skills/aipoch/open-science/scvi-tools/Gen Agent Trust Hub

scvi-tools

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructions and a helper script (kernel.py) for genomic data analysis using established scientific libraries (scvi-tools, scanpy, anndata).
  • [COMMAND_EXECUTION]: The skill includes instructions for installing standard Python packages and provides a template for running compute-intensive tasks on a remote GPU provider (Modal). This is an expected workflow for the described genomic modeling tasks and relies on user-initiated execution.
  • [DYNAMIC_EXECUTION]: The skill uses exec() to load its local kernel.py file into the Python environment. The content of kernel.py is a standard utility function for data type conversion in pandas DataFrames and does not perform any sensitive operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 10:23 PM
Security Audit — agent-trust-hub — scvi-tools