The Agent Skills Directory

[SAFE]: The skill's primary function is to provide visibility into a Kubernetes namespace for troubleshooting purposes. All command patterns (e.g., kubectl get pods, kubectl logs, kubectl top) are standard administrative tools and are used appropriately for diagnostics.
[SAFE]: Command execution via kubectl exec is limited to inspecting file system states (df, du, stat) and database health (redis-cli), which is consistent with the debugging use case. No attempts to modify cluster state or acquire unauthorized privileges were observed.
[SAFE]: The diagnostic script (scripts/diagnose.sh) is self-contained and does not involve external network connections, remote code downloads, or multi-step execution chains that could lead to unauthorized persistence or data exfiltration.
[SAFE]: No obfuscation, prompt injection attempts, or hardcoded credentials were found in the skill metadata, instructions, or scripts.

debug-buttercup