foundation-model-analysis

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows standard biological data analysis workflows using the omicverse library. No malicious behavior, prompt injection, or data exfiltration patterns were identified.
  • [EXTERNAL_DOWNLOADS]: The skill mentions installing common bioinformatics packages (scgpt, geneformer) and downloading model checkpoints. These are expected operations for the stated purpose of analyzing foundation models.
  • [COMMAND_EXECUTION]: The skill utilizes the ov.fm API to execute model inference. The execution is scoped to the analysis of single-cell data provided by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 06:15 AM