diff --git a/docs/reviews/001-waves-1-2-general-review.md b/docs/reviews/001-waves-1-2-general-review.md
index be99a3a..554a2be 100644
--- a/docs/reviews/001-waves-1-2-general-review.md
+++ b/docs/reviews/001-waves-1-2-general-review.md
@@ -10,8 +10,8 @@
 > N-4 (URI flag dropped, a registered fork delta), N-6 (1 ms default
 > stands, cadence carried onto `SqliteOpts`). The task-carried items
 > (lint removal, wave-5 suite adds, release-readiness N-1 review —
-> wave 7 since ADR-024's renumbering) stand as
-> ordered in §7.
+> wave 7 since ADR-024's renumbering; wave 8 since the fuzzing
+> wave's second renumber, 2026-10-10) stand as ordered in §7.
 
 - **Reviewer**: opencode (glm-5.3-flash)
 - **Scope**: general review of everything landed by waves 1–2 (11 tasks, commits
@@ -289,5 +289,6 @@ All gates re-run green after the fixes: build, workspace tests
   `allow` lints in `substrate/mod.rs`, and adjudicate N-6's poll-cadence
   default.
 - **Wave 5 contract suite**: the §4 test adds.
-- **Wave 6 release review**: consciously accept N-1's growth postures or
-  add engine-side hygiene.
\ No newline at end of file
+- **Wave 6 release review** *(the release pass — wave 6 at §7's
+  writing, since renumbered twice, now wave 8)*: consciously accept
+  N-1's growth postures or add engine-side hygiene.
\ No newline at end of file
diff --git a/docs/reviews/002-wave-4-general-review.md b/docs/reviews/002-wave-4-general-review.md
index 9ff18ce..964d5ea 100644
--- a/docs/reviews/002-wave-4-general-review.md
+++ b/docs/reviews/002-wave-4-general-review.md
@@ -247,8 +247,9 @@ covers it, but say so.
   visibility-budgeting one) would prevent surprise.
 - `eprintln!` diagnostics at the swallowed-wake/lag sites match the
   SQLite twin's posture; the logging story is a wave-7 item (a wave-6
-  item when recorded — renumbered by ADR-024; already recorded by the
-  wave-4 gate).
+  item when recorded — renumbered by ADR-024; the release pass, since
+  renumbered again to wave 8 by the fuzzing wave, 2026-10-10; already
+  recorded by the wave-4 gate).
 - F-2..F-4 from `tasks/review-wave-4.md` stand as recorded there.
 
 ## Recommended sequencing
diff --git a/docs/reviews/003-wave-5-general-review.md b/docs/reviews/003-wave-5-general-review.md
index 45d9ed7..9f7b17a 100644
--- a/docs/reviews/003-wave-5-general-review.md
+++ b/docs/reviews/003-wave-5-general-review.md
@@ -233,19 +233,22 @@ explicitly here so the call is recorded rather than implicit.
   row's lapse assertions can only be delayed, never un-lapsed, by slow
   clocks. Supporting the gate's watch-not-block disposition; the
   watch flag stands (the release-readiness watch — wave 7 since
-  ADR-024's renumbering).
+  ADR-024's renumbering; wave 8 since the fuzzing wave's second
+  renumber, 2026-10-10).
 
 ## Recommendation
 
 No changes required to keep wave 5's verdict standing. For wave 6's
 window *(wave 6 at writing — release readiness; renumbered by
 ADR-024, with Finding 1 since routed to the new wave 6 — the mem
-engine — window, and Findings 2–3 to wave 7's pre-release pass, per
+engine — window, and Findings 2–3 to wave 7's pre-release pass,
+the release pass now wave 8 since the fuzzing wave's second
+renumber, 2026-10-10, per
 `docs/plans/implementation.md`'s wave-6 section)*: Finding 1 is a
 decomposable small task (transient-fault seam +
 retry/exhaustion pin); Findings 2–3 can ride the same task or be
 accepted with the code-read as their pin — either way the disposition
 should be recorded. Findings 4–5 need no work. The env-variable
 boundary section above folds into the release-readiness pass
-(wave 7) as a
+(wave 7 at writing — wave 8 since the fuzzing wave's renumber) as a
 verified-now / re-check-at-prepublish item.
\ No newline at end of file
diff --git a/tasks/pg-fix-scheduler-fire-wake.md b/tasks/pg-fix-scheduler-fire-wake.md
index 3d52e69..a771b01 100644
--- a/tasks/pg-fix-scheduler-fire-wake.md
+++ b/tasks/pg-fix-scheduler-fire-wake.md
@@ -179,7 +179,9 @@ Fix-shape decisions of record (what the description left open):
   next two full runs plus six focused runs. Recorded in
   `tasks/review-wave-5.md` §6 with the ledger's own conclusion: the
   "fails again" trigger is met, a dedicated investigative session is
-  owed by the wave-7 watch carriage before any postulate-and-fix. Not
+  owed by the release-readiness watch carriage (wave 7 at writing —
+  wave 8 since the fuzzing wave's renumber) before any
+  postulate-and-fix. Not
   load-bearing here (locks, not scheduler/wakes; scheduler code
   untouched by that row).
 
