The Agent Skills Directory

[SAFE]: The skill documents standard Kubernetes diagnostic procedures using kubectl commands and provides reference manifests for Ingress, Services, and Gateway API.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect injection as it processes user queries regarding cluster connectivity issues.
Ingestion points: User-provided service names, namespaces, and error messages (SKILL.md).
Boundary markers: None explicitly defined in the instructions.
Capability inventory: The skill utilizes kubectl get for status checks and kubectl exec for diagnostic tests like DNS resolution.
Sanitization: Not present, as the skill relies on the agent's interpretation of command output.
[SAFE]: The skill explicitly recommends security-hardening measures, including the enforcement of default-deny NetworkPolicies for both ingress and egress traffic to prevent unauthorized lateral movement and data exfiltration.

kubernetes-networking