The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of cluster management and infrastructure-as-code tools including kubectl, helm, and terraform to provision and manage production environments.
[EXTERNAL_DOWNLOADS]: The skill references and downloads configuration files from well-known project repositories, including the Argo CD installation manifests from the argoproj GitHub organization and Helm charts from Prometheus, AWS, and Kubernetes official registries.
[REMOTE_CODE_EXECUTION]: The skill applies remote manifests and installs remote Helm charts. While this constitutes execution of remote code on the cluster, the sources are well-known official repositories, representing standard operational procedure.
[INDIRECT_PROMPT_INJECTION]: The skill processes external infrastructure configuration data (YAML, HCL), which constitutes a theoretical attack surface. 1. Ingestion points: YAML manifests and Terraform configurations in EXAMPLES.md and REFERENCE.md. 2. Boundary markers: Explicit boundary markers are absent. 3. Capability inventory: The skill utilizes kubectl apply and helm install which modify cluster state. 4. Sanitization: No specific input validation or sanitization logic is defined for the ingested configuration data.

kubernetes-specialist