backend/outputs/eval/summary.json

{
  "timestamp": "2026-03-05T10:14:52.153644",
  "eval_seed": 42,
  "system_accuracy": "50%",
  "system_accuracy_raw": 50.0,
  "avg_throughput_fps": 117.0,
  "risk_level_accuracy": "25%",
  "risk_level_accuracy_raw": 25.0,
  "risk_priority_alignment": "71%",
  "reflection_unit_tests": "4/5",
  "avg_fp_rate": 0.0,
  "avg_bias_correction_pct": 51.25,
  "total_ops_decisions": 4,
  "avg_skip_rate_pct": 98.3,
  "iterations_documented": 9,
  "limitations_documented": 5
}