llava
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and example code for interacting with the LLaVA vision-language model. All external references target reputable open-source repositories and research papers (GitHub: haotian-liu/LLaVA, ArXiv, Hugging Face: liuhaotian).
- [EXTERNAL_DOWNLOADS]: The skill includes standard instructions to clone the official LLaVA repository and install its dependencies via pip. These are well-known and trusted sources in the machine learning community.
- [COMMAND_EXECUTION]: The provided scripts for training (bash scripts/v1_5/pretrain.sh) and model serving (python -m llava.serve.cli) are standard operational commands for this specific open-source tool and do not involve suspicious parameters or hidden execution.
Audit Metadata