GitHub radar
The Codex skill generates illustration images for articles using a character (Xiaohei)
Ian Xiaohei Illustrations is a Codex Skill that summarizes an article, defines important points in it, and generates an illustration for each point. All illustrations have the same character.
This is a Codex Skill that lets you train AIs to draw pictures related to your Chinese articles/blogs/Notion notes/methodology articles etc. It reads an article, identifies important information (key judgment, thinking transformation, pre/post comparison, metaphors) and plans (4–8 images based on these information) before generating images with Codex’s native image generation function. For each image, there’s a corresponding image template (horizontal 16:9 aspect ratio, all white background, black outline, some red/orange/blue Chinese calligraphy, etc.). The same character (Xiaohei, a black silhouette with two white dots and skinny legs) is used across all images, and there are also some rules when generating images (e.g., Xiaohei must be doing something rather than just appearing in the image). This skill is NOT meant for drawing illustrations for businesses, PPT infographics, cartoon icons etc.
Why a vibe-coder should care
The goal of this skill isn’t to make beautiful images, but to ensure consistency in terms of image styles and reproducible results, so that each time you run this skill, you’ll get some kind of “signature” images rather than random ones. Currently, this skill only supports Chinese articles/images with Chinese captions, but the concept of using the same character, same style, same image selection criteria in a single skill file can be extended to other languages. If you’re intrigued by this skill, feel free to give it a shot! It’s open source under MIT License, and the images are generated using Codex’s native image generation function without needing to deploy another model.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Set up this Codex skill from https://github.com/helloianneo/ian-xiaohei-illustrations — clone the repo and copy the ian-xiaohei-illustrations folder into ~/.codex/skills per the README, then run one of the ready-made examples from examples/prompts.md and show me the resulting shot list, the images, and the path they were saved to.
A regular laptop is more than enough — there is no heavy computation here, just skill files for the agent. Images are drawn by Codex's own built-in generator, so you don't need to set up a separate model.
Open on GitHub▌ More finds