tradfi-portfolio-manager

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run local Python scripts (e.g., python3 backtests/v3_allocate_today.py) to retrieve portfolio state and regime scores. It specifies the use of a local virtual environment at /Users/engineer/.venv/bin/python3.
  • [DATA_EXPOSURE]: The agent is directed to read internal repository files such as GOAL.md, strategy/v3-bubble-aware-all-weather.md, and strategy/v3-etf-rationale.md to ground its decision-making. These are local files within the skill's own directory structure.
  • [SAFE]: The skill includes strong safety guardrails, explicitly forbidding the agent from placing real trades. It requires human-in-the-loop confirmation and prohibits the use of leverage or options without explicit sign-off.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 08:08 AM
Security Audit — agent-trust-hub — tradfi-portfolio-manager