Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate or edit images using Google Gemini API via nanobanana. Use when the user asks to create, generate, edit images with nanobanana, or mentions image generation/editing tasks.
.claude/skills/microck-nanobanana-skill/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 105% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 227% | 0% |
Generate or edit images using Google Gemini API through the nanobanana tool.
~/.nanobanana.env or export GEMINI_API_KEY=<your-api-key>python3 -m pip install -r ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/requirements.txt if not installed yet.${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.pybash python3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py --prompt "description of image" --output "filename.png"
bash python3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py --prompt "editing instructions" --input image1.png image2.png --output "edited.png"
1024x1024 (1:1) - Square832x1248 (2:3) - Portrait1248x832 (3:2) - Landscape864x1184 (3:4) - Portrait1184x864 (4:3) - Landscape896x1152 (4:5) - Portrait1152x896 (5:4) - Landscape768x1344 (9:16) - Portrait (default)1344x768 (16:9) - Landscape1536x672 (21:9) - Ultra-widegemini-3-pro-image-preview (default) - Higher qualitygemini-2.5-flash-image - Faster generation1K (default)2K4Kbashpython3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py --prompt "A serene mountain landscape at sunset with a lake"
bashpython3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py \ --prompt "Modern minimalist logo for a tech startup" \ --size 1024x1024 \ --output "logo.png"
bashpython3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py \ --prompt "Futuristic cityscape with flying cars" \ --size 1344x768 \ --resolution 2K \ --output "cityscape.png"
bashpython3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py \ --prompt "Add a rainbow in the sky" \ --input photo.png \ --output "photo-with-rainbow.png"
bashpython3 ${CLAUDE_PLUGIN_ROOT}/skills/nanobanana-skill/nanobanana.py \ --prompt "Quick sketch of a cat" \ --model gemini-2.5-flash-image \ --output "cat-sketch.png"
If the script fails:
GEMINI_API_KEY is exported or set in ~/.nanobanana.env| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 6,990 | 2,744 | -61% | 1 | 1 | 0% | 1,102 | 1,724 | +56% | 0 | 0 | — |
case-02 | fail→fail | 6,004 | 3,309 | -45% | 1 | 1 | 0% | 932 | 1,597 | +71% | 0 | 0 | — |
case-03 | fail→pass | 23,449 | 4,393 | -81% | 1 | 1 | 0% | 4,375 | 1,751 | -60% | 0 | 0 | — |
case-04 | fail→fail | 3,562 | 6,765 | +90% | 1 | 1 | 0% | 525 | 1,604 | +206% | 0 | 0 | — |
case-05 | fail→fail | 12,487 | 10,103 | -19% | 1 | 1 | 0% | 1,965 | 2,085 | +6% | 0 | 0 | — |
case-06 | fail→fail | 5,614 | 6,758 | +20% | 1 | 1 | 0% | 531 | 1,588 | +199% | 0 | 0 | — |
case-07 | pass→fail | 9,286 | 5,791 | -38% | 1 | 1 | 0% | 502 | 1,580 | +215% | 0 | 0 | — |
case-08 | pass→fail | 4,613 | 6,435 | +39% | 1 | 1 | 0% | 665 | 1,533 | +131% | 0 | 0 | — |
case-09 | fail→fail | 39,723 | 2,344 | -94% | 1 | 1 | 0% | 6,166 | 1,637 | -73% | 0 | 0 | — |
case-10 | fail→fail | 5,565 | 6,341 | +14% | 1 | 1 | 0% | 647 | 1,573 | +143% | 0 | 0 | — |
case-11 | fail→fail | 5,346 | 8,198 | +53% | 1 | 1 | 0% | 771 | 1,735 | +125% | 0 | 0 | — |
case-12 | fail→fail | 37,414 | 7,251 | -81% | 1 | 1 | 0% | 6,164 | 1,646 | -73% | 0 | 0 | — |
case-13 | fail→fail | 35,245 | 8,510 | -76% | 1 | 1 | 0% | 6,170 | 1,817 | -71% | 0 | 0 | — |
case-14 | fail→fail | 10,671 | 10,469 | -2% | 1 | 1 | 0% | 1,422 | 1,995 | +40% | 0 | 0 | — |
case-15 | fail→fail | 4,662 | 7,665 | +64% | 1 | 1 | 0% | 565 | 1,683 | +198% | 0 | 0 | — |
case-16 | fail→pass | 8,482 | 2,746 | -68% | 1 | 1 | 0% | 1,458 | 1,775 | +22% | 0 | 0 | — |
case-17 | fail→pass | 5,320 | 2,051 | -61% | 1 | 1 | 0% | 815 | 1,672 | +105% | 0 | 0 | — |
case-22 | pass→pass | 4,287 | 5,534 | +29% | 1 | 1 | 0% | 729 | 2,003 | +175% | 0 | 0 | — |
case-18 | fail→pass | 3,472 | 2,400 | -31% | 1 | 1 | 0% | 512 | 1,676 | +227% | 0 | 0 | — |
case-19 | pass→pass | 7,846 | 3,425 | -56% | 1 | 1 | 0% | 1,145 | 1,810 | +58% | 0 | 0 | — |
case-20 | pass→pass | 18,903 | 13,953 | -26% | 1 | 1 | 0% | 3,313 | 3,764 | +14% | 0 | 0 | — |
case-21 | pass→pass | 8,903 | 11,530 | +30% | 1 | 1 | 0% | 1,697 | 3,123 | +84% | 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 11 counted toward the lift figure. The other 11 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 +14 percentage points is the difference between those two pass rates over the 11 comparable cases. 2 cases got worse with the skill loaded, and they are 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.