The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run within its test suite in tests/support.py to execute its own Python scripts. This is limited to local execution of the skill's own CLI tools during automated verification processes.
[DYNAMIC_EXECUTION]: The tests/support.py file uses the importlib module to dynamically load local scripts for unit testing. This is a standard development practice for isolating components during tests and is restricted to the skill's internal directory.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied Markdown files, which represents an indirect injection surface. However, it mitigates risk by applying robust sanitization using html.escape in scripts/apply_template.py (lines 80-130) for all text and attribute values before rendering the final HTML.
Ingestion points: scripts/wechat_mp.py reads user-provided Markdown files via command-line arguments.
Boundary markers: The skill logic and user data are strictly separated; processing is guided by documentation in SKILL.md that defines clear operational scopes.
Capability inventory: The skill can read local files and write HTML output to the local file system.
Sanitization: The apply_template.py script ensures all Markdown elements are HTML-escaped during the conversion process.

wechat-mp-writer