md-to-html

将Markdown格式的笔记转换为带左侧固定目录大纲的可阅读HTML文件。适用于：(1) 将学习笔记、技术文档转换为可浏览的HTML；(2) 为Markdown文件生成带目录导航的阅读界面；(3) 整理长文档生成静态网站。触发条件：用户说转换HTML、markdown转html、生成HTML笔记、把笔记转成网页等

This skill looks coherent at a high level: the bundled JS/CSS libraries match the described features and Python is the only required binary. The main risk is that the agent will run scripts/md2html.py — the scanner output omitted the full script, so you must inspect scripts/md2html.py before installing or executing. Specifically: (1) Open and review scripts/md2html.py for network calls (requests, urllib, sockets), subprocess.exec/spawn usage, or code that reads paths outside the supplied input file. (2) Search all files (including SKILL.md) for invisible Unicode control characters or suspicious string-handling that might attempt injection. (3) Run the script locally on a non-sensitive test Markdown file in a sandbox or container (no network) to observe behavior and any outbound connections. (4) If you are not comfortable reading the Python file, decline installation or run it in an isolated VM. If you share the contents of scripts/md2html.py I can do a more specific code-level review and update the verdict.

Type: OpenClaw Skill Name: md-to-html Version: 1.0.6 The skill bundle provides a legitimate utility for converting Markdown files to HTML with a table of contents and syntax highlighting. The Python script (scripts/md2html.py) uses the standard 'markdown' library and embeds local CSS/JS resources from the lib/ directory. No network activity, unauthorized file access, or obfuscated code was detected. The behavior is entirely consistent with the documentation in SKILL.md.