feat(contract): D-TRI-1 value-tenant half — autopoiesis triangle (3 SoA lanes)

.claude/board/LATEST_STATE.md

@@ -1,5 +1,34 @@
 # LATEST_STATE — What Just Shipped (read this FIRST)
 
+## 2026-07-17 — branch `claude/review-claude-board-files-nhqgx1` — D-TRI-1 value-tenant half: the autopoiesis triangle lands as 3 SoA lanes
+
+**IN PR.** The value-tenant half of the D-TRI-1 batched mint
+(`triangle-tenants-gestalt-separation-v1.md` §1). Three NEW append-only
+`ValueTenant`s — `FrozenStyle = 10`, `LearnedStyle = 11`, `ExploreStyle = 12`
+— each `ColumnKind::U8 × 12` (12 palette256 atoms), contiguous at `row_offset`
+152 / 164 / 176 (value-slab `[120,156)`), appended **after Kanban** per the
+2026-07-17 operator ruling ("triangle right after the kanban board").
+**Additive, reserve-don't-reclaim, layout-preserving** (Full carve 120→156 B,
+`NODE_ROW_STRIDE` 512 untouched, no `ENVELOPE_LAYOUT_VERSION` bump). Added to
+`ValueSchema::Full` **only** — deliberately NOT `Cognitive`, so entity classes
+(OSINT/PROJECT/ERP/Commerce, which resolve to Cognitive) do **not** inherit a
+thinking-style triangle; the thinking-row schema decision is deferred to P4
+(ancestry pipeline). Slot `f` = `StyleFamily` ordinal 0..11 **or** compiled-
+template step 0..11 (one content-blind register, ClassView-selected reading,
+plan §4). **Atom 0 = null default** (zero-fallback: an un-populated lane reads
+all-null, never a wrong policy). Accessors `NodeRow::{style_lane, set_style_lane,
+triangle_for}` (`triangle_for(f) -> (frozen[f], learned[f], explore[f])`, the
+one-glance per-family read the dispatch/perturbation/learning ancestry pipeline
+resolves against; `set_style_lane` is the owner-`&mut` write that resolves the
+W4b orphan-write flag). **The I-LEGACY field-isolation matrix**
+(`thinking_style_triangle_tenant_carve_field_isolation_matrix`) certifies each
+lane's 12 bytes flip in isolation, key/edges/other-tenant untouched.
+**906 contract lib tests green; clippy `-D warnings` + fmt clean.**
+**Still queued (the classid half — cross-repo OGAR mint):** chess domain 0x06
+concepts, the Tasks-SoA task-row classid (cognitive-task concepts), and
+BoardAggregates @ row_offset 188 — one batched OGAR-originated mint (never solo).
+Refs: STATUS_BOARD D-TRI-1, plan §1/§5/§6, v3-envelope-auditor gate.
+
 ## 2026-07-16 — branch `claude/x265-x266-plans-review-h9osnl` (v5, post-#702) — probe-wave verdicts on main
 
 #702 MERGED (autonomous merge under the operator-granted authority):

.claude/board/STATUS_BOARD.md

@@ -4,7 +4,7 @@ Plan: `.claude/plans/triangle-tenants-gestalt-separation-v1.md`. Design shipped;
 
 | D-id | Title | Repo | Status | Evidence |
 |---|---|---|---|---|
-| D-TRI-1 | Triangle tenant spec (Frozen/Learned/Explore, 12 slots x palette256) through envelope-auditor T1-T6, behind the batched mint | lance-graph | Queued (LAYOUT-GATED) | plan §1, §6 |
+| D-TRI-1 | Triangle tenant spec (Frozen/Learned/Explore, 12 slots x palette256) through envelope-auditor T1-T6, behind the batched mint | lance-graph | **VALUE-TENANT HALF IN PR** — 3 lanes `ValueTenant::{FrozenStyle=10,LearnedStyle=11,ExploreStyle=12}` U8×12 at row_offset 152/164/176 (value-slab [120,156)), additive after Kanban per operator ruling; in `Full` only (NOT Cognitive — entity classes keep their carve), zero ENVELOPE_LAYOUT_VERSION bump; `NodeRow::{style_lane,set_style_lane,triangle_for}` + field-isolation matrix test; v3-envelope-auditor gated. **classid half (chess 0x06 + Tasks-SoA + BoardAggregates @188) still queued — the cross-repo OGAR mint.** | plan §1, §6 |
 | D-TRI-2 | 12-family vs 12-step reading agreement: jc battery (ICC, Pearson/Spearman, Cronbach alpha) over real shader cycles; collapse only on measured identity | lance-graph | Queued | plan §4, §6 |
 | D-TRI-3 | Nail->hammer dispatch probe: object resonance -> atom retrieval vs inverted baseline; no inverted read path exists structurally | lance-graph | Queued | plan §3, §6 |
 | D-TRI-4 | Chess<->thinking transfer measurement (validity + reliability) — the quarantine-lift gate | lance-graph + stockfish-rs | Queued | plan §5, §6 |

