Validate whether a judgment framework (an investment screen, a hiring rubric, a risk score, any rule-based decision system) actually works, by backtesting it on real historical cases under a strict anti-self-deception protocol. Trigger when the user says "backtest my framework", "test if this rubric
Validate whether a judgment framework (an investment screen, a hiring rubric, a risk score, any rule-based decision system) actually works, by backtesting it on real historical cases under a strict anti-self-deception protocol. Trigger when the user says "backtest my framework", "test if this rubric/model actually works", "validate my decision criteria against history", or "stress-test my framewor