three-text-troika-3d

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected across all 11 threat categories.
  • [DYNAMIC_EXECUTION]: The skill utilizes createDerivedMaterial to inject GLSL code into Three.js shaders. This is a standard and safe practice for customizing 3D materials in graphics programming and does not involve executing arbitrary system commands or processing untrusted external script data.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits a potential attack surface for indirect prompt injection as it ingests external text strings and font files for rendering. However, these inputs are used solely for visual rendering within a WebGL context and the skill lacks exploitable capabilities such as network exfiltration or sensitive file writing, posing no significant risk.
  • [EXTERNAL_DOWNLOADS]: The skill references standard font file types (.ttf, .otf, .woff) through relative paths or placeholder URLs (e.g., /fonts/Inter-Bold.woff), which is standard behavior for web-based 3D applications.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 05:48 PM
Security Audit — agent-trust-hub — three-text-troika-3d