qmt-inner-backtest

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for quantitative trading backtesting and does not exhibit any malicious behaviors such as credential theft, persistent access attempts, or unauthorized network communication.
  • [SAFE]: The provided template script, scripts/daily-factors-backtest.py, utilizes standard financial analysis libraries and platform-specific QMT APIs for legitimate market data processing and strategy execution.
  • [SAFE]: The instruction set explicitly requires the agent to interpret user-provided documents (PDFs/images) and present a summary for human verification before generating executable code, which mitigates the risk of indirect prompt injection.
  • [SAFE]: No obfuscation, dynamic code execution from untrusted sources, or privilege escalation patterns were identified in either the instructions or the script template.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 09:52 AM
Security Audit — agent-trust-hub — qmt-inner-backtest