customer-care

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFE
Full Analysis
  • Internal Safety Guidelines: The skill includes explicit instructions for the agent to avoid asking for or repeating sensitive credentials, such as credit card numbers, passwords, or one-time codes. This aligns with security best practices for handling customer data.
  • Controlled Data Ingestion: The skill utilizes specific tools to retrieve order information and store policies. While fetching external data inherently involves a potential surface for indirect prompt injection, the skill mitigates this by instructing the agent to focus on factual reporting based on the retrieved records and by explicitly defining the scope of operations as read-only for this specific flow.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 12:15 AM