ab-testing-analyzer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): No malicious logic or obfuscation was found in the provided skill files. The code performs legitimate statistical calculations.
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill relies on standard data science packages (pandas, numpy, scipy, matplotlib, seaborn, statsmodels) installed from trusted registries like PyPI.
  • [DATA_EXPOSURE] (SAFE): File access is restricted to reading user-provided CSV data and exporting analysis reports, which aligns with the skill's primary purpose. No sensitive environment variables or credentials are accessed.
  • [COMMAND_EXECUTION] (SAFE): The scripts operate within a standard Python execution environment and do not invoke dangerous system commands or subprocesses.
  • [PROMPT_INJECTION] (SAFE): While the skill ingests external CSV data, its function is statistical processing rather than language generation, which mitigates risks associated with indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:10 PM