The Agent Skills Directory

[SAFE]: The skill provides operational instructions and code examples for managing Kubernetes PersistentVolumeClaims (PVCs), StorageClasses, and PersistentVolumes (PVs). All tools and tasks described align with its stated purpose.
[SAFE]: No patterns of prompt injection, data exfiltration, or persistence mechanisms were detected. The skill uses standard tool calls to interact with a Kubernetes cluster.
[INDIRECT_PROMPT_INJECTION]: The skill includes instructions to apply Kubernetes manifests, which creates a potential surface for indirect prompt injection if the manifests contain malicious configurations from untrusted sources. However, this is inherent to the skill's primary function of cluster resource management.
Ingestion points: YAML manifest strings passed to the kubectl_apply tool in SKILL.md examples.
Boundary markers: None explicitly defined in the skill text.
Capability inventory: Includes resource creation (kubectl_apply) and deletion (kubectl_delete) capabilities within the Kubernetes cluster.
Sanitization: None defined; the skill relies on the underlying kubectl implementation for manifest validation.

k8s-storage