The Agent Skills Directory

[SAFE]: No security threats detected. The skill is dedicated to architectural modeling and uses the Blender Python API (bpy) and bmesh for scene manipulation. All functions are transparent and match their documented purposes.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes user-provided vertex coordinates and building dimensions (e.g., in vertices parameter of create_walls_from_outline). Evidence: Ingestion points: vertex lists and dimensions in SKILL.md and scripts/arch_utils.py. Boundary markers: Absent. Capability inventory: Python execution via bpy and allowed Bash tool. Sanitization: Absent. However, given the numeric nature of the expected input, the risk is minimal and considered a standard characteristic of modeling agents.

blender-architect