The Agent Skills Directory

[SAFE]: The skill provides a logical framework and process for strategy backtesting without executing any sensitive operations. It focuses on enforcing data integrity, realistic cost modeling, and statistical rigor in financial modeling.
[SAFE]: References to external data libraries like yfinance and ccxt are standard for the described domain of financial analysis and are used in a documented, non-malicious context. No remote code execution or unauthorized downloads are present.
[SAFE]: The skill explicitly includes safeguards against common modeling errors such as look-ahead bias and survivorship bias, and mandates human sign-off for any strategy that passes the backtest phase.

strategy-discovery-backtest