15 — Supertrack O: final verdict

DESIGN_NEEDS_TARGETED_PROPOSALS_BEFORE_FINAL_APPROVAL

This is not a hardcode, PG-native, scale/feasibility, or read-path failure. The corrected design is internally coherent and has advanced materially. All six blocking RP are structurally resolved, both advisory RP are applied, the "exactly 27 authority surfaces" invariant holds, and Codex's deviations from T1's literal asks are all in the safe direction (fewer surfaces).

The verdict is gated by the single acceptance criterion the T1 handoff named explicitly — "H04/H05/H02 byte-implementable; every sub-payload key resolves to a named table.column" — which is not yet fully met. Three surgical precision gaps remain in CP-06 hash doc 07 (P-01 contradiction, P-02 scope_hash composition, P-03 runtime-aggregate total orders). Leaving them unpinned would re-admit the determinism/divergence loop RP-01 was created to close.

Per-supertrack results

• A blocking-RP resolution: 5 RESOLVED + RP-01 PARTIAL (one CONTRADICTION)
• B RP-01 runtime-evidence: RP01_NEEDS_PROPOSAL
• C RP-02 retention: RP02_VERIFIED
• D RP-03 constraint catalog: RP03_VERIFIED (+ advisory P-04)
• E RP-04 catalog-family: RP04_VERIFIED
• F RP-05 item_payload: RP05_VERIFIED
• G RP-06 same-human: RP06_VERIFIED (slot hash-binding strengthened by P-02)
• H RP-07 retirement-evidence FK: RP07_VERIFIED
• I RP-08 Directus read-contract: RP08_VERIFIED
• J cross-impact / non-regression: CROSS_IMPACT_CLEAN
• K zero-hardcode / disguised-hardcode: ZERO_HARDCODE_VERIFIED
• L PG-first/native/driven: PG_NATIVE_DRIVEN_VERIFIED
• M feasibility / scale: FEASIBILITY_SCALE_VERIFIED
• N proposals: 4 (P-01/P-02/P-03 blocking; P-04 advisory)

What this means

• Codex may not final-approve yet. Apply P-01..P-03 to CP-06 doc 07 (P-04 optional in RP-03 doc 06), recompute H01/H02/H04/H05 over the now-defined columns, re-run the CP-06 determinism fixtures for the four aggregates, then republish for a short T1 re-review.
• No implementation. Stage 2.6B, permit, REAL_RUN, and QT001 apply remain BLOCKED.
• Production remained READ-ONLY throughout this review; no Codex doc was edited.

Path to ready

P-01..P-03 are confined to one doc, add no surface/gate/hash/table, and ship with T1-supplied unambiguous values (artifact_sha256 source; the four slot keys; pinned PK-terminated orders). The next cycle should be fast: Codex edits → republish → short T1 re-review → DESIGN_READY_FOR_CODEX_FINAL_APPROVAL → Codex final approval.