Social Image Card Series Generator (baoyu-xhs-images)
Turns any article or idea into a 1–10 image card series for social media, with 12 styles, 8 layouts and 3 palettes.
Image & VideoIntermediate★ 24,749⑂ 2,764AI score 8/10Last updated: Jul 4, 2026
What it does
Give it an article, doc or topic and it breaks the content into a series of 1–10 infographic image cards tuned for Xiaohongshu, WeChat, Instagram and similar feeds.
- Analyzes the source into
analysis.md(topic, hook potential, audience, swipe flow) - Recommends one of three outline strategies: story-driven, information-dense, or visual-first
- Freely mixes 12 styles (cute, notion, chalkboard, study-notes, screen-print, sketch-notes…) × 8 layouts (sparse, balanced, dense, list, comparison, flow, mindmap, quadrant) × 3 palettes (macaron / warm / neon)
- Ships ~25 presets such as
--preset knowledge-cardor--preset posteras one-word shortcuts - Saves every image's full prompt to
prompts/NN-....mdso runs are reproducible and editable - Uses an image-1 anchor chain (generate the cover first, then pass it as a reference) to keep characters and colors consistent across the set
- Remembers watermark, preferred style/layout/language and backend in an
EXTEND.mdconfig
Actual rendering is delegated to whatever raster backend your runtime exposes (Codex imagegen, Cursor GenerateImage, baoyu-image-gen, …). The skill explicitly forbids faking images with SVG/HTML and forbids patching bad text by painting over the bitmap — it regenerates from a corrected prompt instead.
Who it's for
- Marketers and creators who post card-style carousels on Xiaohongshu, Instagram or WeChat
- Writers who want to repurpose blog posts or newsletters as visual summaries
- Educators who need handwritten-notes or chalkboard style infographics
- Solo operators who want a consistent look across a whole series without a designer
Examples
- "Make 5 Xiaohongshu cards from this blog post" → it analyzes the text, recommends the
cute-sharepreset, and after you confirm produces cover + 3 content cards + closing card. - "Use
--preset knowledge-card --palette macaronfor my AI tools roundup" → notion-style dense knowledge cards in soft macaron colors. - "
--ref brand.png --yes, just generate it" → skips confirmation, applies your brand reference to the cover, then chains the remaining cards off that cover for visual consistency.
· · · Install guide · · ·
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the skills/baoyu-xhs-images folder from the GitHub repo JimLiu/baoyu-skills into my ~/.claude/skills/baoyu-xhs-images/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/JimLiu/baoyu-skills.git /tmp/baoyu-skills && mkdir -p ~/.claude/skills && cp -r /tmp/baoyu-skills/skills/baoyu-xhs-images ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repo:
git clone https://github.com/JimLiu/baoyu-skills.git - Copy the skill into your Claude Code skills folder:
mkdir -p ~/.claude/skills && cp -r baoyu-skills/skills/baoyu-xhs-images ~/.claude/skills/ - Make sure an image backend is available. Runtime-native tools like Codex
imagegenor CursorGenerateImageare picked up automatically; otherwise also install thebaoyu-image-genskill from the same repo. - Restart Claude Code and trigger the skill with something like "turn this article into social image cards" or by naming
baoyu-xhs-images. - On first run it walks you through setup (watermark, default style, language) and stores it in
~/.baoyu-skills/baoyu-xhs-images/EXTEND.md. Edit that file or say "reconfigure preferences" to change it later. - Add
--yesto a request when you want it to skip the confirmation step and generate straight away.
View source on GitHub ↗License: MIT