xhs-html AI Skill

What it is

xhs-html makes multi-page Xiaohongshu image posts. It is one of the 10 Xiaohongshu Skills in the Creator Buddy repository by 空格的键盘 (GitHub user SpaceZephyr). The README says the Skills follow the open Agent Skills standard and run in AI agents such as Claude Code, Codex, Cursor, OpenClaw and Hermes Agent, and you can install just the one you need. Give it an article, note, tutorial, checklist, web page or local document and it splits it into 6 or more 3:4 pages (1080×1440 each) inside one web page file you can screenshot page by page. Before it starts, it offers a few matching styles and a 6 to 9 page plan for you to choose from; the docs say it supports 62 styles.

Highlights:

Splits long content into 6 or more 3:4 pages
62 styles, with style and page count chosen before it starts
Outputs one web page file you can screenshot page by page

How to use:

First, download the repository: `git clone https://github.com/SpaceZephyr/creator-buddy.git`
Then copy the `xhs-Skills/xhs-html` folder into your AI agent's Skill directory.
Finally, send your content to your AI agent and ask it to make a Xiaohongshu image post.

Who it's for:

Creators who often turn tips and tutorials into Xiaohongshu image cards.

Before you start:

You need an AI agent that supports Agent Skills; page screenshots use Playwright and your local Chrome.
Screenshots run in a fresh headless browser that does not use your existing logins; if the page links to outside fonts or images, they load over the internet.
The README shows an MIT license, but the repository has no separate license file.

Based on the project's public documentation. Not tested by AI Inspiration World.