AdaWorldAPI · AdaWorldAPI · Aug 26, 2026 · Aug 26, 2026 · Aug 26, 2026
diff --git a/.claude/board/SUPERSESSION-INDEX.md b/.claude/board/SUPERSESSION-INDEX.md
@@ -70,11 +70,15 @@ Route is **mechanical triage, not a verdict**: `ARCHIVE?` = the plan's own statu
 it shipped; `RESCOPE` = it targets a symbol marked RETIRE; `READ` = neither signal fires
 and a human read decides. Board coverage counts this plan's D-ids cited on the board.
 
+`ARCHIVE?` reads the status's LEADING token only. The first batch it produced was
+**3/3 false positives** -- an unanchored match on a word that did not predicate the
+plan (`ACTIVE ... Phase A/B COMPLETE`, `PROPOSAL. v1 SHIPPED in PR #420`, and a
+`Status legend:` defining the tick mark). All three were live, and archiving them
+would have retired work in flight. A route here is a prompt to read the plan, never
+a licence to act on it.
+
 | route | plan | ruled symbols named | self-declared status | board coverage |
 |---|---|---|---|---|
-| **ARCHIVE?** | `cognitive-substrate-convergence-v2` | `BindSpace`, `CollapseGateEmission`, `GateDecision`, `MergeMode` … | ACTIVE — sprint-11 Phase A/B COMPLETE (pendi | 5/15 |
-| **ARCHIVE?** | `odoo-savant-reasoners-v2` | `BindSpace`, `CollapseGateEmission` | PROPOSAL. v1 SHIPPED in PR #420 (`D-ODOO-SAV | 2/3 |
-| **ARCHIVE?** | `unified-soa-rubikon-integration-v1` | `BindSpace` | legend:** ✅ SHIPPED (verified file:line this | 8/8 |
 | **READ** | `archetype-scaffold-v1` | `GateDecision` | In progress (2026-04-24) | 0/0 |
 | **READ** | `capstone-cognitive-loop-wiring-nan-census-v1` | `GateDecision` | PROPOSED (2026-06-20). The measurement compa | 0/0 |
 | **READ** | `deepnsm-v3-convergence-v1` | `StepMask` | PROPOSED (doc-only). Extends `v3-convergence | 5/5 |
@@ -84,30 +88,32 @@ and a human read decides. Board coverage counts this plan's D-ids cited on the b
 | **READ** | `mask-algebra-revision-read-v1` | `StepMask` | DRAFT, awaiting operator ruling on §5 | 1/3 |
 | **READ** | `persistence-artifact-backed-commit-v1` | `persist_cycle` | RATIFIED (operator ruling 2026-08-09). Phase | 0/0 |
 | **READ** | `r2il-bpe-typed-genetic-recombination-v1` | `GateDecision` | PROPOSAL, §7's three falsifiers now RUN (see | 1/1 |
-| **READ** | `scientific-kg-substrate-v1` | `GateDecision` | (§0):** PROPOSED — **scoping doc**, no code. | 8/9 |
-| **READ** | `self-reasoning-substrate-v1` | `GateDecision` | - **Status:** PROPOSED — doc-only. No code,  | 5/5 |
+| **READ** | `scientific-kg-substrate-v1` | `GateDecision` | PROPOSED — **scoping doc**, no code. Records | 8/9 |
+| **READ** | `self-reasoning-substrate-v1` | `GateDecision` | PROPOSED — doc-only. No code, no contract ch | 5/5 |
 | **READ** | `v3-convergence-wiring-v1` | `GateDecision` | ACTIVE (2026-07-01). Operator: "I'm all in f | 0/0 |
-| **RESCOPE** | `soa-migration-diff-resolution-2026-06-13` | `BindSpace`, `CollapseGateEmission`, `GateDecision`, `MergeMode` … | The SoA / BindSpace / identity plan family,  | 3/5 |
-| **RESCOPE** | `cognitive-substrate-convergence-v1` | `BindSpace`, `CollapseGateEmission`, `GateDecision`, `MergeMode` … | PROPOSAL (sprint-10 architectural decisions  | 5/13 |
