openai-image-gen

Generate batches of images via the OpenAI Images API with a local HTML gallery.

Updated Mar 7, 2026
One-click install
npx skills add https://github.com/shalevamin/Tau_agent --skill openai-image-gen-shalevamin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openai-image-gen
Source: https://github.com/shalevamin/Tau_agent/tree/main/tau-agent-main/skills/openai-image-gen
Command: npx skills add https://github.com/shalevamin/Tau_agent --skill openai-image-gen-shalevamin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Manually crafting prompts and calling the OpenAI Images API one image at a time is slow and repetitive. This Skill automates batch image generation with structured random prompts, model-aware parameter defaults, and a browsable local gallery of results. ## Core Features & Use Cases - Batch Generation: Create multiple images in one run using gpt-image-1, gpt-image-1.5, dall-e-3, or dall-e-2, with automatic per-model defaults for size and quality. - Random Prompt Sampler: Combines subjects, styles, and lighting descriptors into varied prompts when no explicit prompt is given. - Model-Specific Options: Supports background transparency and output formats (png, jpeg, webp) for GPT image models, and vivid/natural styles for dall-e-3. - Use Case: A designer needs 16 concept variations for a landing page hero image. Run the script with a count of 16, then open the generated index.html gallery to review thumbnails and pick the best option. ## Quick Start Run the gen.py script with python3 and your OPENAI_API_KEY set, then open the generated index.html gallery to review the images.

Frequently Asked Questions about openai-image-gen

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

FAQPage Schema
How do I generate multiple images with the OpenAI Images API?

Run the gen.py script with the --count flag to generate multiple images in one batch. Each image is saved to a timestamped output directory along with a prompts.json file and an index.html gallery for browsing results.

What is the difference between gpt-image-1 and dall-e-3 for image generation?

gpt-image-1 supports background transparency, output formats like webp and jpeg, and batch generation, while dall-e-3 is limited to one image per request but offers vivid and natural style options. The script applies correct defaults for each model automatically.

Can I generate transparent background images with the OpenAI API?

Yes, GPT image models support transparent backgrounds via the --background transparent flag. This option is ignored for dall-e-2 and dall-e-3, which do not support the background parameter.

Why does dall-e-3 only generate one image at a time?

The dall-e-3 API only accepts n=1 per request, so the script automatically reduces the count to 1 and prints a warning. Use gpt-image-1 or dall-e-2 if you need multiple images per run.

Why does image generation time out when run through an exec tool?

Image generation requests can take longer than common 30-second exec timeouts. Set a higher timeout such as 300 seconds when invoking the script to avoid premature termination and retries.