What model does the Gemini Image Gen skill use?

It utilizes Google's Nano Banana 2 model, which is the Gemini 3.1 Flash Image Preview.

Can I use existing images as a style reference?

Yes, you can use the -r or --reference flag to provide one or more images that guide the model toward a specific visual style.

Do I need an API key to use this skill?

Yes, you must have a valid GEMINI_API_KEY environment variable configured to authenticate with Google's services.

What file formats does it support for output?

The skill generates high-quality PNG files by default, which can be saved to any path using the -o flag.

Gemini Image Generator

Name: Gemini Image Generator
Author: clsandoval

byclsandoval

0•

デザインツール

Generates custom AI images and visual assets using Google's Gemini 3.1 Flash models directly within your development workflow.

The Gemini Image Generator skill integrates Google's advanced Nano Banana 2 model (Gemini 3.1 Flash Image Preview) into the Claude Code environment. It enables developers to generate high-quality icons, illustrations, and photorealistic images via text prompts. With support for reference images, the skill ensures stylistic consistency across generated assets, making it ideal for rapid UI prototyping, game asset creation, and visual documentation without leaving the terminal.

主な機能

01Text-to-image generation via Google Nano Banana 2 (Gemini 3.1 Flash)

02Optimized prompting for UI icons, game sprites, and illustrations

03Configurable output paths for automated file management

040 GitHub stars

05Seamless integration with Claude's natural language command triggers

06Reference image support for maintaining stylistic consistency

ユースケース

01Generating consistent concept art and sprites for game development

02Rapidly prototyping UI icons and visual elements for web and mobile applications

03Creating custom illustrations and assets for technical documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add clsandoval/monorepo gemini-image-gen

For use in Claude.ai and ChatGPT

主な機能

01Text-to-image generation via Google Nano Banana 2 (Gemini 3.1 Flash)

02Optimized prompting for UI icons, game sprites, and illustrations

03Configurable output paths for automated file management

040 GitHub stars

05Seamless integration with Claude's natural language command triggers

06Reference image support for maintaining stylistic consistency

ユースケース

01Generating consistent concept art and sprites for game development

02Rapidly prototyping UI icons and visual elements for web and mobile applications

03Creating custom illustrations and assets for technical documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add clsandoval/monorepo gemini-image-gen

For use in Claude.ai and ChatGPT