿并经确认发布。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 22,693 | 22,039 | -3% | 1 | 1 | 0% | 3,392 | 1,467 | -57% | 0 | 0 | — |
case-02 | fail→fail | 21,610 | 10,435 | -52% | 1 | 1 | 0% | 965 | 1,070 | +11% | 0 | 0 | — |
case-03 | fail→fail | 22,081 | 18,193 | -18% | 1 | 1 | 0% | 826 | 904 | +9% | 0 | 0 | — |
case-04 | fail→pass | 12,287 | 9,082 | -26% | 1 | 1 | 0% | 1,170 | 762 | -35% | 0 | 0 | — |
case-05 | fail→pass | 22,308 | 10,353 | -54% | 1 | 1 | 0% | 3,188 | 1,017 | -68% | 0 | 0 | — |
case-06 | pass→pass | 16,619 | 8,692 | -48% | 1 | 1 | 0% | 1,641 | 747 | -54% | 0 | 0 | — |
case-12 | fail→pass | 18,828 | 9,592 | -49% | 1 | 1 | 0% | 2,233 | 956 | -57% | 0 | 0 | — |
case-07 | pass→pass | 16,372 | 17,691 | +8% | 1 | 1 | 0% | 1,776 | 2,180 | +23% | 0 | 0 | — |
case-08 | fail→pass | 12,605 | 8,970 | -29% | 1 | 1 | 0% | 1,196 | 777 | -35% | 0 | 0 | — |
case-09 | fail→fail | 25,525 | 8,122 | -68% | 1 | 1 | 0% | 1,746 | 636 | -64% | 0 | 0 | — |
case-10 | fail→fail | 26,599 | 7,436 | -72% | 1 | 1 | 0% | 1,745 | 532 | -70% | 0 | 0 | — |
case-11 | fail→fail | 12,770 | 7,793 | -39% | 1 | 1 | 0% | 1,123 | 562 | -50% | 0 | 0 | — |
case-13 | fail→fail | 21,419 | 17,889 | -16% | 1 | 1 | 0% | 3,096 | 2,661 | -14% | 0 | 0 | — |
case-14 | fail→fail | 16,993 | 14,199 | -16% | 1 | 1 | 0% | 2,210 | 1,777 | -20% | 0 | 0 | — |
case-15 | fail→fail | 25,407 | 14,919 | -41% | 1 | 1 | 0% | 1,375 | 1,789 | +30% | 0 | 0 | — |
case-16 | fail→pass | 15,487 | 10,902 | -30% | 1 | 1 | 0% | 1,931 | 1,137 | -41% | 0 | 0 | — |
case-17 | fail→pass | 10,831 | 9,268 | -14% | 1 | 1 | 0% | 892 | 758 | -15% | 0 | 0 | — |
case-18 | fail→pass | 10,336 | 9,032 | -13% | 1 | 1 | 0% | 796 | 827 | +4% | 0 | 0 | — |
case-19 | pass→pass | 13,550 | 8,738 | -36% | 1 | 1 | 0% | 1,263 | 721 | -43% | 0 | 0 | — |
case-20 | pass→pass | 22,819 | 11,687 | -49% | 1 | 1 | 0% | 3,325 | 1,214 | -63% | 0 | 0 | — |
case-21 | fail→fail | 23,182 | 20,877 | -10% | 1 | 1 | 0% | 3,254 | 2,718 | -16% | 0 | 0 | — |
case-22 | pass→pass | 21,294 | 17,072 | -20% | 1 | 1 | 0% | 2,760 | 2,275 | -18% | 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, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +32 percentage points is the difference between those two pass rates over the 20 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.