Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use this skill when the user asks to generate, draw, create, or revise an image. Supports text-to-image and image-reference generation.
.claude/skills/wecode-ai-image-generation/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 9 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✓→✓ | ＝ Same ✓ | -24% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -2% | 0% |
| case-11 | ✓→✓ | ＝ Same ✓ | -43% | 0% |
| case-13 | ✓→✓ | ＝ Same ✓ | -71% | 0% |
| case-16 | ✓→✓ | ＝ Same ✓ | -40% | 0% |
Call generate_image when the user asks for an image.
prompt.reference_images only when the user supplied image attachments or URLs.error, explain it once and do not retry unchanged parameters.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | pass→pass | 24,253 | 16,740 | -31% | 1 | 1 | 0% | 2,585 | 1,962 | -24% | 0 | 0 | — |
case-01 | fail→fail | 34,620 | 22,890 | -34% | 1 | 1 | 0% | 4,349 | 789 | -82% | 0 | 0 | — |
case-02 | fail→fail | 19,616 | 31,257 | +59% | 1 | 1 | 0% | 2,571 | 1,905 | -26% | 0 | 0 | — |
case-03 | fail→fail | 9,963 | 20,263 | +103% | 1 | 1 | 0% | 769 | 984 | +28% | 0 | 0 | — |
case-04 | pass→pass | 10,012 | 9,702 | -3% | 1 | 1 | 0% | 963 | 946 | -2% | 0 | 0 | — |
case-05 | fail→fail | 3,593 | 10,226 | +185% | 1 | 1 | 0% | 463 | 907 | +96% | 0 | 0 | — |
case-07 | fail→fail | 36,398 | 20,690 | -43% | 1 | 1 | 0% | 5,937 | 834 | -86% | 0 | 0 | — |
case-08 | fail→fail | 22,323 | 16,510 | -26% | 1 | 1 | 0% | 560 | 594 | +6% | 0 | 0 | — |
case-09 | fail→fail | 18,344 | 15,548 | -15% | 1 | 1 | 0% | 1,757 | 612 | -65% | 0 | 0 | — |
case-10 | fail→fail | 4,934 | 26,256 | +432% | 1 | 1 | 0% | 674 | 853 | +27% | 0 | 0 | — |
case-11 | pass→pass | 12,386 | 4,094 | -67% | 1 | 1 | 0% | 1,274 | 726 | -43% | 0 | 0 | — |
case-12 | fail→fail | 8,081 | 10,752 | +33% | 1 | 1 | 0% | 1,272 | 760 | -40% | 0 | 0 | — |
case-13 | pass→pass | 28,975 | 32,133 | +11% | 1 | 1 | 0% | 4,679 | 1,376 | -71% | 0 | 0 | — |
case-14 | fail→fail | 15,883 | 12,872 | -19% | 1 | 1 | 0% | 1,590 | 869 | -45% | 0 | 0 | — |
case-15 | fail→fail | 14,714 | 16,624 | +13% | 1 | 1 | 0% | 1,371 | 1,094 | -20% | 0 | 0 | — |
case-16 | pass→pass | 7,906 | 6,915 | -13% | 1 | 1 | 0% | 1,409 | 847 | -40% | 0 | 0 | — |
case-17 | fail→fail | 17,085 | 13,973 | -18% | 1 | 1 | 0% | 151 | 622 | +312% | 0 | 0 | — |
case-18 | fail→fail | 8,452 | 20,312 | +140% | 1 | 1 | 0% | 1,408 | 777 | -45% | 0 | 0 | — |
case-19 | fail→fail | 14,087 | 29,569 | +110% | 1 | 1 | 0% | 1,780 | 1,132 | -36% | 0 | 0 | — |
case-20 | fail→fail | 5,501 | 11,820 | +115% | 1 | 1 | 0% | 703 | 854 | +21% | 0 | 0 | — |
case-21 | pass→pass | 7,869 | 3,566 | -55% | 1 | 1 | 0% | 1,008 | 537 | -47% | 0 | 0 | — |
case-22 | fail→fail | 11,023 | 10,342 | -6% | 1 | 1 | 0% | 1,566 | 661 | -58% | 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 7 counted toward the lift figure. The other 15 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. A headline lift is not published for this run.
Other measured skills in the registry, with their headline benchmark lift.