What image formats are supported?

The skill supports common formats including PNG, JPEG, and WebP for both input reference images and output generation.

Does this skill require a separate API key?

Yes, you must set the GEMINI_API_KEY environment variable using a key obtained from Google AI Studio to use this skill.

What is the maximum resolution supported?

This skill supports generating images in 1K, 2K, and 4K resolutions with flexible aspect ratios like 16:9, 4:3, and 1:1.

Why does the skill use a file-based approach for API calls?

Large Base64-encoded images exceed standard command-line argument limits. Using temporary JSON files prevents 'argument list too long' errors during the upload process.

Can I edit existing images using this skill?

Yes, you can provide a path to an existing image and use natural language to describe changes, such as adding objects or modifying the background.

Gemini Image Generator

Name: Gemini Image Generator
Author: dair-ai

bydair-ai

•

设计工具

Generates and edits high-quality images directly within your development workflow using Gemini's advanced vision models.

The Image Generator skill integrates Google's Gemini-3-pro-image-preview model into Claude Code, enabling developers to create, edit, and iterate on visuals without leaving the terminal. It supports complex tasks such as text-to-image generation with accurate text rendering, semantic image editing, style transfer, and multi-image composition. By handling the heavy lifting of Base64 encoding and file-based API requests, this skill allows for seamless UI/UX prototyping, logo creation, and visual documentation generation through simple conversational commands.

主要功能

01Real-time visual grounding using Google Search for data-driven image creation

02Multi-image composition to combine elements from up to 14 reference images

0381 GitHub stars

04Semantic image editing including inpainting, outpainting, and element removal

05Precise text rendering within images for logos, diagrams, and infographics

06High-resolution image generation up to 4K with various aspect ratios

使用场景

01Creating custom logos, icons, and technical infographics for documentation

02Prototyping UI/UX designs and generating placeholder assets for web applications

03Iteratively editing existing images to change backgrounds, styles, or specific subjects

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add dair-ai/dair-academy-plugins image-generator

For use in Claude.ai and ChatGPT

Download Skill