+| **RESCOPE** | `soa-migration-diff-resolution-2026-06-13` | `BindSpace`, `CollapseGateEmission`, `GateDecision`, `MergeMode` … | — | 3/5 |
+| **RESCOPE** | `cognitive-substrate-convergence-v1` | `BindSpace`, `CollapseGateEmission`, `GateDecision`, `MergeMode` … | PROPOSAL (sprint-10 architectural decisions  | 6/13 |
+| **RESCOPE** | `cognitive-substrate-convergence-v2` | `BindSpace`, `CollapseGateEmission`, `GateDecision`, `MergeMode` … | ACTIVE — sprint-11 Phase A/B COMPLETE (pendi | 7/15 |
 | **RESCOPE** | `bindspace-singleton-to-mailbox-soa-v1` | `BindSpace`, `CollapseGateEmission`, `ResonanceDto`, `ThinkingStyle` | CONJECTURE / design (migration spec). NOT ye | 18/19 |
 | **RESCOPE** | `callcenter-membrane-v1` | `BindSpace`, `GateDecision`, `MergeMode`, `ThinkingStyle` | Active | 0/0 |
 | **RESCOPE** | `causaledge64-mailbox-rename-soa-v1` | `BindSpace`, `GateDecision`, `MergeMode`, `ThinkingStyle` | Active (draft, 2026-05-14) | 1/10 |
-| **RESCOPE** | `integrated-cognitive-planner-v1` | `BindSpace`, `GateDecision`, `ResonanceDto`, `dispatch_busdto` | ` (:90), `QueryReference{server_id,ref_versi | 2/3 |
+| **RESCOPE** | `integrated-cognitive-planner-v1` | `BindSpace`, `GateDecision`, `ResonanceDto`, `dispatch_busdto` | — | 2/3 |
 | **RESCOPE** | `palantir-parity-cascade-v2` | `BindSpace`, `MergeMode`, `ResonanceDto`, `ThinkingStyle` | plan, not implementation. | 1/17 |
 | **RESCOPE** | `temporal-markov-and-style-classes-v1` | `BindSpace`, `MergeMode`, `StepMask`, `ThinkingStyle` | ACTIVE (operator-ratified 2026-07-10: "other | 16/19 |
 | **RESCOPE** | `unified-soa-convergence-v1` | `BindSpace`, `CollapseGateEmission`, `ResonanceDto`, `ThinkingStyle` | PROPOSAL / integration plan. Design-spec onl | 22/25 |
 | **RESCOPE** | `bindspace-mailbox-soa-dependency-map-v1` | `BindSpace`, `dispatch_busdto`, `persist_cycle` | MAP / preflight. No source wired yet. Read-b | 2/2 |
 | **RESCOPE** | `bindspace-mailbox-soa-w3-w4a-impl-v1` | `BindSpace`, `dispatch_busdto`, `persist_cycle` | v2 — 5-consolidation + 3-brutal-critic pass  | 1/1 |
-| **RESCOPE** | `cognitive-substrate-convergence-v3` | `BindSpace`, `CollapseGateEmission`, `MergeMode` | ACTIVE — sprint-12 Wave F + Wave G complete  | 5/16 |
+| **RESCOPE** | `cognitive-substrate-convergence-v3` | `BindSpace`, `CollapseGateEmission`, `MergeMode` | ACTIVE — sprint-12 Wave F + Wave G complete  | 7/16 |
 | **RESCOPE** | `mailbox-cycle-aware-write-contract-v1` | `BindSpace`, `dispatch_busdto`, `persist_cycle` | CONJECTURE / design. 5+3-gated before code. | 1/1 |
 | **RESCOPE** | `unified-integration-v1` | `BindSpace`, `MergeMode`, `ThinkingStyle` | Active — brainstorm phase complete; delivera | 0/0 |
 | **RESCOPE** | `2026-05-06-splat-osint-ingestion-v1` | `BindSpace`, `MergeMode` | Active — PR 1+2 of 6 in flight on `claude/sp | 1/7 |
 | **RESCOPE** | `Palette256-3DSB-PhiSpiral-attention-integration-plan` | `BindSpace`, `CollapseGateEmission` | — | 0/0 |
-| **RESCOPE** | `anatomy-realtime-v1` | `BindSpace`, `ThinkingStyle` | . | 0/1 |
-| **RESCOPE** | `elegant-herding-rocket-v1` | `BindSpace`, `ThinkingStyle` | ") vs | 0/0 |
+| **RESCOPE** | `anatomy-realtime-v1` | `BindSpace`, `ThinkingStyle` | — | 0/1 |
