worldview-game-chase-route-architecture
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust design process for game development without introducing malicious behaviors. Safety guidelines for microphone input and external API usage are explicitly stated and prioritized.
- [DATA_EXPOSURE_AND_EXFILTRATION]: While the skill mentions the use of microphone APIs and external 3D generation tools, it mandates user authorization, calibration, and deterministic fallbacks. It explicitly prohibits the storage of raw audio and ensures that external tool usage is subordinate to runtime evidence.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted project assets such as sketches and videos. Potential injection risks are mitigated by a structured 'Layer Lock' system that requires each stage of design to be verified in a grey-box runtime before proceeding, preventing instructions in ingested data from overriding intended agent behavior.
Audit Metadata