kubernetes-expert

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting data from Kubernetes resources and container logs, which represents a potential surface for indirect prompt injection if those sources contain adversarial content.\n
  • Ingestion points: Command outputs from kubectl logs, kubectl get, and kubectl describe as documented in SKILL.md.\n
  • Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore' instructions for the data retrieved from the cluster.\n
  • Capability inventory: The skill is authorized to use kubectl exec, helm install, and local file operations (Read, Write, Edit) to manage the environment.\n
  • Sanitization: There are no defined processes for sanitizing or validating the integrity of logs or metadata retrieved from the cluster before the agent processes them.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 06:41 PM
Security Audit — agent-trust-hub — kubernetes-expert