gemini-imagegen

Generate and edit JPEG images via the Gemini API from text prompts.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/mengblom/dex-ghx --skill gemini-imagegen-mengblom
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gemini-imagegen
Source: https://github.com/mengblom/dex-ghx/tree/main/.claude/plugins/compound-engineering/skills/gemini-imagegen
Command: npx skills add https://github.com/mengblom/dex-ghx --skill gemini-imagegen-mengblom

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, Pillow, and includes scripts (resource) components.

What problem does it solve?

This Skill removes the friction of creating, editing, and iterating on images by giving you a repeatable way to call the Gemini API and get back usable image outputs without manual tooling.

Core Features & Use Cases

  • Text-to-Image Generation: Turn prompts into images with controllable resolution and aspect ratio for consistent outputs.
  • Image Editing & Style Transfer: Modify existing images by describing changes in natural language, including stylistic transformations.
  • Multi-turn Refinement & Composition: Iterate on image results across chat turns and combine multiple reference images into a single composed output.
  • Operational Requirements: Uses GEMINI_API_KEY for authentication and supports Gemini 3 Pro image preview as the default model.

Quick Start

Use the gemini-imagegen skill to generate an image from the prompt "Create a photorealistic product mockup of a matte-black stainless steel tumbler on polished concrete" with a wide 16:9 aspect ratio at 2K resolution.

Frequently Asked Questions about gemini-imagegen

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate images from text prompts using the Gemini API?

To generate images from text prompts using the Gemini API, you provide a text description and the skill calls the gemini-3-pro-image-preview model to return a JPEG file. You can configure the aspect ratio and resolution for consistent outputs.

Can I edit an existing image and apply a style transfer with Gemini?

You can edit an existing image and apply a style transfer with Gemini by providing the reference image and describing the desired changes in natural language. The skill processes the modifications and outputs the newly styled JPEG image.

Do I need a GEMINI_API_KEY to use this image generation skill?

Yes, you need a GEMINI_API_KEY to authenticate your requests. This skill requires the key to access the Gemini API and generate or edit images using the gemini-3-pro-image-preview model.

What's the best way to combine multiple reference images into one composition?

The best way to combine multiple reference images into one composition is to use the multi-turn refinement feature. You provide multiple reference images in the chat, and the skill merges them into a single composed output like a logo or mockup.

Does this skill support multi-turn image refinement in chat?

Yes, this skill supports multi-turn image refinement in chat. You can iteratively describe changes to your generated or edited images across conversation turns to achieve your desired result without restarting.

What image formats and resolutions does the Gemini image generation output?

The Gemini image generation outputs JPEG images saved with a .jpg extension. You can configure the aspect ratio, such as a wide 16:9 format, and specify a 2K resolution for your generated images.