skills/jimliu/science-skills/scgpt/Gen Agent Trust Hub

scgpt

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes established and well-known bioinformatics libraries including scGPT, scanpy, and anndata for biological data analysis.
  • [COMMAND_EXECUTION]: Provides implementation details for submitting and managing remote compute jobs through a host.compute interface. This is a standard and expected mechanism for performing GPU-intensive foundation model tasks that exceed local capabilities.
  • [SAFE]: The metadata correctly references the official academic source (Wang Lab at the University of Toronto) for model weights and documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 01:21 AM
Security Audit — agent-trust-hub — scgpt