experimental-design

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses well-known and reputable Python libraries (numpy, pandas, and pyDOE3) for statistical design and data manipulation.
  • [SAFE]: No evidence of prompt injection or attempts to bypass AI safety guidelines were detected in the instructions or reference materials.
  • [SAFE]: No data exfiltration patterns or unauthorized access to sensitive files were identified. The scripts strictly perform local statistical computations and write results to local CSV files.
  • [SAFE]: The package installation instructions use standard tools (uv/pip) to fetch legitimate dependencies from official registries.
  • [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation techniques are present in the provided scripts or documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 08:08 PM
Security Audit — agent-trust-hub — experimental-design