blender-content-safety-and-privacy

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructions for establishing a security boundary when handling untrusted 3D assets, archives, and dependencies.
  • [SAFE]: The guidelines explicitly advise against the automatic execution of embedded scripts and add-ons in Blender files, mitigating risks of unauthorized code execution from untrusted content.
  • [SAFE]: Instructions include protective measures for data privacy, such as excluding sensitive project details from screenshots and preventing credentials from being placed in metadata.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface where untrusted external content could be used to influence agent behavior and provides mitigation strategies.
  • Ingestion points: External .blend files, USD layers, textures, and archives as described in SKILL.md.
  • Boundary markers: The skill instructs the agent to treat third-party content as untrusted and use isolated profiles.
  • Capability inventory: Notes that Blender content can contain embedded scripts, drivers, and add-ons with executable capabilities.
  • Sanitization: Recommends verifying file digests, inspecting metadata, and rejecting archive entries that attempt path traversal.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 06:06 PM
Security Audit — agent-trust-hub — blender-content-safety-and-privacy