k8s-storage
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill includes capabilities to read cluster data (e.g.,
get_events,describe_pod) which could contain untrusted input from the environment. This data is handled by an agent with write permissions (kubectl_apply), creating a potential surface for indirect injection. - Ingestion points:
SKILL.md(Troubleshooting Storage section viaget_eventsanddescribe_podtools). - Boundary markers: Absent.
- Capability inventory:
kubectl_apply,kubectl_delete(cluster resource modification). - Sanitization: Absent.
- [Command Execution] (SAFE): The skill utilizes tools to execute Kubernetes administrative tasks. These operations are explicitly declared and align with the primary purpose of storage management.
Audit Metadata