NewSmartMoney-Cub
A financial decision harness compares deterministic rules with optional Jev typed judgments over fixed evidence cases.
Added to Jevfast
NewHow it uses Jev
Track-specific typed answers become a structured decision. Deterministic Python validates temporal boundaries and required output fields, then evaluator compares the proposed decision with the supplied outcome.
What Jev decides
Illustrative source-defined Jev questionExample answers · not a recorded Jev response · Source ↗
Question 1 · evidence sufficiency
YOUR APP
INSTRUCTION
Assess whether the evidence is sufficient to justify the trade action.
STATE
The repository-owned `case_tr_dev_01` fixture state contains case_id `case_tr_dev_01`, symbol `TR.DE01`, decision_time `2026-09-15T09:30:00Z`, available_at at the same time, one earlier data source marked stale, loss_amount 300.0, return_pct -14.0, and synthetic execution notes.
JEV · CHOICE
- sufficient
- partial
- insufficient
- contradictory
App workflow
- Load evidence case
Read a fixed financial case with source and timing metadata.
- Evaluate
Run deterministic rules and, when configured, ask the Jev backend typed judgments.
- Check outcome
Validate required fields and evaluate the proposed decision against the recorded outcome.
Why it is interesting
It pairs probabilistic semantic judgments with deterministic financial constraints and a frozen, auditable evaluation set.