k8s-cluster-ops

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). Outsider-authored free text can enter the LLM context via Kubernetes API outputs that include user-controlled fields (e.g., kubectl logs / kubectl describe pod / kubectl get events in the helper scripts), where log lines, event messages, and annotations are arbitrary text produced by workloads or cluster users not chosen by the operating user.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill includes explicit system installation commands (apt-get/yum/brew) that modify the host and may require elevated privileges and also advises updating kubeconfig (e.g. ~/.kube/config), so it pushes the agent to change the machine's state.