blender-modeling-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Blender command-line interface to execute local Python scripts (
blender --python create_asset.py). This is the primary intended function for automating 3D modeling and is implemented using standard execution patterns. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided descriptions and reference images. While this presents a theoretical attack surface common to agentic workflows, the skill mitigates risk through a 'Modeling Contract' phase and structured, staged reviews that emphasize verification against established technical criteria.
- [DYNAMIC_EXECUTION]: Procedural geometry is generated via a local helper script (
lathe_mesh.py) and LLM-generated asset scripts. The logic is transparent and focuses on mathematical mesh construction without utilizing unsafe deserialization or library injection techniques.
Audit Metadata