{"name":"Phronesis Decision Quality Bench","schema_version":"1.0","generated_at":"2026-08-28T01:14:30Z","modules":[{"module_id":"P-RES","status":"insufficient_sample","n_resolved":0,"sample_basis":"0 forecast_resolution_event rows","score":null,"slices":[],"notes":"P-RES needs the forecast_resolution_event kind populated; until forecasts resolve, this module reports honest-empty","family":"bench_specific"},{"module_id":"P-CAL","status":"insufficient_sample","n_resolved":0,"sample_basis":"0 scored forecast(s) in the public ledger","score":null,"slices":[],"notes":"P-CAL emits a score once at least 30 forecasts are scored against ground truth; current scored count = 0","family":"assurance_grade"},{"module_id":"P-EVD","status":"insufficient_sample","n_resolved":0,"sample_basis":"0 Decision Assets in the public ledger","score":null,"slices":[],"notes":"no Decision Assets to grade yet","family":"assurance_grade"},{"module_id":"P-ACT","status":"insufficient_sample","n_resolved":0,"sample_basis":"","score":null,"slices":[],"notes":"P-ACT lands in Bench Phase 2 — action_boundary scoring requires DA spec Phase 4 (live action_boundary computation) + resolved outcomes","family":"assurance_grade"},{"module_id":"P-SHP","status":"insufficient_sample","n_resolved":0,"sample_basis":"","score":null,"slices":[],"notes":"P-SHP lands in Bench Phase 1 — sharpness measures the usefulness of calibrated forecasts","family":"bench_specific"},{"module_id":"P-BBL","status":"insufficient_sample","n_resolved":0,"sample_basis":"","score":null,"slices":[],"notes":"P-BBL lands in Bench Phase 1 — edge vs naive / market / expert / model baselines","family":"bench_specific"},{"module_id":"P-ROB","status":"insufficient_sample","n_resolved":0,"sample_basis":"","score":null,"slices":[],"notes":"P-ROB lands in Bench Phase 2 — robustness ties to the RAMPART surface + evidence_graph.contradiction_flags","family":"bench_specific"},{"module_id":"P-COST","status":"insufficient_sample","n_resolved":0,"sample_basis":"","score":null,"slices":[],"notes":"P-COST lands in Bench Phase 2 — needs the per-DA retrieval cost ledger (Decision Feed Phase 1 follow-on)","family":"bench_specific"}],"notes":["Phase 0 — name the stick. P-CAL / P-EVD / P-RES live; the rest publish honest insufficient_sample. Every score carries n_resolved + sample_basis so a reader sees its strength."]}