xiaohongshu-image-generator

根据用户提示词生成小红书配图。使用 HTML + CSS 设计可视化卡片，通过 Playwright 浏览器截图生成图片。触发场景：(1) 用户要求生成小红书笔记配图 (2) 需要生成竖版封面卡片图 (3) 用 HTML 模板生成营销配图

This skill appears to do what it claims: generate HTML templates, host them locally, and capture a screenshot for an image. Before installing/using, consider: (1) run it in a confined environment (container or restricted workspace) to isolate the local HTTP server; ensure the server is bound to localhost and not exposed externally; (2) the Python generator inserts user text directly into the HTML without escaping — avoid passing untrusted HTML or scripts in prompts; (3) the workflow expects a browser/Playwright-like screenshot capability (the SKILL.md uses a browser action) — ensure your runtime provides that and you trust the tool that will open the page; (4) review templates if you will include sensitive content. If you need higher assurance, ask the author for the full generate_image.py code (verify no external network calls) or run the script in a sandbox first.

Type: OpenClaw Skill Name: xiaohongshu-image-generator Version: 1.0.0 The skill bundle is a legitimate tool designed to generate social media images (Xiaohongshu style) by creating HTML/CSS cards and rendering them via a local HTTP server for screenshotting. The Python script `scripts/generate_image.py` uses standard libraries to handle template formatting and local hosting, and the HTML templates contain no external scripts or tracking elements. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found.