nemoclaw-user-overview

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill exclusively contains documentation in Markdown format regarding the NemoClaw stack and its security architecture. No executable code or scripts are present.
  • [EXTERNAL_DOWNLOADS]: The documentation describes the legitimate process of downloading version-controlled blueprints from NVIDIA's official repositories for environment orchestration.
  • [DATA_EXFILTRATION]: Details the implementation of security measures such as credential stripping and inference routing through host-managed providers to prevent the leakage of sensitive keys from the sandbox.
  • [PROMPT_INJECTION]: Describes the surface for indirect prompt injection inherent in AI agents but focuses on the built-in isolation and network policies designed to contain such risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 11:35 PM