The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/md_to_png.py uses subprocess.run to call a secondary screenshot script (html-cut). The command is executed using a list of arguments rather than a raw shell string, which significantly reduces the risk of command injection.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted Markdown content provided by the user, which presents a surface for indirect prompt injection. The skill includes basic HTML escaping to mitigate rendering-based attacks, though it lacks explicit boundary markers.
Ingestion points: The script scripts/md_to_png.py reads local Markdown files provided via command-line arguments.
Boundary markers: The skill does not implement boundary markers or instructions to ignore instructions embedded within the Markdown files.
Capability inventory: The skill has the capability to read local files, write temporary HTML files, and execute subprocesses.
Sanitization: The script performs manual HTML escaping of special characters (<, >, &) in the _escape_html function before interpolating content into the HTML template.

md-to-png