The Agent Skills Directory

[SAFE]: No security issues detected. The skill is purely informational and serves as a router to other tools. It does not perform any file system, network, or process operations.
[DATA_EXFILTRATION]: No network operations or sensitive data access patterns were identified. The skill's output is limited to text recommendations.
[PROMPT_INJECTION]: The skill includes a security rule that instructs the agent to treat infrastructure details or credentials supplied by the user as data only, which is a best practice for preventing instruction injection via user-supplied content.

ops-guide