social-card-ai-visual-generator

Generates themed Xiaohongshu visual card HTML from confirmed storyboard and fact data.

5|1|Updated Jul 31, 2026
One-click install
npx skills add https://github.com/shiker1996/wechat-editroom --skill social-card-ai-visual-generator-shiker1996
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: social-card-ai-visual-generator
Source: https://github.com/shiker1996/wechat-editroom/tree/main/skills/social-card-ai-visual-generator
Command: npx skills add https://github.com/shiker1996/wechat-editroom --skill social-card-ai-visual-generator-shiker1996

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires puppeteer, and includes scripts (resource) and references (resource) components.

What problem does it solve? Turning a confirmed storyboard into a set of visually consistent, screenshot-ready Xiaohongshu cards requires manual design work; this Skill translates verified facts and a theme spec into a complete 375x667 HTML document with clear visual hierarchy and rhythm. ## Core Features & Use Cases - Storyboard-to-HTML generation: Reads card-plan.json, fact sheets, and a theme design spec, then writes a complete ai-beautified.html with themed CSS, components, and decorations. - Semantic component mapping: Maps facts like metrics, steps, risks, code, and conclusions to components such as metric-focus, process-rail, warning-panel, and terminal-panel. - Layout and visual contract enforcement: Applies a fixed 375x667 canvas, safe areas, typography scales, and per-page visual focus rules from built-in references. - Use Case: After a content pipeline finalizes a storyboard for a product launch, use this Skill to produce a multi-page Xiaohongshu card set where each page has one primary visual focus and the theme's textures and colors are visibly applied. ## Quick Start Generate the AI visual HTML for the current candidate's storyboard using the confirmed card plan and theme design spec.

Frequently Asked Questions about social-card-ai-visual-generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate Xiaohongshu visual cards from a storyboard?

Provide a confirmed card-plan.json with page order, roles, and content blocks plus a theme design spec. The Skill reads all workspace files, plans per-page visual focus and components, then writes a complete ai-beautified.html with themed CSS.

What input files does the social card visual generator require?

It requires card-plan.json as the authoritative source for page order and facts, plus ai-visual-card-plan.json, fact sheets, social-theme-design-spec.md, and a theme snapshot. Missing or conflicting files fall back to card-plan.json as the source of truth.

What canvas size and layout rules apply to Xiaohongshu cards?

Pages are fixed at 375x667 pixels with safe areas of at least 16px horizontally and 28px vertically. Typography, spacing, and content-stack structure follow the layout guide, with content pages targeting 60-80% visual occupancy.

Can the generated card HTML use external fonts or images?

No. The HTML must be self-contained for direct screenshotting: no scripts, event handlers, @import, external fonts, remote images, or url() references are allowed. All styling and decorations must be inline CSS.

What are the limitations of AI visual card generation?

It cannot change page order, invent facts, or alter numbers, names, and causal claims from the storyboard. Page count may deviate by at most 2 from the storyboard baseline, and it does not perform fact collection, event analysis, or programmatic rendering.