What models are supported by this skill?

This skill primarily utilizes the gemini-3-pro-image-preview model, which is optimized for high-quality generation, editing, and multi-turn refinement tasks.

Why does the skill documentation emphasize using .jpg extensions?

The Gemini API returns images in JPEG format by default. Saving these as .png without explicit conversion causes media type mismatch errors in many image viewers.

How many reference images can I use for a single generation?

You can provide up to 14 reference images, allowing the model to combine elements, styles, or subjects from multiple sources into a single composition.

Can I edit existing images with this skill?

Yes, the skill supports image-to-image workflows where you provide an existing image and a text prompt to describe the desired modifications or additions.

Does this skill require an API key?

Yes, you must have the GEMINI_API_KEY environment variable set in your environment for the skill to authenticate with Google's GenAI services.

Gemini Image Generation

Name: Gemini Image Generation
Author: gvkhosla

bygvkhosla

•

Herramientas de Diseño

Generates and edits high-resolution images using the Gemini API with support for multi-turn refinement and reference-based composition.

The Gemini Image Generation skill empowers Claude to interact directly with Google's Gemini API to create, modify, and refine visual assets. It supports a wide range of tasks from generating photorealistic scenes and stylized art to creating product mockups and logos with embedded text. With advanced features like Google Search grounding for real-time data visualization and the ability to compose images from up to 14 reference sources, this skill provides a robust framework for handling complex image-centric workflows. It includes specific guardrails for file handling, ensuring generated JPEGs are saved with correct extensions to maintain compatibility across development environments.

Características Principales

0135 GitHub stars

02Multi-turn refinement for iterative image editing through conversational chat

03High-resolution output supporting 1K, 2K, and 4K resolutions

04Real-time data visualization via Google Search grounding integration

05Flexible aspect ratios ranging from square (1:1) to ultra-wide (21:9) and portrait (9:16)

06Composition of new images using up to 14 reference images as input

Casos de Uso

01Editing existing visual assets by applying style transfers or adding elements via semantic masking

02Automated generation of product mockups and marketing assets from text descriptions

03Iterative logo design and UI element creation using multi-turn conversational feedback

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add gvkhosla/compound-engineering-pi gemini-imagegen

For use in Claude.ai and ChatGPT

Download Skill