simready-foundation-conform-fet-023-robot-materials

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a validation command uv run --python 3.12 validate-simready-profile to verify repairs against specified profiles. This is a standard part of the SimReady conformance workflow.
  • [DATA_EXFILTRATION]: The skill reads local specification and requirement files (e.g., nv_core/sr_specs/docs/features/FET_023-robot_materials.md) to establish a 'Source of Truth'. It does not perform network operations or attempt to exfiltrate data.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external USD assets and validation reports as primary inputs.
  • Ingestion points: The skill ingests usd_asset files and validation_report data (SKILL.md).
  • Boundary markers: No explicit delimiters or boundary markers are defined for isolating untrusted data content.
  • Capability inventory: The skill has file system write capabilities (staged output creation) and command execution capabilities (uv run) (SKILL.md).
  • Sanitization: No explicit sanitization or validation of the internal content of USD files or reports is mentioned, though the repair logic is described as mechanical and deterministic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 09:31 AM
Security Audit — agent-trust-hub — simready-foundation-conform-fet-023-robot-materials