+| **RESCOPE** | `elegant-herding-rocket-v1` | `BindSpace`, `ThinkingStyle` | — | 0/0 |
 | **RESCOPE** | `kognitionswirtschaft-v1` | `GateDecision`, `ThinkingStyle` | PROPOSED (unbuilt, unprobed). Operator-initi | 1/7 |
 | **RESCOPE** | `lance-graph-ontology-v5` | `BindSpace`, `GateDecision` | Drafted (2026-05-07). Picks up where v4 (`cl | 3/16 |
 | **RESCOPE** | `north-star-integration-v1` | `CollapseGateEmission`, `ThinkingStyle` | RATIFIED (council resolved + gates ratified  | 1/1 |
+| **RESCOPE** | `odoo-savant-reasoners-v2` | `BindSpace`, `CollapseGateEmission` | PROPOSAL. v1 SHIPPED in PR #420 (`D-ODOO-SAV | 2/3 |
 | **RESCOPE** | `ogar-ar-shape-endgame-v1` | `GateDecision`, `ThinkingStyle` | when filed:** PLAN (pre-council). Becomes PL | 0/0 |
 | **RESCOPE** | `rung-ladder-grounding-v1` | `CollapseGateEmission`, `GateState` | PROPOSAL (the most-obvious first grounding p | 0/4 |
 | **RESCOPE** | `rung-persona-orchestration-v1` | `BindSpace`, `ThinkingStyle` | PROPOSAL (sibling to `rung-mul-grounding-v1` | 4/9 |
@@ -119,7 +125,7 @@ and a human read decides. Board coverage counts this plan's D-ids cited on the b
 | **RESCOPE** | `bindspace-columns-v1` | `BindSpace` | Active | 0/0 |
 | **RESCOPE** | `codec-sweep-via-lab-infra-v1` | `BindSpace` | — | 0/0 |
 | **RESCOPE** | `cycle-coherent-soa-snapshot-v1` | `CollapseGateEmission` | Queued | 0/6 |
-| **RESCOPE** | `dacr7-band-reading-contract-v1` | `ThinkingStyle` | + the temporal doctrine | 3/5 |
+| **RESCOPE** | `dacr7-band-reading-contract-v1` | `ThinkingStyle` | — | 3/5 |
 | **RESCOPE** | `foundry-consumer-parity-v1` | `BindSpace` | Active | 0/0 |
 | **RESCOPE** | `foundry-roadmap-unified-smb-medcare-v1` | `BindSpace` | Active | 0/0 |
 | **RESCOPE** | `lf-integration-mapping-v1` | `BindSpace` | Active (2026-04-25) | 0/0 |
@@ -133,9 +139,10 @@ and a human read decides. Board coverage counts this plan's D-ids cited on the b
 | **RESCOPE** | `sql-spo-ontology-bridge-v1` | `BindSpace` | Active | 0/0 |
 | **RESCOPE** | `super-domain-rbac-tenancy-v1` | `BindSpace` | Active | 23/41 |
 | **RESCOPE** | `thought-cycle-soa-awareness-integration-v1` | `BindSpace` | integration plan. No implementation claimed  | 0/0 |
-| **RESCOPE** | `unified-ogit-architecture-v1` | `ThinkingStyle` | , **code refs**, and where it | 0/0 |
+| **RESCOPE** | `unified-ogit-architecture-v1` | `ThinkingStyle` | — | 0/0 |
+| **RESCOPE** | `unified-soa-rubikon-integration-v1` | `BindSpace` | — | 8/8 |
 
-- **ARCHIVE?**: 3
-- **RESCOPE**: 47
+- **ARCHIVE?**: 0
+- **RESCOPE**: 50
 - **READ**: 12
 - ruled symbols tracked: 14
diff --git a/.claude/board/entries/2026-08-26-e-the-archive-route-was-3-of-3-wrong-1.md b/.claude/board/entries/2026-08-26-e-the-archive-route-was-3-of-3-wrong-1.md
@@ -0,0 +1,103 @@
+## 2026-08-26 — E-THE-ARCHIVE-ROUTE-WAS-3-OF-3-WRONG-1
+
+**Status:** FINDING (measured) · **Confidence:** HIGH
+
+First routing pass over `SUPERSESSION-INDEX.md`, taking the `ARCHIVE?` batch
+first because it was the smallest and looked the most decided. **All three were
+false positives, and every one of them is live work.** Archiving on the route
+would have retired a plan mid-flight.