.claude/board/TECH_DEBT.md

@@ -1,5 +1,14 @@
 # Technical Debt Log — Open + Paid (double-entry, append-only)
 
+## TD-TRI-1-P4-OBLIGATIONS — two forward obligations the v3-envelope-auditor flagged on the D-TRI-1 value-tenant half (2026-07-17, PR #717)
+
+**Surfaced by the v3-envelope-auditor** on the LAYOUT-CLEAN verdict for the triangle value tenants (`FrozenStyle`/`LearnedStyle`/`ExploreStyle`, PR #717). Neither gates #717 (the layout is additive + byte-isolated); both are **P4 (ancestry-pipeline) obligations** that must land WITH the reader that decodes these lanes:
+
+1. **The 226-atom cognitive palette256 codebook MUST reserve index 0 as the null / no-op atom.** The triangle's zero-fallback (`atom 0 = null default`, `NodeRow::{style_lane,triangle_for}`) is byte-safe today (reserved tail zero-fills, no persisted corpus wrote `[152,188)`), but the *reading* semantics are unbacked: that codebook does not exist in code yet (grep finds only the plan doc + the new doc-comments). When it lands, a zeroed slot must decode to "no policy," NOT "verb 0" (a live, wrong policy). Recommend a codebook-level `assert!`/test reserving index 0. This is the le-contract §3b split (bytes-don't-move is proven; the reading is a jc-pillar obligation).
+2. **The P4 thinking-row `ValueSchema` MUST include `FrozenStyle`/`LearnedStyle`/`ExploreStyle`** — or thinking rows must resolve to `Full`. The triangle is in `Full` ONLY (not `Cognitive`, deliberately — entity classes OSINT/PROJECT/ERP keep their carve). Reads are schema-independent (accessors read raw slab bytes), so no read-side gap today; the risk is *persist-side*: a thinking row with POPULATED triangle data that resolves to `Cognitive` would have those bytes dropped on a `field_mask`-driven bake. Today `DEFAULT → Full` covers unmapped rows.
+
+Both recorded so P4 (D-TRI-2/3 ancestry-pipeline consolidation) inherits them. Refs: PR #717, plan §1/§4, `E-*` (P4 wiring session to append the reader-side finding).
+
 ## TD-PLANNER-STYLE-DEFAULT-DRIFT-1 — PAID (2026-07-17, same day as surfaced)
 
 **Paid** in `crates/lance-graph-planner/src/thinking/style.rs` `default_modulation`. The 5 formerly-`_ => default()` families (`Convergent`/`Systematic`/`Divergent`/`Diffuse`/`Peripheral`) now carry the canonical `UNIFIED_STYLES`≡`StyleParams` values on the three MEASURED dims: `Convergent` 0.75/4/0.10, `Systematic` 0.70/5/0.10, `Divergent` 0.40/10/0.70, `Diffuse` 0.45/8/0.40, `Peripheral` 0.20/20/0.60. The **4 planner-specific dims** (`depth_bias`/`breadth_bias`/`noise_tolerance`/`speed_bias`) have NO canonical source, so they stay at `FieldModulation::default()` via `..default()` struct-update — filled only what was measured, fabricated nothing. The `_` fallback is REMOVED → the match is exhaustive over all 12 `StyleFamily` variants (a future variant is now a compile error, not a silent default). Regression test `formerly_defaulted_families_match_canonical_on_measured_dims` pins the 5 families' 3 dims + asserts the 4 unmeasured dims stayed at default; `every_family_has_a_nondefault_or_explicit_modulation` pins exhaustiveness. 218 planner lib tests green (+2), clippy `-D warnings` exit 0. Behavior change: the 5 families now search with their real per-family resonance/fan_out/exploration instead of the flat default — a correctness improvement (the drift D-TSC-1b measured is retired; re-running the probe would now show Mode A ≈ Mode B). Numeric half of census question O5 answered. See `E-PLANNER-STYLE-DRIFT-PAID-1`.