The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill invokes Kubernetes-specific tools (velero_backup_create_tool, kubectl_apply, etc.) for administrative tasks. These calls are consistent with the skill's description for managing backups and restores.
[DATA_EXPOSURE] (SAFE): The skill references the handling of Kubernetes secrets during restoration processes. This is an expected and necessary behavior for a backup/restore utility and does not constitute unauthorized data exposure.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill exhibits an indirect prompt injection surface as it processes data from the cluster environment and possesses write capabilities.
Ingestion points: Resource lists and backup metadata retrieved via velero_backups_list_tool and velero_restores_list_tool in SKILL.md.
Boundary markers: Absent; the skill does not explicitly use delimiters to separate cluster data from instructions.
Capability inventory: Cluster state modification via kubectl_apply and velero_restore_create_tool in SKILL.md.
Sanitization: Absent; the skill relies on the underlying CLI tools for validation of resource names and manifests.

k8s-backup