omicverse-single-cell-lda-topic-clustering

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill provides instructions for performing LDA topic clustering on single-cell data using the omicverse library. All operations described, including model fitting and label prediction, are standard practices in bioinformatics.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the omicverse and mira libraries as its primary computational backends. These are established open-source tools within the genomics research community.
  • [COMMAND_EXECUTION]: The skill includes a validation script (scripts/lda_smoke.py) used during acceptance testing to verify the availability of required functions. This script generates synthetic data and performs basic checks locally.
  • [DATA_EXFILTRATION]: The skill includes an ondisk=True option that writes temporary training and test datasets to the local filesystem. This is a documented behavior for handling large datasets and does not involve network exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 05:43 PM
Security Audit — agent-trust-hub — omicverse-single-cell-lda-topic-clustering