living-ui-creator
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates extensive safety instructions that align with secure development lifecycle standards. It explicitly forbids hardcoding credentials and provides a controlled platform environment for application scaffolding.\n- [PROMPT_INJECTION]: Instructions include clear operational constraints and prohibitions against modifying system-level configurations, ensuring the agent remains within its intended functional scope.\n- [DATA_EXFILTRATION]: The skill utilizes a bridge architecture for third-party service integrations, effectively isolating sensitive authentication data from the application code and preventing accidental data exposure.\n- [INDIRECT_PROMPT_INJECTION]: The architecture for app-to-agent triggers mandates that all runtime parameters be treated strictly as data, with explicit instructions to ignore any potential command content within those parameters.
Audit Metadata