The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes kubectl and jq commands to retrieve deployment information, application health status, and sync errors from the argocd namespace in a Kubernetes cluster.- [COMMAND_EXECUTION]: Accesses system logs from the argocd-application-controller and retrieves configuration data from ConfigMaps like argocd-rbac-cm and argocd-notifications-cm for auditing purposes.- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection (Category 8) by processing untrusted external data.
Ingestion points: External data enters the agent's context through the output of kubectl commands when reading resource names, status conditions, and sync result messages in SKILL.md.
Boundary markers: The scripts do not implement delimiters or explicit 'ignore' instructions to distinguish between the cluster data being analyzed and the agent's instructional logic.
Capability inventory: The skill possesses the capability to execute shell commands via kubectl and read cluster-wide configurations.
Sanitization: There is no evidence of text sanitization or validation performed on the data retrieved from the Kubernetes API before it is passed to the agent for analysis.

managing-k8s-argocd-deep