The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the blender command-line interface to execute a specific Python script located at skills/blender-render-and-export/scripts/render_or_export.py. This is a standard automation pattern for Blender-based workflows.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-owned Blender files (.blend), which are complex binary formats that can contain embedded scripts or drivers. While required for the skill's functionality, this ingestion of untrusted data represents an attack surface.
Ingestion points: User-supplied scene file at /absolute/caller-owned/hero.blend.
Boundary markers: None identified in the prompt instructions to isolate user content from the rendering logic.
Capability inventory: Execution of the blender binary with background processing and script execution capabilities.
Sanitization: None; the skill relies on the caller-owned file being safe for rendering.

ovrtx-hero-render