The Agent Skills Directory

[SAFE]: The skill follows established best practices for document generation by separating content (Markdown) from presentation logic (Word reference templates). This architecture ensures that the rendering process is deterministic and consistent.
[COMMAND_EXECUTION]: The skill uses a shell script scripts/render.sh to interface with the system's pandoc and unzip utilities. The script correctly quotes variables to prevent command injection and utilizes standard system tools for their intended purposes without escalating privileges.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied Markdown content, which creates a potential surface for indirect prompt injection. However, this risk is mitigated by the use of pandoc, a robust and specialized parser that treats input as document structure rather than executable instructions.
Ingestion points: Markdown content files passed as the first argument to scripts/render.sh.
Boundary markers: None explicitly implemented; the system relies on the structural constraints of the Markdown format and pandoc's parser.
Capability inventory: Execution of pandoc and unzip within the local shell environment.
Sanitization: Shell arguments are properly quoted in scripts/render.sh to ensure data is treated as file paths and not as shell commands.

docx