Open-Source Skill Converts Long-Form Text to Xiaohongshu Image Sets with 5 Styles
yihui_indie · x · 2026-08-07
Developer @xiaohua888 open-sourced an Agent Skill (xhs-article-images-skill) that automatically renders long-form Markdown articles into Xiaohongshu-ready 3:4 (1242×1656) image sets.
Key Features:
- Pain Point Solved: Addresses image loss issues in Xiaohongshu's native backend by using a deterministic pipeline (Markdown → markdown-it parsing → CSS styling → Playwright screenshots), bypassing text blurring or omissions common with text-to-image models.
- Smart Layout: Features semantic pagination to prevent orphaned titles at page bottoms, intelligently manages text-image fitting, and splits long paragraphs by sentence.
- Multiple Styles: Includes 5 built-in styles such as 'Rational Warm', 'Elegant Yellow', and 'Inspiration Notes'.
- Compatibility: Easily installable via npx skills add, supporting major AI coding tools like Cursor, Claude Code, and Codex.
More from coding & agent
- Dev Releases Local Terminal Voice Input Extension Based on Nemotron ASR — Danmoreng · 2026-08-07
- Practical Tip: Using Apple Reminders as an AI Agent Kanban Board — willccbb · 2026-08-07
- OpenAI, Microsoft, Cursor, and Others Unite on Shared AI Agent Plugin Standard — The Decoder · 2026-08-07
- Consult Ralph: An Open-Source Deep Research Agent for Consultants — tom_doerr · 2026-08-07
- Winning a Kaggle Silver Medal Using Claude Code — AndLukyane · 2026-08-07
- GitHub Outage Exposes Agent Bottleneck: Centralized Control Planes Limit Automation — amu4biz · 2026-08-07