This skill should be used when the user wants to generate AI images, create pictures, make transparent PNGs, perform style transfer, edit images with AI, or use the nano-banana CLI. Covers Gemini 3.1 Flash (fast, cheap) and Gemini 3 Pro (highest quality) image generation with multi-resolution output (512-4K), aspect ratios, reference images, green screen transparency, and cost tracking. Triggered by phrases like "generate an image", "create a picture", "make an icon", "transparent sprite", "nano banana", or "edit this image with AI".
From nano-banananpx claudepluginhub lifegenieai/lifegenie-claude-marketplace --plugin nano-bananaThis skill uses the workspace's default tool permissions.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Proposes cuts, reorganization, and simplification to improve document structure, clarity, and flow while preserving comprehension. Use for structural or editorial reviews.
AI image generation CLI. Default model: Gemini 3.1 Flash Image Preview (Nano Banana 2).
When the user says "init", "setup nano-banana", or "install nano-banana":
# Install dependencies for the bundled CLI tool
cd "${CLAUDE_PLUGIN_ROOT}/tools" && bun install
Prerequisites:
curl -fsSL https://bun.sh/install | bash-t): FFmpeg + ImageMagick
brew install ffmpeg imagemagickwinget install ffmpeg and winget install ImageMagickAPI key setup:
mkdir -p ~/.nano-banana
echo "GEMINI_API_KEY=<ask user for their key>" > ~/.nano-banana/.env
Get a Gemini API key at: https://aistudio.google.com/apikey
Always invoke via:
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "prompt" [options]
| Option | Default | Description |
|---|---|---|
-o, --output | nano-gen-{timestamp} | Output filename (no extension) |
-s, --size | 1K | Image size: 512, 1K, 2K, or 4K |
-a, --aspect | model default | Aspect ratio: 1:1, 16:9, 9:16, 4:3, 3:4, etc. |
-m, --model | flash | Model: flash/nb2, pro/nb-pro, or any model ID |
-d, --dir | current directory | Output directory |
-r, --ref | - | Reference image (can use multiple times) |
-t, --transparent | - | Generate on green screen, remove background (FFmpeg) |
--api-key | - | Gemini API key (overrides env/file) |
--no-search | - | Disable Google Search grounding |
--costs | - | Show cost summary |
| Alias | Model | Use When |
|---|---|---|
flash, nb2 | Gemini 3.1 Flash | Default. Fast, cheap (~$0.067/1K image). Supports image search grounding |
pro, nb-pro | Gemini 3 Pro | Highest quality needed (~$0.134/1K image). Web search grounding only |
| Size | Cost (Flash) | Cost (Pro) |
|---|---|---|
512 | ~$0.045 | Flash only |
1K | ~$0.067 | ~$0.134 |
2K | ~$0.101 | ~$0.201 |
4K | ~$0.151 | ~$0.302 |
Supported: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 4:5, 5:4, 21:9
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "minimal dashboard UI with dark theme"
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "cinematic landscape" -s 2K -a 16:9
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "quick concept sketch" -s 512
# Default (Flash - fast, cheap)
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "your prompt"
# Pro (highest quality)
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "detailed portrait" --model pro -s 2K
# Edit existing image
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "change the background to pure white" -r dark-ui.png -o light-ui
# Style transfer - multiple references
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "combine these two styles" -r style1.png -r style2.png -o combined
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "robot mascot character" -t -o mascot
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "pixel art treasure chest" -t -o chest
The -t flag automatically prompts the AI to generate on a green screen, then uses FFmpeg colorkey + despill to key out the background and remove green spill from edge pixels. Pixel-perfect transparency with no manual prompting needed.
To get a specific output dimension:
-r flag: your reference/style image-r flag: blank image in target dimensionsbun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "pixel art character in style of first image, 256x256" -r style.png -r blank-256x256.png -o sprite
Flash model automatically uses Google Search (web + image) to ground generated images in real-world references. The output shows which queries and sources were used. Disable with --no-search if not needed.
# Grounding helps with real-world subjects - the model searches for visual references
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "the Golden Gate Bridge at sunset" -a 16:9 -s 2K
# Disable grounding for purely creative/abstract prompts
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "abstract geometric pattern" --no-search
Every generation is logged to ~/.nano-banana/costs.json. View summary:
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" --costs
Write prompts as natural descriptions, not keyword lists:
"A cozy coffee shop interior with warm lighting and wooden furniture""coffee shop, cozy, warm, wood, interior"Include art style, lighting, and composition when they matter:
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "clean SaaS dashboard with analytics charts, white background"
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "cyberpunk cityscape at sunset" -a 16:9 -s 2K
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "Premium SaaS chat interface, dark mode, minimal, Linear-style aesthetic"
bun run "${CLAUDE_PLUGIN_ROOT}/tools/cli.ts" "mobile app onboarding screen" -a 9:16
The CLI resolves the Gemini API key in this order:
--api-key flagGEMINI_API_KEY environment variable.env file in current directory.env file next to the CLI script~/.nano-banana/.envGet a key at: https://aistudio.google.com/apikey