rwkv-architecture

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and technical documentation are focused on educational content and project setup with no malicious intent detected.
  • [EXTERNAL_DOWNLOADS]: The skill lists standard machine learning libraries (torch, transformers, rwkv, pytorch-lightning, deepspeed, wandb, ninja) for installation via official package managers and registries. These are appropriate for the skill's stated purpose.
  • [COMMAND_EXECUTION]: Code snippets provide examples for running model inference, training, and state management using the RWKV library in a local environment. These patterns are standard for ML development.
  • [SAFE]: All external URLs point to legitimate and well-known sources including arXiv for research papers, official GitHub repositories for the RWKV project, and the official project wiki.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 05:57 AM