gke-service-networking
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill provides instructions and templates for executing
gcloudandkubectlcommands. These commands are used to configure critical infrastructure components such as Cloud Armor security policies, firewall rules, and Kubernetes Gateways/Ingress. Users should ensure these commands are executed with the appropriate permissions and that the resulting configurations align with their organization's security posture. - Indirect Prompt Injection Surface: Several assets (e.g.,
assets/backendconfig.yaml,assets/gateway.yaml) and command examples inSKILL.mduse placeholders like{security_policy_name}or{app_name}. This creates an ingestion point for user-supplied data. If this data is not properly validated before being interpolated into manifests and applied viakubectl, it could potentially be used to inject unintended configurations. - Ingestion points: Placeholder values in YAML manifests within the
assets/directory and variables ingcloudcommand snippets. - Boundary markers: The skill does not define specific delimiters or validation logic for the placeholder values, relying on the agent or user for correct substitution.
- Capability inventory: The skill utilizes
kubectl applyandgcloud computecommands to modify cluster and cloud resources. - Sanitization: No explicit sanitization or filtering of the user-provided placeholder values is implemented within the skill templates.
Audit Metadata