gcp-resource-optimizer

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious code or patterns were detected. The skill consists of documentation and command-line templates for legitimate cloud resource management tasks such as right-sizing instances, setting budget alerts, and cleaning up unused resources.
  • [PROMPT_INJECTION]: The skill displays a surface for indirect prompt injection because it instructs the agent to interpolate user-provided identifiers (such as PROJECT_ID, ACCOUNT_ID, and BILLING_ACCOUNT_ID) into shell commands and BigQuery SQL queries. This is an inherent property of tools designed to manage specific cloud environments and is not considered malicious.
  • Ingestion points: User-provided environment variables and identifiers used in SKILL.md and references/cost-queries.md.
  • Boundary markers: Not present; the skill assumes the provided identifiers are trusted.
  • Capability inventory: The skill uses the gcloud CLI and BigQuery for data retrieval and resource modification across various files including SKILL.md and references/cost-queries.md.
  • Sanitization: No explicit sanitization or validation of these identifiers is performed within the provided templates.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 09:13 PM
Security Audit — agent-trust-hub — gcp-resource-optimizer