diff --git a/tasks/release-renumber-sweep.md b/tasks/release-renumber-sweep.md
index 020debb..984921b 100644
--- a/tasks/release-renumber-sweep.md
+++ b/tasks/release-renumber-sweep.md
@@ -1,7 +1,7 @@
 ---
 id: release-renumber-sweep
 name: Renumber sweep — historical "wave 7" references to the release pass
-status: pending
+status: completed
 depends_on: []
 scope: narrow
 risk: trivial
@@ -50,14 +50,14 @@ The distinguishing rule: a reference is stale only if it means the
 
 ## Acceptance Criteria
 
-- [ ] Every stale "wave 7"-means-release-pass reference in
+- [x] Every stale "wave 7"-means-release-pass reference in
       `docs/reviews/` and `tasks/` annotated or reworded per the rule;
       genuine wave-7 (fuzzing) references untouched
-- [ ] The sweep's grep command + hit disposition list recorded in
+- [x] The sweep's grep command + hit disposition list recorded in
       Notes (so the next session doesn't re-derive the inventory)
-- [ ] No historical record's meaning altered — annotations only where
+- [x] No historical record's meaning altered — annotations only where
       the phrasing is load-bearing
-- [ ] `taskgraph validate` green after any task-file edits
+- [x] `taskgraph validate` green after any task-file edits
 
 ## References
 
@@ -69,8 +69,106 @@ The distinguishing rule: a reference is stale only if it means the
 
