gke-productionize
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill uses standard Google Cloud CLI (
gcloud) and Kubernetes CLI (kubectl) commands to discover the state of the environment. These are routine operations for infrastructure management and are scoped to cluster and workload metadata. - Indirect Prompt Injection Surface: As an orchestrator, the skill processes configuration data from GKE clusters and workloads. While this presents a surface for indirect instructions, the skill defines a deterministic scoring rubric and delegates logic to specialized domain skills, which limits the potential for unexpected behavior from untrusted data.
- Vendor-Owned Resources: The skill instructions involve standard Google Cloud resources and tools (GKE, gcloud, kubectl) which are consistent with the skill's purpose and its author identity.
Audit Metadata