kimberlying/build-xhs-content-pack — explained in plain English
Analysis updated 2026-05-18
Turn a long X post or article into a set of Xiaohongshu image cards ready to post.
Convert book notes or reading summaries into a shareable knowledge card series.
Generate a matching WeChat public account article alongside a social media post.
Preview four different visual styles for a post before committing to one.
| kimberlying/build-xhs-content-pack | aa2448208027-code/localaihotswap | amapvoice/pilottts | |
|---|---|---|---|
| Stars | 39 | 39 | 39 |
| Language | Python | Python | Python |
| Setup difficulty | easy | moderate | hard |
| Complexity | 2/5 | 3/5 | 4/5 |
| Audience | writer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires the Codex assistant with the skill copied into its skills directory, plus the Pillow Python package for image rendering.
Build-xhs-content-pack is a Codex Skill, meaning it is a packaged workflow that plugs into the Codex AI coding assistant rather than a standalone app you run on its own. Its job is to take long form source material, things like an X post, a book's notes, a PDF, a video transcript, or just scattered personal notes, and turn it into a ready to publish content package for Xiaohongshu, the Chinese social platform also known as Little Red Book, along with a matching WeChat public account article and its own set of images. Feed it a topic and some source material, and it produces a content brief, a summary of sources used, eight to twelve image cards sized for Xiaohongshu posts, a full size preview image, a title and caption with tags, a longer WeChat article, a horizontal WeChat banner image, a pre-publish checklist, and a template for reviewing performance after the post goes live. The author built it after repeatedly doing this process by hand with Codex and eventually packaging the steps into a reusable skill. A distinguishing feature is that it offers four different visual styles for the generated cards, ranging from a magazine style knowledge layout to a book notes style, a clean tech tutorial look, and a business analysis look with a financial terminal feel. If you do not pick a style up front, it will generate four cover options in each style so you can choose visually instead of describing fonts or colors yourself. There are three modes: a standard mode that shows you all the card text first so you can approve it before any images are generated, a full mode that produces everything in one pass, and a text only mode that skips image generation entirely. Installation means copying the skill folder into your Codex skills directory and installing a small Python dependency, Pillow, used for rendering the images. The author is explicit that this is an experimental workflow still under active testing, and that the tool does not fabricate data, testimonials, or guarantees, and requires a human to confirm final facts, wording, and anything published, especially around sensitive topics like investing or health. The full README does not mention a license.
A Codex Skill that turns long articles, book notes, or PDFs into a ready to publish Xiaohongshu content pack with matching WeChat article and images.
Mainly Python. The stack also includes Python, Pillow, Codex.
The project does not state a license, so its reuse terms are unclear.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly writer.
This repo across BitVibe Labs
double-check against the repo, no cap.