gke-productionize
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted configuration data from the GKE environment which could potentially contain malicious instructions.
- Ingestion points: Discovery phase commands like 'kubectl get deployment -o yaml' and 'gcloud container clusters describe' in SKILL.md read data directly from the environment.
- Boundary markers: No explicit markers or instructions are provided to the agent to ignore potentially malicious embedded content in the retrieved YAML or JSON configurations.
- Capability inventory: The skill utilizes cloud and container CLI tools and delegates tasks to multiple other skills with various capabilities across the GKE ecosystem.
- Sanitization: There is no evidence of input validation or content filtering for the data retrieved from the cluster before it enters the agent context for assessment.
Audit Metadata