verl-rl-training
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and standard configuration files for an established open-source machine learning library. No malicious patterns, obfuscation, or unauthorized data access were found.
- [EXTERNAL_DOWNLOADS]: Mentions installation of standard industry-standard libraries including
verl,torch,ray, andvllm. These references point to official package registries and the official ByteDance/Volcengine GitHub repository. - [COMMAND_EXECUTION]: Provides standard command-line instructions for distributed training and environment setup, which are consistent with the skill's purpose for machine learning research and development.
Audit Metadata