gcp-resource-optimizer
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and command templates for GCP resource management. No malicious instructions, obfuscation, or unauthorized data access patterns were identified.
- [COMMAND_EXECUTION]: The skill includes numerous gcloud command templates for resource inspection and billing management. These are standard for cloud optimization tasks and correctly use placeholders (e.g., ACCOUNT_ID, PROJECT_ID) to ensure user control.
- [PROMPT_INJECTION]: 1. Ingestion points: gcloud list commands and BigQuery billing exports (SKILL.md, cost-queries.md). 2. Boundary markers: Absent. 3. Capability inventory: gcloud CLI and BigQuery SQL. 4. Sanitization: Absent. The skill contains an attack surface for indirect prompt injection via cloud metadata, which is expected for cloud management tools.
Audit Metadata