+
+The classifier searched the whole status string for a shipped-word:
+
+```python
+SHIPPED = re.compile(r'\b(SHIPPED|COMPLETE[D]?|SUPERSEDED|DONE|CLOSED|LANDED)\b')
+route   = "ARCHIVE?" if SHIPPED.search(st) else ...
+```
+
+An unanchored search has no notion of **what** shipped. The three hits:
+
+| plan | matched on | what the word actually predicated |
+|---|---|---|
+| `cognitive-substrate-convergence-v2` | `COMPLETE` | a *phase* — the status opens `ACTIVE`, and the very next word is `(pending merges)` |
+| `odoo-savant-reasoners-v2` | `SHIPPED` | its **predecessor** — `PROPOSAL. v1 SHIPPED in PR #420`; this plan exists *because* v1's shape was judged wrong |
+| `unified-soa-rubikon-integration-v1` | `SHIPPED` | a **legend** — `Status legend:` defining what ✅ means, captured as if it were a status |
+
+Verified against `STATUS_BOARD.md` rather than the headers, anchoring the join on
+each row's own leading cell (a loose `grep D-CSV-11` returns the D-CSV-14 row,
+which *mentions* it — same bad-join class as the earlier filename-stem inflation):
+
+- `cognitive-substrate-convergence-v2` — 12 Shipped, 4 **In PR**, 1 Queued. The
+  four cite **#388/#390**; the repo is at #1044. Not archivable — its board rows
+  are *stale*, which is a different defect and a different fix.
+- `odoo-savant-reasoners-v2` — all five of its own D-ids **Queued**. Never
+  started.
+- `unified-soa-rubikon-integration-v1` — Queued, one OBE, one In PR.
+
+All three name a RETIRE symbol, so all three route **RESCOPE**. `ARCHIVE?` → 0,
+`RESCOPE` 47 → 50, `READ` unchanged at 12.
+
+**The fix is an anchor, not a longer word list.** A stricter alternation cannot
+help: any pattern matching `COMPLETE` fires on case 1, any matching `SHIPPED`
+fires on case 2. Only the status's *leading* token predicates the plan.
+`STATUS` additionally now refuses `Status legend:`.
+
+Both halves falsified rather than assumed — it fires on `SHIPPED in #420`,
+`DONE (2026-06-01)`, `✅ SHIPPED`, `Superseded by v3`; it is silent on all three
+false positives. **The rule is not dead:** `odoo-source-extraction-v1` carries a
+leading-shipped status tree-wide, so a category reading 0 inside the table is a
+measurement, not a broken guard. (My first can-it-fire run reported a spurious
+`False` because I fed `SHIPPED` a raw line; the generator only ever feeds it the
+captured group. The test was wrong, not the code.)
+
+**Two limits found while measuring, both still open:**
+
+1. **`crates` hit counts include tombstone comments.** `CollapseGateEmission`
+   shows 5 crate files and is genuinely **gone** — all six occurrences are
+   comments recording its removal. The column measures textual mention, so a
+   dead symbol reads as live.
+2. **The stale-status class is real and unmeasured.** Four D-CSV rows still say
+   `In PR (#388/#390)` hundreds of PRs later. Nothing reconciles a board row
+   against its PR's actual outcome. This is the "claim-vs-reality drift" check
+   already named as the highest-value unrun scriptable class.
+
+**The generalizable rule:** a mechanical route is a *prompt to read*, never a
+licence to act. The value of taking the smallest, most-decided-looking batch
+first was that it was small enough to check every member — and checking every
+member is what found that the category was empty.
+
+---
+
+**⊕ CORRECTION (same day, codex P2 on #1046 — verified, and it was right).**
+The first fix skipped `Status legend:` **by name**. That is whack-a-mole, not a
+fix: `STATUS` still searched the whole document and took the first hit anywhere,
+so for `unified-soa-rubikon-integration-v1` the next match became the section
+heading `## 6. Honest status (no overclaim)` at line 181 — and the table duly
+reported **`(no overclaim)`** as that plan's self-declared status. Codex also
+named the escalation I had not: any mid-document heading whose text after
+`status` leads with a shipped token (`## 9. Status: SHIPPED items`) recreates
+the false `ARCHIVE?` outright.
