metaverse-expert
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides educational content and C# code templates for VR interaction and multiplayer networking. A manual review of the instructions and reference scripts found no evidence of malicious patterns, obfuscation, or unauthorized data access. The code follows standard industry practices for the Unity game engine.
- [INDIRECT_PROMPT_INJECTION]: A theoretical attack surface was identified in the
AvatarManagercode template (references/EXAMPLES.md), which describes a pattern for loading NFT metadata from external IPFS or blockchain sources. While this represents a surface for ingesting untrusted data, it is a static educational example rather than active automation logic. Evidence Chain: 1. Ingestion point:AvatarManager.LoadNFTMetadatafunction inreferences/EXAMPLES.md. 2. Boundary markers: Absent in the code template. 3. Capability inventory:Bash,Write,Read, andWebSearchtools. 4. Sanitization: Not implemented in the provided template.
Audit Metadata