The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using kubectl to manage and inspect Kubernetes resources. It specifically uses kubectl exec to run diagnostic commands such as curl for metrics/health checks and cat for configuration verification inside the ingress controller pods.
[DATA_EXFILTRATION]: Network operations are limited to internal cluster diagnostics, specifically targeting localhost endpoints within the pods via kubectl exec. No evidence of data transmission to external or untrusted domains was found.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted data (annotations and metadata) from Kubernetes Ingress resources. However, it mitigates risk by using structured parsing with jq and focusing on read-only discovery of resource states.

managing-nginx-ingress