Can I edit existing images with this skill?

Yes, you can pass existing images along with text prompts to perform edits, apply style transfers, or add new elements conversationally.

What file extension should I use for saved images?

Gemini returns images in JPEG format by default. You should always use the .jpg extension when saving to avoid media type mismatch errors.

Is an API key required to use Gemini ImageGen?

Yes, you must have a valid Google Gemini API key set in your environment variables as GEMINI_API_KEY.

Does this skill support different aspect ratios?

Yes, it supports a wide variety of ratios including 1:1, 16:9, 9:16, 3:2, 4:3, 21:9, and several others for portrait and landscape needs.

What model does this skill use for image generation?

The skill defaults to the gemini-3-pro-image-preview model (Nano Banana Pro), which is optimized for high-quality generation ranging from 1K to 4K resolutions.

Gemini Image Generation

Name: Gemini Image Generation
Author: Folly-Partners

byFolly-Partners

设计工具

Generates and edits high-resolution images using the Gemini Pro API directly within your development environment.

关于

This skill integrates Google's Gemini API (Nano Banana Pro) into Claude Code, enabling developers to create, refine, and manipulate images using natural language prompts. It supports high resolutions up to 4K, custom aspect ratios, and advanced features like multi-turn chat refinement and composition from up to 14 reference images. Whether you need to generate UI mockups, design logos with embedded text, or perform complex image-to-image edits with Google Search grounding, this skill provides a robust programmatic interface for visual asset creation.

主要功能

Optimized prompting patterns for photorealism, logos, and stylized art
Composition from up to 14 reference images for complex scenes
High-resolution generation (up to 4K) with customizable aspect ratios
0 GitHub stars
Multi-turn refinement for iterative image editing via chat
Real-time image grounding using Google Search data integration

使用场景

Visualizing real-time information or data concepts through search grounding
Creating and refining brand logos with specific typography
Generating production-ready UI assets and product mockups

关于

主要功能

Optimized prompting patterns for photorealism, logos, and stylized art
Composition from up to 14 reference images for complex scenes
High-resolution generation (up to 4K) with customizable aspect ratios
0 GitHub stars
Multi-turn refinement for iterative image editing via chat
Real-time image grounding using Google Search data integration

使用场景

Visualizing real-time information or data concepts through search grounding
Creating and refining brand logos with specific typography
Generating production-ready UI assets and product mockups