gke-tpu-dynamic-slices-monitoring
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill involves administrative operations such as patching resource metadata via kubectl patch and modifying cluster configurations via gcloud container clusters update. These actions are intended for troubleshooting and decommissioning purposes, are clearly marked as high-risk, and require the agent to obtain explicit user confirmation before proceeding.
- [PROMPT_INJECTION]: The skill uses user-supplied values (Project ID, Cluster Name, and Slice Name) directly in command execution strings. This represents an indirect prompt injection surface as there are no boundary markers, validation rules, or escaping mechanisms defined to sanitize the input before it is processed by the shell environment.
Audit Metadata