
Gpt Image
Wraps the GPT Image 2 API with a 162-prompt reference gallery, a prompt-craft checklist, and a CLI that routes text-to-image, multi-reference edits, and inpainting through the right endpoints. You ask for a poster or UI mockup, it searches the bundled gallery for matching patterns, optionally confers on direction, then calls the packaged gpt-image binary or generate.py script. The operating loop is built to avoid reinstalling tools or rewriting image code on every run. Quality and size flags map to concrete use cases: low for drafts, high for Chinese text and diagrams, portrait for posters, 4k for print. Reads OPENAI_API_KEY from the environment and bills your account. Useful if you generate images often enough that you want curated prompt patterns and stable CLI behavior instead of one-off SDK calls.
- skills.sh installs
- 585
- GitHub stars
- 6.1K
- last updated
- May 23, 2026
Install
$ npx -y skills add wuyoscar/gpt-image2-skill --skill gpt-image --agent claude-codeInstalls into .claude/skills of the current project.
Files
gpt-image
Agent runbook for GPT Image 2 / 2.5 / 2.6 generation/editing. Use the prompt library + packaged CLI. Do not reimplement image API code.
Operating loop
Sponsors
Details
- Repository
- github.com/wuyoscar/gpt-image2-skill
- Categories
- Backend & APIs, AI & Agent Building, CLI & Terminal, Generative Media
- License
- MIT
- First seen
- Jun 11, 2026