Agent Skills: openai-image-generator

openai-image-generator

Instructions

Use this skill to generate images using OpenAI's gpt-image-2 model. The skill supports:

• Text-to-image generation from prompts (images.generate)
• Image editing / composition with one or more reference images (images.edit)
• Size presets (square, landscape, portrait, 2K) or a raw WxH string
• Quality levels (low / medium / high / auto)
• Custom output paths

The API key should be set via the OPENAI_API_KEY environment variable.

Parameters

• --prompt (required): The text prompt describing the image to generate
• --output (required): Output file path for the generated image
• --reference: Optional reference image path. Can be specified multiple times. When provided, the edits endpoint is used.
• --size: Output size. Accepts a preset (square, landscape, portrait, 2K) or a raw WxH string (e.g. 1920x1080). Default: square (1024x1024).
• --quality: low, medium, high, or auto (default: auto)

Size constraints for gpt-image-2: max edge 3840px, both edges multiples of 16, aspect ratio ≤ 3:1, total pixels between 655,360 and 8,294,400.

Examples

Basic text-to-image generation

./scripts/generate.py --prompt "A serene mountain landscape at sunset" --output images/landscape.png

With reference image(s) for editing / composition

./scripts/generate.py --prompt "Same character but wearing a party hat" --reference images/character.png --output images/party.png

./scripts/generate.py --prompt "Gift basket containing items from reference images" --reference a.png --reference b.png --output basket.png

Landscape, high quality

./scripts/generate.py --prompt "Cinematic cyberpunk skyline" --size landscape --quality high --output skyline.png

Custom raw size

./scripts/generate.py --prompt "Abstract banner" --size 1920x1088 --output banner.png

Setup

Before first use, set up the virtual environment:

cd scripts && python3 -m venv venv && ./venv/bin/pip install -r requirements.txt

Set your API key:

export OPENAI_API_KEY="your-api-key-here"