skills/scottstts/threejs-awesome-graphics-agent-skills/threejs-procedural-architecture/Gen Agent Trust Hub
threejs-procedural-architecture
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill instructions in
SKILL.mdare focused on technical guidance for building architectural generators. While it uses firm language regarding following examples, it does not attempt to override the AI's safety protocols or system instructions. - [DATA_EXPOSURE]: No hardcoded credentials, API keys, or sensitive file paths were detected. The generation logic relies on numeric seeds and mathematical parameters.
- [REMOTE_CODE_EXECUTION]: There are no patterns involving the download and execution of remote scripts. The JavaScript files contain local logic for geometry processing.
- [EXTERNAL_DOWNLOADS]: The skill uses
three(Three.js) as a dependency for 3D rendering and shader logic (viathree/webgpuandthree/tsl). This is a well-known and trusted library for web-based graphics. - [DYNAMIC_EXECUTION]: The use of
Fninshadow-clipmaps.jsis part of the Three.js Shading Language (TSL) for GPU shader generation. This is standard functionality for modern Three.js applications and does not represent an arbitrary code execution risk on the host system. - [OBFUSCATION]: The code in
building-system.jsandshadow-clipmaps.jsis technical and math-heavy, but it is not obfuscated. There are no hidden strings or encoded payloads. - [INDIRECT_PROMPT_INJECTION]: The skill processes building configuration settings. However, these parameters are only used for geometric calculations (clamping, lerping, mesh writing) and do not interact with dangerous system capabilities.
Audit Metadata