managing-k8s-pod-disruption
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell scripts that invoke
kubectlandjqto interact with the cluster API. - Evidence: Multiple script blocks in SKILL.md use these utilities to discover and filter Kubernetes resources.
- [DATA_EXFILTRATION]: The skill accesses configuration data from the Kubernetes cluster across all namespaces.
- Evidence: Commands such as
kubectl get pdb --all-namespacesandkubectl get events --all-namespacesexpose cluster metadata and status information. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by processing potentially untrusted data from the Kubernetes API.
- Ingestion points: Kubernetes resource names, selectors, and event messages in SKILL.md.
- Boundary markers: Absent.
- Capability inventory: Shell command execution via
kubectlandjq, and local file write access to/tmp/in SKILL.md. - Sanitization: Absent.
Audit Metadata