# FLR-019 deterministic approval-flow lab

Implement the bounded evaluator in `starter.py` for the eight supplied synthetic approval-flow cases. No network calls are needed. Keep required policy gate, technical readiness, reviewer eligibility/availability, queue age, review latency, serial steps, re-review count, and authorized alternatives as separate evidence dimensions. A long queue alone is never enough to classify an approval bottleneck. Preserve OPEN state when policy authority is unknown, and do not treat a faster path as closure unless the required control is preserved.

Run `python run_checks.py` to verify fixture shape before implementing the evaluator.