op7418

op7418 / guizang-social-card-skill

Public

πŸͺ§ Claude Code / Codex skill β€” generate Xiaohongshu carousels & WeChat 21:9+1:1 cover pairs. Editorial Γ— Swiss visual systems, 28 layouts, 10 themes, single-file HTML β†’ PNG. 小纒书图文 + 公众号封青对

github.comop7418guizang-social-card-skill agent-skill ai-agent anthropic claude-code claude-skill
51
5
80% credibility
Found May 28, 2026 at 282 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
HTML
AI Summary

This is a design template system that gives AI coding assistants the ability to create professional social media card images. It provides 28 pre-built layout templates and 10 color themes in two visual styles (magazine-style Editorial and clean Swiss International), specifically for creating Xiaohongshu carousel posts and WeChat cover pairs. The user describes what they want to create, their AI assistant walks through a structured workflow to design and render the images, and the user can iterate until satisfied.

How It Works

1
πŸ’‘ You have content worth sharing

You've written an article, have product photos, or want to turn your ideas into scroll-stopping social media images.

2
πŸ”Œ You connect your AI assistant to this skill

Your AI coding assistant gets a new ability - it now knows how to design social cards just like a magazine art director would.

3
You pick your visual style
πŸ“–
Editorial style

Inspired by magazines like Monocle and Kinfolk - restrained, atmospheric, great for travel and lifestyle content

πŸ“
Swiss style

Bold grids and sharp typography - perfect for product reviews, tutorials, and data-focused content

4
🎨 Your AI picks a layout and creates your images

From 28 pre-designed layouts, your assistant builds beautiful 3:4 carousel cards for Xiaohongshu or paired covers for WeChat.

5
πŸ‘€ You review the results

The images appear in your project folder. You look at them and decide if they hit the mark or need tweaking.

6
πŸ”„ You ask for adjustments

Want a different color scheme? Different images? A tighter layout? Your assistant refines the designs until you're happy.

πŸŽ‰ Your social cards are ready

Sharp, magazine-quality images are ready to post - formatted perfectly for your target platform.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 282 to 51 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is guizang-social-card-skill?

This is a Claude Code skill that generates social media card sets for Chinese platforms. You give it an article, product notes, or screenshots, and it outputs Xiaohongshu carousels or paired WeChat covers in two visual styles: Editorial (magazine-inspired, narrative) and Swiss International (grid-first, data-focused). The workflow runs entirely in HTML, using Playwright to render single-file HTML directly to PNG. It ships with 28 layout templates and 10 color themes, but custom colors are locked down to keep output consistent.

Why is it gaining traction?

The niche is specific but underserved. Developers building content for Chinese social platforms usually cobble together Figma exports or manual screenshots. This automates that pipeline inside an AI coding agent. The two visual systems cover the main use cases: lifestyle and storytelling content gets Editorial treatment, while product reviews and data summaries get Swiss grids. The image sourcing workflow (falling back from user photos to Unsplash to Pexels to Wallhaven) means you can generate complete cards without hunting for assets manually. The built-in validator catches overflow and typography violations before you ship.

Who should use this?

Content creators and marketers who work with Chinese social platforms and want to automate visual production. Developers building tools for Chinese audiences will find this useful for generating card decks programmatically. It's less useful if you need horizontal slide decks (there's a sister project for that), or if you're working with platforms outside the Xiaohongshu/WeChat ecosystem.

Verdict

A well-scoped tool for a specific workflow, with solid documentation and a clear aesthetic philosophy. At 51 stars, it's early-stage but the documentation is thorough and the skill structure is production-ready. The AGPL-3.0 license matters if you're embedding this in commercial SaaS. Credibility score sits at 0.8% -- respectable for a niche tool, but treat it as a solid experiment rather than enterprise-tested software. Worth trying if your workflow involves generating social cards programmatically through Claude Code.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.