threejs-loaders

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes configuration examples that download Draco decoders and KTX2 transcoders from well-known services, including Google's static storage (gstatic.com) and the jsDelivr CDN. These are standard, industry-recognized sources for these dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from 3D model files (such as GLB, OBJ, and FBX) which may contain attacker-controlled metadata or animations.
  • Ingestion points: Data ingestion occurs through GLTFLoader.load(), OBJLoader.load(), and FBXLoader.load() methods documented in SKILL.md.
  • Boundary markers: Absent; there are no specific markers or instructions to isolate or ignore potentially malicious content embedded within the 3D assets.
  • Capability inventory: The skill demonstrates capabilities for network operations (fetch, loader.load) and file system interaction for asset management.
  • Sanitization: Absent; the code relies on standard library parsing without additional validation of the ingested asset data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 01:00 PM