+
+The class, not the instance: **a status is metadata, not prose.** Extraction now
+runs over `plan_head()` — the preamble, extended through any *leading*
+status-titled section (`## §0 — Status` is metadata; line 181 is not). A strict
+preamble cut was measured first and was too tight: it lost
+`self-reasoning-substrate-v1`, whose real status sits in exactly such a `## §0`
+section. A 40-line window was also measured — its only marginal catch was itself
+prose, so the principled rule beat the magic constant on the evidence.
+
+Nine junk statuses disappear, not one: alongside `(no overclaim)` the table had
+been reporting `` ` (:90), `QueryReference{server_id,ref_versi ``, `") vs `,
+`, **code refs**, and where it `, and `. ` as statuses. Two more rows were
+near-misses recovered by allowing a parenthetical (`**Status (§0):** PROPOSED`).
+Twelve rows now read `—`; an honest absence beats a fabricated status.
+
+Routes are unchanged (`ARCHIVE?` 0 · `RESCOPE` 50 · `READ` 12) and generation is
+still idempotent. The falsifier suite gained codex's escalation case and stays
+two-sided: fires on `SHIPPED in #420` / `DONE (2026-06-01)` / `SUPERSEDED by v3`,
+silent on all five negatives.
+
+**What this adds to the entry above:** the original finding was that a route is a
+prompt to read, never a licence to act. This correction is the same error one
+level up — I fixed the *instance* I had measured rather than the *class* it
+belonged to, and shipped a table with a fabricated value in it. Naming the
+failing input is not the same as naming the failure.
diff --git a/.claude/board/entries/README.md b/.claude/board/entries/README.md
@@ -163,3 +163,4 @@ up in exactly one of them, never both.
 | 2026-08-06 | `E-AN-IDENTITY-SLOT-IS-NOT-A-RAIL-REF-WHICH-IS-WHY-A-WIDE-CARVING-CAN-BE-CORRECT-1` | the byte-axis rule is a rule about REFERENCES, and an exact identity is not one | [2026-08-06-e-an-identity-slot-is-not-a-rail-ref-which-is-why-a-wide-carving-can-be-correct-1.md](2026-08-06-e-an-identity-slot-is-not-a-rail-ref-which-is-why-a-wide-carving-can-be-correct-1.md) |
 | 2026-08-06 | `E-A-SORTED-CODEBOOK-ORDINAL-IS-A-PROPERTY-OF-THE-KEY-SET-NOT-THE-KEY-1` | a within-book bijectivity witness cannot see a between-book shift, and that is the gap review found | [2026-08-06-e-a-sorted-codebook-ordinal-is-a-property-of-the-key-set-not-the-key-1.md](2026-08-06-e-a-sorted-codebook-ordinal-is-a-property-of-the-key-set-not-the-key-1.md) |
 | 2026-08-06 | `E-A-REPEATABLE-TRANSFER-IS-NOT-IDEMPOTENCE-OVER-A-MULTI-FILE-DIRECTORY-1` | the PR #901 review round: four corrections, one of them load-bearing on a safety claim | [2026-08-06-e-a-repeatable-transfer-is-not-idempotence-over-a-multi-file-directory-1.md](2026-08-06-e-a-repeatable-transfer-is-not-idempotence-over-a-multi-file-directory-1.md) |
+| 2026-08-26 | `E-THE-ARCHIVE-ROUTE-WAS-3-OF-3-WRONG-1` | the index's ARCHIVE? batch was 3/3 false positives — an unanchored shipped-word match with no notion of what shipped; all three plans were live | [2026-08-26-e-the-archive-route-was-3-of-3-wrong-1.md](2026-08-26-e-the-archive-route-was-3-of-3-wrong-1.md) |
diff --git a/.claude/tools/supersession_index.py b/.claude/tools/supersession_index.py
@@ -48,17 +48,43 @@ def count(pat, *dirs):
 board = "\n".join(open(f,encoding='utf-8',errors='ignore').read()
                   for f in glob.glob(".claude/board/entries/*.md")+[".claude/board/EPIPHANIES.md"])
 AWARE = re.compile(r'v3/COMPONENT-MAP|\.claude/v3|RETIRE|REPURPOSE|D-PERT')
-SHIPPED = re.compile(r'\b(SHIPPED|COMPLETE[D]?|SUPERSEDED|DONE|CLOSED|LANDED)\b')
-STATUS = re.compile(r'[Ss]tatus:?\*{0,2}\s*([^\n|]{0,52})')
+# Anchored to the status's LEADING token on purpose. An unanchored search matched
+# a word that did not predicate the plan -- "ACTIVE ... Phase A/B COMPLETE",
+# "PROPOSAL. v1 SHIPPED in PR #420" -- routing live plans to ARCHIVE?. 3/3 of the
+# first ARCHIVE? batch were that error (2026-08-26 routing pass).
