session 2026-07-28: governance block closed — CLAUDE.md drift 9→0 (REVIEWED-76/77/79/80), PENDING.md split 1848→430 + archive, wake-digest SessionStart hook, doctrine ids live, PENDING-79/80/81

REVIEWED.md

@@ -719,4 +719,32 @@ born-digital lane (whole-EPUB injection → per-spine) + promotion.scale_applica
   - **Q4** PROPOSAL/MINOR confirmed; the FIX reading declined on the vacuity argument.
   - **Q5** Edit 2 kept (§V reads as a prose-framed universal alone).
   - **Carry-forward (recorded here, deliberately NOT in spec text):** the primitives that would serve a future non-text kind already exist ratified — a character-bearing image is content, hash-fixed, mapped under a catalogued normalization, never silently dropped (REVIEWED-70 Q3); and the Loeb sidecar already holds a typed `{kind: 'glyph', file: …}` entry. Nothing is reserved and nothing is promised, but a successor arriving at the door this edit opens should find that the architecture's own primitives were already pointing that way.
-**If AUTHORIZED:** Landed as spec v2.8.0's predecessor v2.7.0 (2026-07-25) + the sidecar-schema generality note (FIX). Tag commits REVIEWED-75.
+**If AUTHORIZED:** Landed as spec v2.8.0's predecessor v2.7.0 (2026-07-25) + the sidecar-schema generality note (FIX). Tag commits REVIEWED-75.
+
+## REVIEWED-76 — PENDING-76 — Authorization class follows claim class
+**Date:** 2026-07-28
+**Decision:** REJECTED — withdrawn by the executor (proposer) following the jurist's remand.
+**Notes:**
+**Reason.** The proposal would have amended Constitutional Constraint #1 so that
+  machine-verifiable state claims fell to `[FIX]` while doctrine remained `[ESCALATE]`. The jurist remanded it by running the package's own Part IV.2 refinement back across its Part II census — a check the executor drafted and did not perform. **Required count returned 0 of 11:** nine of the false claims fail the doctrine test (every dead tool name was welded to the directive instructing its use), and two were steward-held. Of the defects that were `[FIX]`-eligible, all five were structural and none was a state claim. The amendment's target category was empty: it authorised no edit that existing law did not already permit.
+**What survived the proposal.** Q2, severed and ratified as a standing epistemic standard, with the addition that **a negative command result requires a positive control** — an absence proves nothing until the instrument is shown capable of detecting presence. Q4's preference for sunset over revocation. Both are now load-bearing in `governance-drift-check.py` and `wake-digest.py`.
+**How the underlying problem was actually solved.** The jurist's reframing — that the drift was a symptom and the disease was operational configuration filed in a constitutional instrument — was accepted by the steward and executed on 2026-07-28 under existing law, with no amendment: REVIEWED-79 (doctrine preserved, instrument retargeted, state moved to `MEMORY.md`) and REVIEWED-80 (doctrine ids). **Empirical close: the drift count went 9 → 0 without the rule change the proposal argued was necessary.** That is a retrospective refutation of its premise, and it is checkable by running `governance-drift-check.py`.
+**Do not revisit without new evidence** — specifically, a defect class that is machine-verifiable, *not* steward-held, and *not* welded to a directive. None of the eleven examined met that bar.
+
+## REVIEWED-77 — PENDING-77 — CLAUDE.md structural repair
+**Date:** 2026-07-28
+**Decision:** AUTHORIZED
+**Notes:**
+**If AUTHORIZED:** Applied 2026-07-28. Four of five defects fixed directly; the fifth (empty `### L1 Active Workstream` stub) was superseded by REVIEWED-79 leg C, which deleted the region.
+
+## REVIEWED-79 — PENDING-79 — CLAUDE.md doctrine preservation (legs A, B, C)
+**Date:** 2026-07-28
+**Decision:** AUTHORIZED
+**Notes:**
+**If AUTHORIZED:** All three legs applied 2026-07-28. Leg A: §MemPalace → §Memory Discipline, seven doctrine units preserved, tool roster and hook claim removed. Leg B: two rules hoisted to §Session Discipline. Leg C: §Active Projects reduced to a pointer at MEMORY.md. Verified: governance-drift-check 9 → 0.
+
+## REVIEWED-80 — PENDING-80 — Doctrine IDs, pilot on §Memory Discipline
+**Date:** 2026-07-28
+**Decision:** AUTHORIZED
+**Notes:**
+**If AUTHORIZED:** Applied 2026-07-28. Seven ids defined; drift-check §6 active and clean. Follow-on now unblocked: skills cite `D:` ids instead of paraphrasing doctrine.