The Agent Skills Directory

[SAFE]: The skill is purely informational and contains no executable code or scripts.
[NO_CODE]: The analysis confirmed that no scripts, binaries, or configuration files for automated execution are included in this skill.
[INDIRECT_PROMPT_INJECTION]: The skill incorporates external data via the $ARGUMENTS variable.
Ingestion points: The $ARGUMENTS placeholder in SKILL.md allows user-provided data to influence the agent's context.
Boundary markers: None are present to delimit the $ARGUMENTS input.
Capability inventory: None. The skill does not utilize any tools, file-system access, or network operations.
Sanitization: None. The input is interpolated directly into the prompt.

qa-retrospective