Refactor high-complexity React components in Dify frontend. Use when `pnpm analyze-component...
npx skills add ccalebcarter/purria-skills --skill "gemini-image-generator"
Install specific skill from multi-skill repository
# Description
Generate images using Google Gemini AI with text prompts and reference images. Use when creating game assets, concept art, UI mockups, promotional images, or any visual content. Supports text-to-image, image-to-image with style transfer, and multiple output sizes. Requires GEMINI_API_KEY environment variable. Triggers on requests for AI image generation, concept art, visual assets, or Gemini images.
# SKILL.md
name: gemini-image-generator
description: Generate images using Google Gemini AI with text prompts and reference images. Use when creating game assets, concept art, UI mockups, promotional images, or any visual content. Supports text-to-image, image-to-image with style transfer, and multiple output sizes. Requires GEMINI_API_KEY environment variable. Triggers on requests for AI image generation, concept art, visual assets, or Gemini images.
Gemini Image Generator
Generate images using Google Gemini's image generation capabilities.
Prerequisites
- Python 3.8+
- Google AI Studio API key
- Virtual environment with dependencies
Setup
# Navigate to scripts directory
cd scripts
# Create virtual environment
python3 -m venv venv
# Install dependencies
./venv/bin/pip install -r requirements.txt # Unix
# or
.\venv\Scripts\pip install -r requirements.txt # Windows
# Set API key
export GEMINI_API_KEY="your-api-key" # Unix
# or
$env:GEMINI_API_KEY = "your-api-key" # PowerShell
Get your API key from Google AI Studio.
Usage
Basic Text-to-Image
python generate.py --prompt "A serene mountain landscape at sunset" --output landscape.png
With Reference Image (Style Transfer)
python generate.py --prompt "Same scene but in winter" --reference landscape.png --output winter.png
Prompt Engineering Tips
For best results, structure prompts as:
[Subject] + [Style] + [Composition] + [Technical] + [Mood]
Example for game assets:
"A bio-mimetic robot with Art Nouveau brass gears and botanical vine patterns,
centered composition on transparent background, flat vector style suitable for
game sprite, warm golden hour lighting, whimsical and charming mood"
Style keywords that work well:
- Art styles: Art Nouveau, steampunk, Studio Ghibli, pixel art, vector illustration
- Technical: transparent background, game sprite, icon, UI element, seamless texture
- Mood: whimsical, dramatic, cozy, ethereal, vibrant
Parameters
| Parameter | Required | Description |
|---|---|---|
--prompt |
Yes | Text description of desired image |
--output |
Yes | Output file path (.png) |
--reference |
No | Reference image for style guidance |
Troubleshooting
| Error | Solution |
|---|---|
| API key not valid | Check GEMINI_API_KEY is set correctly |
| 403 Forbidden | API key may have IP restrictions |
| Model not found | Model names change; check Google AI docs |
| No image generated | Try simpler prompt, check API quota |
Integration with Game Assets Team
This skill is the primary image generation tool for the game-assets-team skill. Use it for:
- Concept art exploration
- UI element generation
- Character/Simulin designs
- Background and environment art
- Promotional materials
Always follow the art direction guidelines in game-assets-team for consistent visual style.
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.