feiskyer/claude-code-settings / skills/nanobanana-skill
nanobanana-skill
VerifiedGenerate or edit images via Google Gemini (nanobanana). This is the DEFAULT image skill — use whenever the user asks to generate, create, or edit an image and does NOT name another provider. Triggers: "nanobanana", "generate image", "create image", "edit image", "图片生成", "生成图片", "AI绘图", "图片编辑". Do NOT use for diagrams (架构图/流程图/时序图) — draw those with Mermaid or code instead.
Nanobanana Image Generation Skill
Generate or edit images using Google Gemini API through the nanobanana tool.
Requirements
- GEMINI_API_KEY: Must be configured in
~/.nanobanana.envorexport GEMINI_API_KEY=<your-api-key> - Python3 with dependent packages installed: google-genai, Pillow, python-dotenv. They could be installed via
python3 -m pip install -r ${CLAUDE_SKILL_DIR}/requirements.txtif not installed yet. - Executable:
${CLAUDE_SKILL_DIR}/nanobanana.py
Instructions
For image generation
-
Ask the user for:
- What they want to create (the prompt)
- Desired aspect ratio/size (optional, defaults to 9:16 portrait)
- Output filename (optional, auto-generates UUID if not specified)
- Model preference (optional, defaults to gemini-nano-banana-2.1)
- Resolution (optional, defaults to 1K)
-
Run the nanobanana script with appropriate parameters:
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py --prompt "description of image" --output "filename.png" -
Show the user the saved image path when complete
For image editing
-
Ask the user for:
- Input image file(s) to edit
- What changes they want (the prompt)
- Output filename (optional)
-
Run with input images:
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py --prompt "editing instructions" --input image1.png image2.png --output "edited.png"
Available Options
Aspect Ratios (--size)
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-wide
Models (--model)
gemini-nano-banana-2.1(default) - Latest, fast, high qualitygemini-3.1-flash-image- Previous fast modelgemini-3-pro-image- Higher quality for detailed/complex scenesgemini-2.5-flash-image- Legacy model; no Google Search or thinking (auto-disabled), always outputs 1K regardless of--resolution
Resolution (--resolution)
1K(default)2K4K
Other Options
--no-search- Disable Google Search grounding (enabled by default)--no-think- Disable thinking/reasoning mode
Examples
Generate a simple image
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py --prompt "A serene mountain landscape at sunset with a lake"
Generate with specific size and output
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py \
--prompt "Modern minimalist logo for a tech startup" \
--size 1024x1024 \
--output "logo.png"
Generate landscape image with high resolution
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py \
--prompt "Futuristic cityscape with flying cars" \
--size 1344x768 \
--resolution 2K \
--output "cityscape.png"
Edit existing images
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py \
--prompt "Add a rainbow in the sky" \
--input photo.png \
--output "photo-with-rainbow.png"
Use pro model for higher quality
python3 ${CLAUDE_SKILL_DIR}/nanobanana.py \
--prompt "Detailed portrait of a cat in watercolor style" \
--model gemini-3-pro-image \
--output "cat-portrait.png"
Error Handling
If the script fails:
- Check that
GEMINI_API_KEYis exported or set in ~/.nanobanana.env - Verify input image files exist and are readable
- Ensure the output directory is writable
- If no image is generated, try making the prompt more specific about wanting an image
Best Practices
- Be descriptive in prompts - include style, mood, colors, composition
- For logos/graphics, use square aspect ratio (1024x1024)
- For social media posts, use 9:16 for stories or 1:1 for posts
- For wallpapers, use 16:9 or 21:9
- Start with 1K resolution for testing, upgrade to 2K/4K for final output
- Use gemini-nano-banana-2.1 (default) for most tasks; try gemini-3-pro-image for complex, detail-heavy scenes
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |