Where are the generated images saved?

By default, images are saved to your macOS Desktop and automatically opened in the Preview app, though you can specify custom output paths.

Can I edit existing images with this skill?

Yes, the skill supports image-to-image tasks like restyling, adding elements, and removing backgrounds using the Gemini Pro and Flash models.

Which AI models does this skill use?

It uses Google's Nano Banana 2 (Gemini) for editing and reference-based tasks, and Imagen 4 for high-quality photorealistic text-to-image generation.

Does it support different aspect ratios?

Yes, you can specify various ratios like 1:1, 16:9, 9:16, 4:3, and more to suit social media, banners, or print requirements.

How do I choose between photorealistic and stylized images?

Use the --model imagen flag for photorealistic results or omit it to use the default Gemini Flash model for faster, more flexible artistic generation and editing.

Image Generator & Editor

Name: Image Generator & Editor
Author: adamlevoy

byadamlevoy

0•

デザインツール

Generates and edits high-quality images directly within Claude Code using Google's Gemini and Imagen models.

The image-gen skill empowers Claude Code with professional-grade visual generation and editing capabilities by leveraging Google's advanced Nano Banana and Imagen 4 APIs. It allows users to create photorealistic images, design logos, perform complex style transfers, and execute precise image edits—like background removal or restyling—through simple natural language commands. With support for multiple aspect ratios, batch generation, and macOS integration, it streamlines the creative workflow for developers and designers who need quick, high-quality visual assets without leaving their terminal environment.

主な機能

01Flexible output control with custom aspect ratios, resolutions (up to 4K), and batch counts

020 GitHub stars

03Advanced image manipulation: background removal, restyling, and style transfer

04Seamless macOS integration with Keychain security and automatic Preview opening

05Multi-model support including Imagen 4 for photorealism and Gemini for editing

06Support for multi-turn editing by referencing previously generated image paths

ユースケース

01Editing existing product photos to remove backgrounds or apply specific visual styles

02Iterating on UI/UX design concepts using style transfer from reference images

03Generating placeholder assets, logos, and hero banners for web development projects

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add adamlevoy/claude-plugins image-gen

For use in Claude.ai and ChatGPT