The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard diagnostic commands to investigate service health and connectivity.
Evidence: Commands including dig, curl, openssl, ss, journalctl, pg_isready, and kubectl are defined in the 'Discriminating-signal table' in SKILL.md for fault localization.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external system signals, which is a potential vector for indirect prompt injection.
Ingestion points: Data enters the agent context through the output of commands such as journalctl, curl, and kubectl describe pod (defined in SKILL.md).
Boundary markers: The skill mandates an 'AI Self-Check' and strict operational 'Rules' (e.g., 'No fix before localization', 'Surface errors') that limit the agent's susceptibility to embedded instructions by requiring evidence-based reasoning.
Capability inventory: The skill utilizes standard system utilities including kubectl, dig, curl, openssl, ss, journalctl, and pg_isready (referenced in SKILL.md).
Sanitization: The skill focuses on error reporting and differentiating failure modes rather than explicit input sanitization.

debug-triage