scientific-meta-analysis

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides standard Python implementations for meta-analysis statistics (Hedges' g, Odds Ratio, etc.) and uses standard visualization libraries. The code is transparent and follows academic data processing conventions.
  • [EXTERNAL_DOWNLOADS]: Dependencies are restricted to reputable and well-known Python packages (scipy, statsmodels). References to external tools focus on established scientific databases like PubMed, EuropePMC, and Crossref, which are appropriate for the skill's research-oriented purpose.
  • [DATA_EXFILTRATION]: File system interactions are limited to saving analytical results (CSV files) and visualizations (PNG plots) in local directories. There are no network operations that exfiltrate data to untrusted domains.
  • [COMMAND_EXECUTION]: No usage of subprocess, os.system, or other shell execution patterns was found. The skill operates entirely within the Python runtime environment for mathematical analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 04:43 AM
Security Audit — agent-trust-hub — scientific-meta-analysis