nat-user-rules
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
natCLI (a component of the NVIDIA NeMo Agent Toolkit) viauv runfor discovering registered components like functions, evaluators, and providers. These are standard diagnostic commands relevant to the toolkit's primary purpose. - [INDIRECT_PROMPT_INJECTION]: The 'Skill Evolution' section describes a process where the agent may update its own instruction sets based on user feedback or command failures. While this involves processing external input to modify behavior, it is a standard design pattern for self-improving agents within a controlled development environment and does not contain malicious triggers.
Audit Metadata