audience-segment-builder
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows security best practices by treating all external data sources as untrusted.\n- [PROMPT_INJECTION]: The skill addresses the risk of indirect prompt injection by instructing the agent to treat all user-provided data exports (CSV, GA4) as untrusted and to specifically ignore any instructions contained within those files.\n- [DATA_EXFILTRATION]: The skill prevents unauthorized data exposure by mandating that the agent use aggregated or hashed descriptions of customer data instead of raw personally identifiable information (PII) such as emails or phone numbers.
Audit Metadata