Xhs Md2img

Convert Markdown text to beautiful Xiaohongshu (XHS) style card images with 5 themes, deterministic browser screenshot rules, auto-pagination, smart title ex...

This skill will render your Markdown in a headless-browser pipeline and may send the input (and generated images) to third-party image/LLM services (DashScope / Google Gemini / others) and optionally upload results to OSS. The metadata only declares DASHSCOPE_API_KEY, but the docs reference additional env vars (LLM_API_KEY, LLM_BASE_URL, OSS credentials) and heavy Python dependencies (playwright, openai, oss2). Before installing or enabling: 1) Confirm which API keys the skill actually requires and where data will be sent/stored; 2) Do not provide sensitive or private content unless you trust the remote providers and the key owners; 3) Ensure your runtime has the expected tooling (python + playwright) or ask the author for an explicit install spec; 4) Ask the publisher to declare all optional/used environment variables and explain OSS upload behavior and retention/logging policies. If you cannot verify those points, treat the skill with caution or run it in an isolated environment.

Type: OpenClaw Skill Name: xhs-md2pic Version: 0.1.1 The skill bundle is a legitimate utility designed to convert Markdown text into styled images for social media (Xiaohongshu). It uses standard libraries such as Playwright for headless browser rendering and integrates with established AI APIs (DashScope and Gemini) for content formatting and background generation. While it requires network and filesystem permissions, these are strictly aligned with its stated purpose of fetching AI-generated assets and managing templates/output. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found across the code or documentation.