Lucent-Financial-Group · AceHack · May 16, 2026 · May 16, 2026
diff --git a/docs/ALIGNMENT.md b/docs/ALIGNMENT.md
@@ -1096,6 +1096,17 @@ taxonomy below is what those tools implement.
   re-alignment (e.g. primary-research-focus upgrade
   landing today) or a drift we caught late.
 
+**Per-round meta-cognition measurables** (per [B-0037.1](../docs/research/meta-cognition-survey-2026-04-21.md)):
+
+- **`self-corrections-per-round`.** Count of dated revision blocks on memories, BACKLOG rows, ADRs, commits within a round. Target: rising as factory matures. Grep `docs/` + `memory/` for dated revision blocks per round boundary.
+- **`overclaim-self-tags-per-round`.** Count of `overclaim*` tags written by the factory before external correction. Target: rising is healthy; zero is suspicious. `grep -r "overclaim" --include="*.md"` gated by round date range.
+- **`revision-blocks-per-round`.** Count of all dated revision blocks across memory + doc layers within a round. Target: rising with justifications logged. Same grep as above, broader pattern.
+- **`decohere-star-self-detected-events-count`.** Count of `decohere*` events the factory detects in its own work before external correction. Target: rising. `grep -r "decohere\*" --include="*.md"` per round.
+- **`meta-check-execution-rate`.** Ratio of round-closes that actually ran the meta-check step (see [B-0037.2](../docs/backlog/P2/B-0037.2-meta-cognition-round-close-checklist.md)). Target: 100% once the ROUND-HISTORY.md round-close meta-check template (landed by B-0037.2) is wired into the close ritual. Currently not yet automated; agent self-report in commit bodies.
+- **`meta-drift-detection-lag-rounds`.** How many rounds pass before a decayed audit-discipline is caught. Target: low and falling. Not yet automated; requires retrospective survey.
+
+The `meta-check-execution-rate` and `meta-drift-detection-lag-rounds` are honestly labelled *not yet automated* — they depend on subsequent infrastructure to be countable.
+
 ### Multi-round / research-grade metrics (work in progress)
 
 - **Calibration-honesty trajectory (SD-1).**

diff --git a/docs/BACKLOG.md b/docs/BACKLOG.md
@@ -379,7 +379,7 @@ are closed (status: closed in frontmatter)._
 - [ ] **[B-0037](backlog/P2/B-0037-meta-cognition-first-class-factory-discipline.md)** Meta-cognition as first-class factory discipline — survey, audit cadence, measurables
 - [x] **[B-0037.1](backlog/P2/B-0037.1-meta-cognition-taxonomy-survey-doc.md)** B-0037.1 — Meta-cognition taxonomy survey doc
 - [x] **[B-0037.2](backlog/P2/B-0037.2-meta-cognition-round-close-checklist.md)** B-0037.2 — Per-round meta-check checklist in ROUND-HISTORY.md template
-- [ ] **[B-0037.3](backlog/P2/B-0037.3-meta-cognition-measurables-wiring.md)** B-0037.3 — Meta-cognition measurables wired into alignment-trajectory dashboard
+- [x] **[B-0037.3](backlog/P2/B-0037.3-meta-cognition-measurables-wiring.md)** B-0037.3 — Meta-cognition measurables wired into alignment-trajectory dashboard
 - [ ] **[B-0037.4](backlog/P2/B-0037.4-meta-cognition-distributed-vs-concentrated-adr.md)** B-0037.4 — ADR for distributed-vs-concentrated meta-cognition framework
 - [x] **[B-0040](backlog/P2/B-0040-actor-model-factory-register-lens.md)** Actor model as factory-operational-register lens — Hewitt 1973 / Meijer / Akka / Orleans / Service Fabric
 - [ ] **[B-0042](backlog/P2/B-0042-bungie-corpus-priority-seed.md)** Bungie corpus priority seed — Halo / Destiny / Marathon / Myth / Oni / Pathways Into Darkness / "Grimwar

diff --git a/docs/backlog/P2/B-0037.3-meta-cognition-measurables-wiring.md b/docs/backlog/P2/B-0037.3-meta-cognition-measurables-wiring.md
@@ -1,7 +1,7 @@
 ---
 id: B-0037.3
 priority: P2
-status: open
+status: closed
 title: "B-0037.3 — Meta-cognition measurables wired into alignment-trajectory dashboard"
 created: 2026-05-10
 last_updated: 2026-05-10
@@ -54,3 +54,33 @@ meta-cognition discipline part of Zeta's primary-research-focus trajectory.
 
 - Distributed-vs-concentrated ADR (B-0037.4).
 - Automating the measurements (future child of B-0037.3 if needed).
+
+## Resolution
+
+Closed 2026-05-16 via picking up the #2-Ready overlay (the
+`classification: blocked` field was stale because B-0037.1 merged
+earlier this session via PR #3859; B-0037.2 closed in PR #3888
+landed the related ROUND-HISTORY.md template).
+
+**Deliverable shipped this PR**:
+
+`docs/ALIGNMENT.md` §"Measurability — what we count" → §"Per-round
+metrics" → appended a "Per-round meta-cognition measurables" subsection
+with all 6 measurables from B-0037.1 taxonomy:
+
+- `self-corrections-per-round` (concrete grep: dated revision blocks)
+- `overclaim-self-tags-per-round` (concrete grep: `overclaim` matches)
+- `revision-blocks-per-round` (concrete grep: same broader pattern)
+- `decohere-star-self-detected-events-count` (concrete grep: `decohere\*` matches)
+- `meta-check-execution-rate` (honestly labelled not-yet-automated)
+- `meta-drift-detection-lag-rounds` (honestly labelled not-yet-automated)
+
+**Acceptance check**:
+
+- ✅ ALIGNMENT.md §"Per-round metrics" lists all 6 measurables with
+  definitions + measurement procedures
+- ✅ 4 of 6 measurables have concrete grep commands (above 3-required threshold)
+- ✅ No existing measurables modified — pure additive subsection
+- ✅ `dotnet build -c Release` unaffected — pure doc addition
+
+**Composes with**: B-0037.2 closed PR #3888 (the round-close meta-check template that this row's `meta-check-execution-rate` measurable cites). B-0037.4 (distributed-vs-concentrated ADR) remains separate scope.