blender-scene-assembly
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized commands were detected. The skill contains explicit prohibitions against downloading external assets or installing third-party extensions, effectively mitigating remote code execution risks.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface through the ingestion of external 3D files.
- Ingestion points: Importing FBX, OBJ, and glTF assets from external sources as described in SKILL.md.
- Boundary markers: Requirements for explicit input definitions and approved asset roots in examples/request.md.
- Capability inventory: Modification of scene hierarchy, object transforms, and visibility settings in SKILL.md.
- Sanitization: Instructions in SKILL.md to disclose untested format features and verify scene integrity through save/reopen checks.
Audit Metadata