 ## Notes
 
-> To be filled by implementation agent
+> Filled by implementation agent (2026-10-10).
+
+**Sweep command** (what found the inventory):
+
+```
+grep -rn "wave. 7\|wave-7" docs/reviews/ tasks/
+```
+
+filtered by hand against the distinguishing rule (a reference is
+stale only if it means the *release pass*; fuzzing-wave and
+`review-wave-7` references are correct).
+
+**Hit disposition list** — every non-excluded hit and how it was
+disposed. Annotation style throughout: the bracketed parenthetical
+wave note (the `review-wave-5.md` ADR-024 precedent's shape, applied
+to the second renumber — "wave 8 since the fuzzing wave's renumber
+(2026-10-10)"), appended to the historical phrasing, which is kept
+verbatim:
+
+- `docs/reviews/001-waves-1-2-general-review.md` header note
+  ("release-readiness N-1 review — wave 7 since ADR-024's
+  renumbering") — **annotated** (added the wave-8 note; historical,
+  load-bearing).
+- `docs/reviews/001-waves-1-2-general-review.md` §7's "Wave 6
+  release review" bullet — **annotated** (this one is
+  double-removed: wave 6 at §7's writing, wave 8 now; the historical
+  wave-6 label kept, parenthetical trail note added).
+- `docs/reviews/002-wave-4-general-review.md` security-posture bullet
+  ("the logging story is a wave-7 item (a wave-6 item when recorded —
+  renumbered by ADR-024; …)") — **annotated** (a third-generation
+  trail annotation: wave 6 → 7 by ADR-024 → 8 by the fuzzing wave;
+  not in the task's original site list — found by the grep-for-more).
+- `docs/reviews/003-wave-5-general-review.md` flake-watch echo ("the
+  release-readiness watch — wave 7 since ADR-024's renumbering") —
+  **annotated**.
+- `docs/reviews/003-wave-5-general-review.md` Recommendation's
+  renumber-trail parenthetical ("Findings 2–3 to wave 7's pre-release
+  pass") — **annotated** (the parenthetical was itself one renumber
+  old; extended to the second).
+- `docs/reviews/003-wave-5-general-review.md` Recommendation's closing
+  ("folds into the release-readiness pass (wave 7)") — **annotated**.
+- `tasks/review-wave-5.md` AC list ("since renumbered to wave 7 when
+  the mem engine inserted at wave 6, ADR-024") — **annotated**.
+- `tasks/review-wave-5.md` §6 watch flag ("since renumbered to wave 7,
+  ADR-024") — **annotated**.
+- `tasks/review-wave-5.md` ledger update ("owed by the wave-7 watch
+  carriage") — **annotated** (carriage named as the release-readiness
+  watch, wave-8 note attached; historical wave-7 phrasing kept).
+- `tasks/review-wave-6.md` References ("Findings 2–3's wave-7 park") —
+  **reworded** (live pointer pointing a future reader at the wrong
+  wave — now "pre-release-pass park" with the wave-8 note).
+- `tasks/review-wave-6.md` Finding (wake-ordering, "recorded for the
+  wave-7 pre-release pass") — **annotated**.
+- `tasks/review-wave-6.md` Summary ("dispositions for the wave-7
+  pre-release pass") — **annotated**.
+- `tasks/review-wave-4.md` gate record ("the logging story is a wave-7
+  item (release readiness)") — **annotated** (not in the original
+  site list — found by the grep-for-more).
+- `tasks/pg-fix-scheduler-fire-wake.md` flake-ledger note ("owed by
+  the wave-7 watch carriage") — **annotated** (not in the original
+  site list — found by the grep-for-more; mirrors the wave-5 ledger
+  wording it echoes).
+
+**Untouched, verified correct** (the distinguishing rule's exception
+arm): the `fuzz-*` tasks and `tasks/review-wave-7.md` (genuine
+fuzzing-wave/review references — wave 7 since 2026-10-10 is those
+waves' correct number); `tasks/release-publish-prep.md` (points at
+`tasks/review-wave-7.md`, a genuine wave-7 record pointer, not a
+release-pass number); `tasks/review-wave-8.md` ("stale wave-7
+references gone or spot-checked" — a wave-8 record auditing this
+sweep, correct as written).
+
+**Found outside scope, left untouched**: `docs/architecture/README.md`
+("wave 6 — release readiness renumbered to wave 7") and
+`docs/architecture/decisions/024-mem-engine.md` ("release readiness
+renumbered to wave 7"). Both are historical renumber-trail records
+accurate at their writing (the fuzz renumber was later the same day),
+and ADR-024 is an accepted immutable ADR — history is not rewritten
+in place. Recorded here so the residue doesn't re-derive; the task's
+mandate and the plan's §Wave 8 historical note cover `docs/reviews/`
+and `tasks/` only. If anyone wants these swept too, it's a deliberate
+record-hygiene call, not a loose end from this task.
+
+**Word-variant sweep**: case-insensitive `wave-7|wave 7|wave7` checked;
+no additional phrasings beyond the inventory above.
 
 ## Summary
 
-> To be filled on completion
\ No newline at end of file
+> Filled on completion (2026-10-10).
+
+All stale "wave 7"-means-release-pass references in `docs/reviews/`
+(001 header + §7 bullet, 002 security bullet, 003 flake-watch echo +
+Recommendation trail + closing) and `tasks/` (review-wave-5 AC + §6
+watch flag + ledger update, review-wave-6 References + Finding +
+Summary, review-wave-4 gate record, pg-fix-scheduler-fire-wake ledger
+note) disposed: 12 annotated in place, 1 References entry reworded
+(live pointer), zero meaning-altering edits — every historical
+"wave-7" phrasing kept and the wave-8 note attached per the
+renumber-annotation precedent. Genuine fuzzing-wave/review-7
+references untouched; two out-of-scope architecture-doc trail records
+left and documented in Notes. `taskgraph validate` green; no code
+touched, so build/test/clippy/fmt gates unaffected (docs + task-file
+edits only).
\ No newline at end of file
diff --git a/tasks/review-wave-4.md b/tasks/review-wave-4.md
index 102c1d1..daedec8 100644
--- a/tasks/review-wave-4.md
+++ b/tasks/review-wave-4.md
@@ -224,7 +224,8 @@ acceptance criteria — not test-name-trust):
   file holds; `#[non_exhaustive]` policy respected (none on `PgOpts`);
   core has no driver deps; `eprintln!` diagnostics at the five
   swallowed-wake/lag sites match the SQLite twin's posture (the
-  logging story is a wave-7 item (release readiness), not a wave-4 defect).
+  logging story is a wave-7 item (release readiness) — wave 8 since
+  the fuzzing wave's renumber — not a wave-4 defect).
 - **Gates**: `cargo build`, `cargo test` server-less (25 core + 188
   sqlite + 111 pg-skip + 10 + 10 suite + 3 + 9 harness), clippy
   `-D warnings`, fmt — all green; the pg lib suite 111/111 green
diff --git a/tasks/review-wave-5.md b/tasks/review-wave-5.md
index 039faf2..5914bc9 100644
--- a/tasks/review-wave-5.md
+++ b/tasks/review-wave-5.md
@@ -69,7 +69,8 @@ Primary lenses:
       implementation.md updated (wave table + review rounds)
 - [x] Findings recorded; the next decomposition may proceed
       (release readiness at gate time — since renumbered to wave 7
-      when the mem engine inserted at wave 6, ADR-024)
+      when the mem engine inserted at wave 6, ADR-024; wave 8 since
+      the fuzzing wave's renumber, 2026-10-10)
 
