k8s-cloudnative-pg-troubleshooting
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates strong security guardrails by repeatedly instructing the agent to never read, describe, or list Kubernetes Secrets to prevent credential exposure.
- [SAFE]: All investigation commands are restricted to standard diagnostic tools such as
kubectl get,describe,logs, andeventsfocused on cluster health and status. - [SAFE]: The skill enforces a "read-first, ask before changing" policy and explicitly states that the agent should never execute proposed fixes directly, instead handing them to the user for review and execution.
- [SAFE]: External references point exclusively to official CloudNativePG documentation and verified GitHub repositories for the project.
Audit Metadata