imagegen

Generate and edit raster images using built-in image tools or a GPT Image CLI fallback.

1|Updated Jul 2, 2026
One-click install
npx skills add https://github.com/tarabakz25/my-skills --skill imagegen-tarabakz25
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: imagegen
Source: https://github.com/tarabakz25/my-skills/tree/main/.system/imagegen
Command: npx skills add https://github.com/tarabakz25/my-skills --skill imagegen-tarabakz25

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai, pillow, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Creating bitmap assets like photos, illustrations, sprites, mockups, and transparent cutouts normally requires design tools or manual API scripting; this Skill gives agents a governed workflow for generating and editing raster images with clear mode selection and output handling. ## Core Features & Use Cases - Dual-mode generation and editing: Uses the built-in image_gen tool by default, with a scripts/image_gen.py CLI fallback (generate, edit, generate-batch subcommands) for explicit API/model control via gpt-image-2 or gpt-image-1.5. - Transparent background workflow: Generates subjects on flat chroma-key backgrounds and removes them locally with remove_chroma_key.py, escalating to true native transparency via gpt-image-1.5 only after user confirmation. - Structured prompt engineering: Classifies requests into a use-case taxonomy (product-mockup, ui-mockup, photorealistic-natural, etc.) and normalizes prompts into a labeled spec with constraints and invariants. - Use Case: A developer needs a hero image for a landing page; the agent generates a product-mockup image with the built-in tool, validates it, and copies the final PNG into the project workspace. ## Quick Start Ask the agent to generate a hero image of your product for the landing page and save it into the project workspace.

Frequently Asked Questions about imagegen

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

FAQPage Schema
How do I generate an image with a transparent background?▼

Generate the subject on a flat chroma-key background with the built-in image_gen tool, then run remove_chroma_key.py locally to produce an alpha PNG. True native transparency requires the CLI fallback with gpt-image-1.5 --background transparent, used only after user confirmation.

How do I batch generate multiple image assets?▼

On the default path, issue one built-in image_gen call per requested asset. In explicit CLI mode, use the generate-batch subcommand with a JSONL file of prompts, an --out-dir, and --concurrency to run jobs in parallel.

What is the difference between gpt-image-2 and gpt-image-1.5?▼

gpt-image-2 is the CLI default with flexible sizes up to 3840px and always-high input fidelity, but it does not support background=transparent. gpt-image-1.5 supports true transparent backgrounds and input_fidelity settings, limited to fixed sizes like 1024x1024.

Does the CLI fallback require an OpenAI API key?▼

Yes, scripts/image_gen.py requires OPENAI_API_KEY set as an environment variable plus network access for live API calls. The built-in image_gen tool mode does not require the key, and --dry-run works without it.

When should I not use AI image generation for assets?▼

Avoid it when extending an existing SVG/vector icon or logo system, creating simple diagrams better done in HTML/CSS/canvas, or editing assets that already exist in an editable native format. Those tasks are better handled with code-native output.