Futarchists Options

Futarchists Optionsis a modular system for discovering, testing, and executing options strategies. Its architecture separates deterministic Python functions from agent-driven research. Python controls every process that must be exact, reproducible, and auditable, including data preparation, feature calculation, genome generation, crossover, mutation, backtesting, fitness scoring, risk controls, portfolio construction, and trade execution. The same inputs should always produce the same outputs. Agents operate above this deterministic layer. They form hypotheses, study market regimes, interpret results, investigate failed strategies, and propose new features or experiments. Agents cannot directly promote strategies or execute trades. Their ideas must be converted into structured inputs and verified by the Python pipeline. Strategy discovery is modeled after genetic sequencing because each strategy is a combination of interacting traits. A strategy genome can encode its options structure, signed signals, entry threshold, target delta, expiration range, and spread width. Crossover combines traits from successful parents, while mutation introduces controlled changes that may produce stronger offspring. Each resulting strategy is expressed as an executable phenotype and tested for performance, robustness, drawdown, uncertainty, correlation, and portfolio contribution. Every genome retains a stable identity, lineage, validation record, and promotion history. This structure allows agents to expand the research frontier while Python maintains scientific rigor, traceability, risk discipline, and execution reliability.