skills/nvidia/simready-foundation/simready-foundation-conform-fet-004-robot-newton/Gen Agent Trust Hub
simready-foundation-conform-fet-004-robot-newton
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from validation reports, profile TOML files, and user requests to identify failing requirements.
- Ingestion points: The workflow in
SKILL.mdinvolves reading external validation reports and profile configurations. - Boundary markers: The skill does not define specific delimiters for separating user-provided report content from instructions.
- Capability inventory: The skill utilizes the
workspace upgradetool and validation gates to modify and verify assets. - Sanitization: No specific filtering or escaping for external report data is described.
- [COMMAND_EXECUTION]: The instructions refer to using developer-specific tools, such as the
workspace upgradecommand and feature adapters, to author runtime-specific data and run validation gates. - [SAFE]: All referenced domains (
nvidia.com) and internal repository paths (nv_core/) are consistent with the skill's stated purpose and authored context. No obfuscation, data exfiltration, or privilege escalation patterns were detected.
Audit Metadata