NeoSyntropyFramework

Guide / Validation

Turn requirements into explicit gates

Use deterministic validation for hard rules and semantic validation for evidence-based judgments, then branch explicitly.

Use code for hard policy

Pythonpython
@functional_validation_node(id="CheckRefund")
def check_refund(ctx):
    return (
        float(ctx.state.get("amount", 0)) <= 100
        and bool(ctx.state.get("receipt_attached"))
    )

Use few-shot evidence for semantic quality

Provide accepted and rejected examples that demonstrate the boundary, require a typed valid/reason result, and route invalid or uncertain outcomes to correction or review.