gke-ai-troubleshooting-tpu-metrics-monitoring
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill incorporates external variables such as
{cluster_name}and{node_name}directly into PromQL queries. If these inputs are sourced from untrusted external data without validation, they could potentially be used to manipulate query logic. - Ingestion points: Placeholders in
SKILL.mdincluding{project_id},{cluster_name},{location},{node_name}, and{node_pool_name}are used to construct queries. - Boundary markers: The skill does not currently define explicit delimiters or instructions for the agent to treat these specific inputs as untrusted data.
- Capability inventory: The skill provides a variety of PromQL queries in Steps 3 through 7 for cluster and node status monitoring.
- Sanitization: There are no explicit sanitization or verification steps for the placeholder values defined within the instruction files.
Audit Metadata