---

name: gemini-image description: Generate images using Google Gemini/Imagen API. Use when the user asks to create, generate, or make an image, picture, photo, illustration, artwork, or visual content. allowed-tools: Bash(python:*), Read, Write

Gemini Image Generation

Generate images using Google's Gemini and Imagen models.

Prerequisites

The user must have:

1. A Google API key with Gemini access set as GEMINI_API_KEY environment variable
2. Python packages: pip install google-genai pillow

Available Models

Generating Images

Use the script at ~/.claude/skills/gemini-image/scripts/generate.py:

Copypython ~/.claude/skills/gemini-image/scripts/generate.py \
  --prompt "A serene mountain landscape at sunset" \
  --output ~/Pictures/mountain.png \
  --model gemini-2.5-flash-image

Script Arguments

Aspect Ratio Options

• 1:1 - Square (default)
• 16:9 - Widescreen landscape
• 9:16 - Portrait/mobile
• 4:3 - Standard landscape
• 3:4 - Standard portrait
• 3:2, 2:3 - Photo aspect ratios

Usage Workflow

1. Ask the user what they want to generate
2. Craft a detailed prompt (more detail = better results)
3. Choose appropriate model and settings
4. Run the generation script
5. Show the user where the image was saved
6. Offer to regenerate with different settings if needed

Prompt Tips

• Be specific and descriptive
• Include style keywords: "photorealistic", "watercolor", "digital art", "3D render"
• Mention lighting: "golden hour", "studio lighting", "dramatic shadows"
• Specify perspective: "aerial view", "close-up", "wide angle"
• Add mood: "serene", "energetic", "mysterious"

Example Prompts

Product shot:

CopyProfessional product photography of a sleek smartphone on a marble surface,
soft studio lighting, clean white background, 8K detail

Artwork:

CopyDigital painting of a cyberpunk city at night, neon lights reflecting on
wet streets, flying cars, towering skyscrapers, cinematic composition

Portrait:

CopyProfessional headshot of a business executive, neutral background,
soft lighting, confident expression, high resolution

Troubleshooting

• API key error: Ensure GEMINI_API_KEY is set in your environment
• Package not found: Run pip install google-genai pillow
• Rate limit: Wait a few seconds and retry, or use a different model
• Content blocked: Rephrase the prompt to avoid restricted content