gemini-image_skill

This skill generates high-quality images from prompts using Google's Gemini and Imagen models, offering multiple outputs, aspect ratios, and resolutions.
  • Python

1

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill akrindev/google-studio-skills --skill gemini-image

  • SKILL.md10.7 KB

Overview

This skill generates high-quality images from text prompts using Google Gemini and Imagen models via executable scripts. It supports custom aspect ratios, multiple image variations, and high-resolution outputs up to 4K for professional assets. Use it to produce photorealistic or artistic visuals, batch images, and integrate images into content pipelines.

How this skill works

Run the provided image-generation script with a text prompt and optional flags (model, size, aspect, number, output directory, and person policy). The script selects the requested Gemini or Imagen model, renders 1–4 PNG images, and saves them with predictable filenames (timestamped by default). It prints saved file paths and basic error messages for issues like unsupported sizes or model availability.

When to use it

  • Create visual assets from descriptive text for marketing, blogs, or product pages
  • Generate multiple design variations for A/B testing or creative exploration
  • Produce images in specific resolutions and aspect ratios (social, thumbnail, print)
  • Create photorealistic product shots using Imagen 4 or high-detail 4K assets with Gemini 3 Pro
  • Batch-produce images as part of an automated content pipeline

Best practices

  • Write concise, descriptive prompts including style, lighting, and composition details
  • Start with 1K/2K previews before committing to 4K to save time and cost
  • Choose gemini-3-pro-image-preview for maximum quality and 4K, imagen-4.0 for photorealism, and gemini-2.5-flash for speed
  • Use --num to generate variations and test prompts iteratively
  • Respect person generation policies and regional restrictions when enabling person controls

Example use cases

  • Generate a 1:1 2K image for an Instagram post: quick social media assets
  • Create a 16:9 2K YouTube thumbnail with vibrant composition and readable text
  • Produce four variations of a concept for design review and A/B testing
  • Render a 4K architectural visualization for print or high-end presentations
  • Combine with text generation to produce product descriptions and matching product images

FAQ

Use gemini-3-pro-image-preview for the highest quality and 4K outputs; imagen-4.0 is best for photorealistic product shots.

How many images can I generate per request?

The script supports generating 1–4 images per request using the --num flag.

Why did generation fail with 'unsupported size'?

Some models restrict aspect ratios and max sizes; switch to a supported size or use gemini-3-pro-image-preview for 4K.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational