expertapp-receptionist
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, such as direct prompt injection, persistence, or obfuscation, were detected. The skill's behavior is consistent with its stated purpose of managing bookings.
- [CREDENTIALS_UNSAFE]: The skill handles an API key by instructing the agent to request it from the user at runtime. It explicitly forbids logging or exposing the key, adhering to secure development practices.
- [PROMPT_INJECTION]: The skill ingests data from the expert.com.jm API (including service and booking names). While this is an indirect prompt injection surface, the risk is low given the restricted scope of the agent's capabilities and lack of dangerous execution patterns.
- [EXTERNAL_DOWNLOADS]: Network operations are restricted to the primary service domain (expert.com.jm). No downloads of external code or scripts from unknown sources were identified.
Audit Metadata