The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the execution of Python and Node.js scripts (render_xhs.py, publish_xhs.py, render_xhs.js) to perform content rendering and platform interaction. These scripts are invoked via the command line and handle local file operations.
[EXTERNAL_DOWNLOADS]: The installation process involves downloading the Chromium browser binary via the Playwright library (playwright install chromium), which is necessary for the image rendering functionality. This download is from a well-known, trusted source (Microsoft/Playwright).
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied Markdown data to generate visual content, which represents an indirect injection attack surface.
Ingestion points: markdown_file argument processed by the rendering scripts in scripts/.
Boundary markers: No explicit boundary markers or "ignore" instructions are present in the provided templates.
Capability inventory: The skill can write files to the local directory (rendered PNG images) and perform network operations to publish content to Xiaohongshu.
Sanitization: Content is processed using standard libraries (markdown for Python, marked for Node.js) and rendered within the Playwright browser environment, which provides process-level isolation.

xhs-note-creator