The Agent Skills Directory

[SAFE]: The skill provides standard instructional content for Blender animation using the official bpy API. All procedures described are consistent with established 3D animation workflows and contain no malicious code.- [REMOTE_CODE_EXECUTION]: Recipe 6 describes the setup of Blender Drivers, which involves scripted Python expressions. This is an intended feature of Blender; the skill properly identifies the software's built-in security toggle ('Allow Driver Python Expression') as a prerequisite for use.- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads object and scene metadata and has the capability to execute code. Evidence: 1. Ingestion points: mcp__blender__get_scene_info and mcp__blender__get_object_info (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Bash and mcp__blender__execute_blender_code (SKILL.md). 4. Sanitization: Absent. This exposure is typical for Blender automation tools and is managed by application-level safety settings.

blender-animation