NYC

histolab

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The code interacts with local file paths for reading slide images and saving processing outputs. No hardcoded credentials or data exfiltration attempts were detected.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill uses standard scientific Python packages including histolab, numpy, matplotlib, and opencv. There is no evidence of downloading and executing remote scripts or dynamic code evaluation from untrusted sources.
  • [Command Execution] (SAFE): The skill does not invoke any system commands, shell processes, or external binaries.
  • [Persistence Mechanisms] (SAFE): No patterns for maintaining persistent access, such as modifying shell profiles or creating scheduled tasks, were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:23 PM