+SHIPPED = re.compile(r'^\W*(SHIPPED|COMPLETE[D]?|SUPERSEDED|DONE|CLOSED|LANDED)\b', re.I)
+# `(?!\s*legend)` so "Status legend:** OK SHIPPED (verified...)" is not read as a status.
+# Searched over the PREAMBLE only (text before the first `## ` heading), because a
+# plan's status is metadata, not prose. Skipping `legend` by name was whack-a-mole:
+# the next unanchored hit was the section heading `## 6. Honest status (no overclaim)`,
+# which the table then reported as a status -- and any mid-document heading whose text
+# after `status` leads with a shipped token would have recreated the false `ARCHIVE?`.
+# (codex P2 on #1046, verified.)
+STATUS = re.compile(r'[Ss]tatus(?!\s*legend)\s*(?:\([^)]{0,12}\))?:?\*{0,2}\s*([^\n|]{0,52})')
 DID = re.compile(r'\b(D-[A-Z]{2,}[A-Z0-9]*(?:-[A-Z0-9]+)*)\b')
 
+def plan_head(txt):
+    """The plan's metadata region: the preamble, extended through any LEADING
+    section whose heading is itself about status (`## §0 - Status`). Status is
+    metadata, not prose -- searching the whole document scraped section headings
+    and mid-document sentences as if they were statuses."""
+    parts = re.split(r'(?m)^(## .*)$', txt)
+    out = [parts[0]]
+    for i in range(1, len(parts) - 1, 2):
+        if re.search(r'status', parts[i], re.I):
+            out.append(parts[i + 1])
+        else:
+            break
+    return "\n".join(out)
+
 rows = []
 for p, txt in plans.items():
     named = sorted(s for s in syms if re.search(rf'\b{re.escape(s)}\b', txt))
     if not named or AWARE.search(txt): continue
     dids = set(DID.findall(txt))
     cited = sum(1 for d in dids if d in board)
-    st = (STATUS.search(txt).group(1).strip() if STATUS.search(txt) else '')
+    head = plan_head(txt)
+    st = (STATUS.search(head).group(1).strip() if STATUS.search(head) else '')
     retires = [s for s in named if syms[s][0].startswith('RETIRE')]
     route = ("ARCHIVE?" if SHIPPED.search(st) else
              "RESCOPE"  if retires else
@@ -134,6 +160,12 @@ def plans_for(sym):
 out("Route is **mechanical triage, not a verdict**: `ARCHIVE?` = the plan's own status says\n")
 out("it shipped; `RESCOPE` = it targets a symbol marked RETIRE; `READ` = neither signal fires\n")
 out("and a human read decides. Board coverage counts this plan's D-ids cited on the board.\n\n")
+out("`ARCHIVE?` reads the status's LEADING token only. The first batch it produced was\n")
+out("**3/3 false positives** -- an unanchored match on a word that did not predicate the\n")
+out("plan (`ACTIVE ... Phase A/B COMPLETE`, `PROPOSAL. v1 SHIPPED in PR #420`, and a\n")
+out("`Status legend:` defining the tick mark). All three were live, and archiving them\n")
+out("would have retired work in flight. A route here is a prompt to read the plan, never\n")
+out("a licence to act on it.\n\n")
 out("| route | plan | ruled symbols named | self-declared status | board coverage |\n")
 out("|---|---|---|---|---|\n")
 for route, n, stem, named, st, cited, tot in sorted(rows, key=lambda r: (r[0], -r[1])):