tooluniverse-scvi-tools
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strict preflight checks in
_scvi_common.pythat verify the presence of required scientific libraries (scvi-tools,anndata,scanpy) before execution, preventing unexpected failures. - [DYNAMIC_EXECUTION]: The use of
__import__inscripts/_scvi_common.pyis used to check module availability. The module names are hardcoded in the scripts, which eliminates the risk of arbitrary module loading from untrusted input. - [SAFE]: The skill adheres to an 'Honesty contract' which ensures that the AI agent does not fabricate results and honestly reports hardware usage (CPU vs GPU).
- [SAFE]: Data handling is performed locally on user-provided AnnData (
.h5ad) files using standard bioinformatics libraries. No external network calls or data exfiltration patterns were detected.
Audit Metadata