ovphysx-output-read

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill instructions and associated references align with official documentation for NVIDIA Omniverse physics APIs and follow standard development practices for resource management and native API interaction.
  • [EXTERNAL_DOWNLOADS]: The skill references technical documentation from the official NVIDIA Omniverse GitHub documentation site (nvidia-omniverse.github.io). This is a trusted vendor resource.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing simulation data which constitutes a potential ingestion point. 1. Ingestion points: Simulation output results (prim lists, attributes, and tensors) accessed via PhysX.read and ovphysx_read. 2. Boundary markers: Data is processed as structured tensors and attribute tokens. 3. Capability inventory: Data is used to inform simulation state and stage writes. 4. Sanitization: Not explicitly detailed as the processed data consists of numeric simulation attributes and interned tokens.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 02:49 PM
Security Audit — agent-trust-hub — ovphysx-output-read