wibi-style
47 installable visual-design skills for Codex, with reusable Python tooling and versioned packages. Turn photos into artwork. Try online: style.abdc.online
What is wibi-style?
47 installable visual-design skills for Codex, with reusable Python tooling and versioned packages. Turn photos into artwork. Try online: style.abdc.online
Overview of wibi-style
What it is
wibi-style is a collection of 47 installable visual-design skills for OpenAI Codex, provided as reusable Python tooling and versioned packages. It transforms user-submitted photos into finished artwork without requiring prompt engineering or parameter tuning. The project is created and maintained by @威比 Hunter Wei.
Key features
- 47 individually installable visual style Skills, each in an isolated directory.
- Skills are designed to work with OpenAI Codex; users select a style, supply a photo, and receive a completed design artifact.
- Each Skill introduces itself with an interactive welcome card on its first turn.
- Skills are organized by subject (e.g., portraits, pets, scenes) with style names, invocation slugs, and photo recommendations.
- A live preview is available at the Wibi Style Lab website.
Use cases
- Turning selfies, portraits, or pet photos into stylized artwork such as halftone posters, manga avatars, or crayon sketches.
- Creating themed posters, stickers, or collages from group shots, street photos, or travel scenes.
- Generating lo-fi avatars or vintage-style home journal illustrations.
- Applying multi-step effects (e.g., cropping before applying a style) for advanced use.
Getting started
To install a single Skill in Codex, provide the URL https://github.com/Vieeeeeee/wibi-style/tree/main/skills/ and have the user run or install that URL in Codex. Each style resides in an isolated directory; only the selected Skill is downloaded. To use, in a new task submit Use $call-name on this photo.
License and pricing
The project is open source. The license is not stated in the metadata (NOASSERTION).
Based on the project's public README and repository metadata.