The Agent Skills Directory

[SAFE]: The skill consists entirely of reference documentation and implementation recipes for the VitePress framework. The provided files contain legitimate configuration examples and instructional content without any malicious behavior.
[EXTERNAL_DOWNLOADS]: The documentation includes examples of fetching data from external API endpoints (e.g., api.example.com) and installing standard development dependencies like sass, less, and markdown-it-mathjax3. These are documented as part of the tool's standard data-loading and styling features.
[COMMAND_EXECUTION]: The skill provides a reference for the VitePress CLI, documenting standard commands for initialization, development, and production building (vitepress dev, vitepress build, etc.).
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: Processes Markdown content, local data files, and responses from external APIs during the site build process.
Boundary markers: No specific security boundaries are defined; the skill follows the standard VitePress build pipeline.
Capability inventory: Accesses the local file system (via Node.js fs), performs network requests (via fetch), and executes site compilation via the CLI.
Sanitization: Content rendering is managed by the VitePress/Vue compiler, which is a standard behavior for this toolset. This represents a functional surface inherent to the tool's purpose.

vitepress