 ## References
 
@@ -217,7 +218,8 @@ the pattern (a prior unreproducible flake hinting at a real engine
 issue) deserves that attention. Not blocking this gate: 13+ subsequent
 clean pg runs across both rows since the failures, all conditions
 covered. **Flagged for watch in the next wave window** (release
-readiness at gate time — since renumbered to wave 7, ADR-024).
+readiness at gate time — since renumbered to wave 7, ADR-024; wave 8
+since the fuzzing wave's renumber, 2026-10-10).
 
 *Ledger update (2026-10-10, from the `pg-fix-scheduler-fire-wake`
 session):* `row_lock_ttl_expiry_and_reacquisition` failed once more —
@@ -229,7 +231,8 @@ runs. Unrelated mechanism to that task's change (scheduler fire wake;
 locks untouched). Per this ledger's own prescription the row now
 meets the "fails again" trigger — a dedicated investigative session
 (instrumentation + focused repro loop) is owed by the wave-7 watch
-carriage before any postulate-and-fix. Recorded here for the ledger's
+carriage (the release-readiness watch — wave 8 since the fuzzing
+wave's renumber) before any postulate-and-fix. Recorded here for the ledger's
 honesty; not blocking the fire-wake task.
 
 **7. Findings.** No code, suite-row, or stamp defects requiring change
diff --git a/tasks/review-wave-6.md b/tasks/review-wave-6.md
index 0743dbd..2af33b9 100644
--- a/tasks/review-wave-6.md
+++ b/tasks/review-wave-6.md
@@ -70,7 +70,8 @@ contract and the pinned ADR-024 text, and audit:
 - docs/architecture/engine-mem.md
 - tasks/review-wave-3.md / tasks/review-wave-4.md (the gate pattern)
 - docs/reviews/003-wave-5-general-review.md (Finding 1's disposition
-  trail; Findings 2–3's wave-7 park)
+  trail; Findings 2–3's pre-release-pass park — wave 8 since the
+  fuzzing wave's renumber, 2026-10-10)
 
 ## Notes
 
@@ -125,8 +126,9 @@ are recorded with dispositions, none mem-divergent from the engines.
   share the shape in principle (offset assigned pre-commit, wake
   post-commit — the "pg-shaped story" the mem docs cite), and nothing
   contract-side pins wake-fire order vs offset order. Disposition:
-  this is a contract-level tolerance nobody pinned, not a mem-local
-  divergence — recorded for the wave-7 pre-release pass to either
+  this is a   contract-level tolerance nobody pinned, not a mem-local
+  divergence — recorded for the pre-release pass (wave 7 at writing —
+  wave 8 since the fuzzing wave's renumber) to either
   document as tolerated or amend as ADR text (wake order per stream =
   offset order). Mem's declared posture is current-thread-clean
   (engine-mem.md) and the suite's mem column proves it under the
@@ -257,7 +259,8 @@ arithmetic equivalence actually asserted — and the wasm32 acceptance
 items verified. The unpinned contract-adjacent observations (the
 multi-thread wake-ordering tolerance, three ADR-023 §2 table
 classifications, the receiver try_recv closed shape) are recorded
-above with their dispositions for the wave-7 pre-release pass; none
+above with their dispositions for the pre-release pass (wave 7 at
+writing — wave 8 since the fuzzing wave's renumber); none
 blocks. The rider (`pg-fix-scheduler-tick-seam`) closed Finding 1.
 Task files audited: all eleven dependency tasks completed with
 Notes/Summary filled.
\ No newline at end of file