The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted local file content for printing, which could contain instructions that influence the agent's behavior if the agent processes the file contents.
Ingestion points: The FILE argument in the scripts/print script and usage examples provided in SKILL.md.
Boundary markers: There are no markers or safety instructions to separate file content from agent prompts.
Capability inventory: The skill uses the lp system command which reads files for printing.
Sanitization: No sanitization of file content is performed.
[COMMAND_EXECUTION]: The scripts/print bash script constructs a command using the lp utility. It passes the file path to lp without the -- argument separator. A maliciously named file starting with a hyphen (e.g., -r) could be interpreted as a command flag, potentially causing the utility to perform unintended actions like deleting the file after printing.

print