nemo-guardrails
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- SAFE (SAFE): The skill is explicitly designed for safety and security. Analysis of the logic and example workflows confirms its primary function is to block malicious or unsafe inputs and outputs.
- EXTERNAL_DOWNLOADS (LOW): The skill utilizes the
nemoguardrailsPython package. While the author is not explicitly listed in the trusted organizations list, the package is a reputable industry-standard library developed by NVIDIA for the specific purpose of the skill. - PROMPT_INJECTION (SAFE): The skill provides defensive mechanisms against prompt injection, such as patterns to detect jailbreak attempts and instructions to refuse them.
- DATA_EXFILTRATION (SAFE): The skill includes patterns for PII (Personally Identifiable Information) masking and does not contain any code that attempts to access sensitive system files or send data to untrusted external domains.
Audit Metadata