asset-optimization
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines legitimate 3D modeling optimization workflows and asset validation procedures.
- [SAFE]: No evidence of prompt injection, data exfiltration, or unauthorized command execution was found.
- [SAFE]: External metadata references point to the author's official GitHub profile, which is consistent with the skill's identity and purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from 3D scenes (such as object names and material counts) via MCP queries. While this is an ingestion surface, the operations are limited to reporting and standard cleanup within the local environment.
- Ingestion points: MCP scene queries for polycount, material counts, and object naming in SKILL.md.
- Boundary markers: Not explicitly defined for the data ingestion.
- Capability inventory: MCP queries and cleanup operations limited to the host application context.
- Sanitization: No explicit sanitization of scene data is mentioned.
Audit Metadata