nemo-rl-brev-etiquette
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides bash snippets for initializing directory structures on /ephemeral volumes and checking disk capacity using standard commands such as df -h and mkdir -p.
- [SAFE]: Secret management guidelines for loading authentication tokens from /home/ubuntu/RL/.env follow industry-standard practices for local secret storage. The skill includes explicit instructions for the agent to never print, cat, log, or commit these values, preventing accidental credential exposure.
- [SAFE]: The skill processes experiment artifacts and log files stored on ephemeral volumes. While these can originate from external sources, the skill only provides storage management guidance and does not execute untrusted data dynamically.
- [SAFE]: Self-referential safety claims and evaluation benchmarks are present in the documentation but were evaluated as informational metadata rather than authoritative safety signals.
Audit Metadata