world-streaming
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of high-level architectural instructions (SKILL.md) and configuration/evaluation data (JSON files). No executable code (Python, JavaScript, or shell scripts) is included.
- [SAFE]: References to external documentation are limited to official Epic Games documentation for Unreal Engine, which is a well-known service. The reference is used neutrally for illustrative purposes.
- [SAFE]: The 'adversarial' evaluation case in evals.json is designed to test the robustness of world-streaming logic (e.g., preventing data loss during memory pressure) rather than attempting to subvert agent security.
- [SAFE]: No obfuscation, prompt injection, or data exfiltration patterns were detected. The skill focuses on technical implementation details for game residency, fast travel, and delta persistence.
Audit Metadata