P3D Pack 1 Phase 4D — Decision Memo Report

Date: 2026-05-11 Author: Opus 4.7

---

Status flags

phase4d_status=PASS
mode=DECISION_MEMO_ONLY
phase4c_dryrun=PASS_ACCEPTED
species_decision_locked=false
composition_decision_locked=false
seed_allowed=false
backfill_allowed=false
migration_allowed=false
recommendation_provided=true
requires_GPT_User_decision=true
post_implementation_design_rule_acknowledged=true

Paths

Artifact	Path
Decision memo	...design/p3d-pack1-phase4d-species-mapping-decision-memo.md
Execution options	...design/p3d-pack1-phase4d-execution-options-design.md
This report	...reports/p3d-pack1-phase4d-decision-memo-report.md

Recommended option split (NOT locked)

Context	Lean	Reasoning
12 pilot rows	E3 (defer) or E1 (gom) — both acceptable	Pilot artifacts; will be reclassified regardless
Future law_unit	F4 (defer to Phase 5)	Nesting decision determines composition determines species. Solving species before nesting = guessing.
unit_version	G1 (no mapping)	Subordinate, 0 births, no effect
Combined	Option 2 (defer all to Phase 5)	One coherent pass: nesting → composition → species → mapping → backfill. No premature classification. No tech debt.

Key insight

The dependency chain is: nesting decision → composition → species → mapping → backfill → compliance. The bottleneck is the nesting decision (Phase 5 migration design), not the species choice. All evidence is gathered; what's needed is an ARCHITECTURAL decision about TAC→IU structural fidelity.

Confirmation

GPT + User decision required. No agent, no seed, no migration. Choose from 5 execution options, or propose alternative.