What is xhs-article-to-images?

This repository is a workflow tool that converts long Markdown articles into image cards formatted for Xiaohongshu, the Chinese social media platform sometimes called Little Red Book. The output images are sized at 1080 by 1440 pixels, matching the 3:4 portrait format Xiaohongshu uses for image-and-text posts. The project is designed to work as a skill (a reusable plugin) for AI coding agents such as Codex or Claude. Rather than running a fully automated parser on its own, the intended workflow is: an AI agent reads the article, decides how to split it into logical page units, fills those pages into HTML templates, and then uses a built-in Playwright browser to render and export each page as a PNG file. This approach gives more control over how a long article gets divided, compared to a rigid automatic splitter. The repository includes five visual themes: a clean editorial style labeled E, a warm apricot-coffee look labeled A, a misty pine blue labeled B, an ink-and-gold style labeled C, and a night-reading gold foil labeled D. Different page layouts are available, including covers, body text pages, image-and-text combinations, quote cards, list pages, and ending pages. To use it, you clone the repository, run npm install, and install the Playwright Chromium browser. You then link the folder into your Codex or Claude skills directory. From there, two commands handle the work: one initializes a task folder for your article, and one renders the folder into PNG files. The output lands in a subfolder inside your task directory. A scale flag lets you request a higher-resolution render, such as 2x. The repository is MIT licensed. The README is written in Chinese, so readers who do not read Chinese will need a translation tool to follow the full documentation.