The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using pandoc for document rendering and open or xdg-open to display the resulting files. These are standard system operations for the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes document content from outputs/<slug>.md which may originate from untrusted external sources or previous agent steps. While the conversion commands are standard, processing untrusted documents carries a inherent risk that the content might contain instructions designed to influence the agent.
Ingestion points: The content of files located at outputs/<slug>.md (SKILL.md).
Boundary markers: Absent; the file content is processed directly by the rendering utility without explicit delimiters or instructions to ignore embedded commands.
Capability inventory: File system read/write access via pandoc and the ability to launch system viewers via open/xdg-open (SKILL.md).
Sanitization: The instructions do not specify any validation or sanitization of the input artifact content before it is passed to the conversion tools.

preview