The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns or executable code detected. The skill consists exclusively of Markdown documentation and structural guidelines for an AI agent's reasoning process.
[Indirect Prompt Injection] (INFO): The skill manages untrusted user-provided tasks. While this constitutes an inherent injection surface, the skill implements procedural safeguards by requiring a separate 'Plan' phase and an 'Iterate' review phase that explicitly includes assessing security risks before and after code changes.
[COMMAND_EXECUTION] (INFO): The skill mentions running validation commands (e.g., tests, lint, type-check) in the 'Implement' phase. This is a standard development workflow and does not involve hardcoded or unsafe command construction within the skill itself.

rpi