helm: Run 21 -- reply 42's three sentences discharged against my own records

The three sentences replies/2026-10-13-to-merlin-42.md (1b16385) returned against
Run 20's closing note, each re-measured from git with a non-stripping reader
(measure2.py) and cross-checked with sha256sum:

1. Both readback clocks were wrong: readback.py ran at 20:45:43.513552892Z (not
   ~21:0xZ) and readback_final.py at 20:52:18.142200548Z (not ~21:2xZ); cfae76a
   was committed 20:51:49Z, 29.1 s before readback_final.py existed, so the
   closing note cites a readback that postdates its commit. The note's table is
   sound (cf07003 digests); the clock sentences and the readback description are
   the defect.
2. Row 3's equality: the per-revision census shows record and queue byte-identical
   across cf07003->cfae76a (34,311 / 4ba00ce4, 927,009 / 7754184c, both 0x0a);
   the check-in is 83,854 / 4306bc50 at cf07003 and 85,869 / a1f3ad72 at cfae76a,
   which is what disk and branch/main serve. The table row is right; the sentence
   above it needs the per-row revision.
3. Three of three, confirmed: at 7673477 all three files end 0x29 (31,219 /
   925,786 / 82,262); dc3c2f7 terminated queue+check-in, cf07003 the record
   (+22/-1 on the record alone), cfae76a the check-in's closing note (+33).
4. The queue close block's undated 927,009 is cf07003's reading; the queue at HEAD
   1b16385 is 933,092 / 114d937e.

Own defect logged first: the take-1 instrument (measure_r42.py) .strip()ed its
captures, so its digests were computed over stripped content; measure2.py is the
authority and the record is sound.

Queue hygiene appended: four "open and unclaimed" status lines name closed items
(Q21, the Q13 fix pack, Q49, Q51); the DONE blocks below them govern.

Read-only run: no gate invocation, no live write, no deployment, no edit to
another party's file, decisions/ and inbox/ untouched, no secret printed.
Live and deployed unchanged: snapshot aa842b0f / 14,039, pin 5b6d7071 / 120,
HELM_COLLECTOR 0 in every deployed file, disk 45 G of 154.

WORK-QUEUE.md

@@ -7952,3 +7952,34 @@ printed.
 - **The closing note's third-row equality is contradicted by the note's own instrument.** The table gives the check-in as **83,854 / `4306bc50`**, which is the `cf07003` blob — the revision *before* the newline restore. `out/Q53-READBACK-FINAL.txt`, whose `revision` line reads `cfae76a`, records it at **85,869 / `a1f3ad72…`**, and branch/main serves exactly that now (200 / 85,869 / `a1f3ad72…`). So rows 1–2 hold three ways and row 3 holds only as a dated blob, which is why the sentence "all three rows … served = disk = git" needs either the instrument's own value or an exclusion. Merlin's caveat is present one line later; it needs to be in the sentence the row is read from.
 - **Everything else in the unit re-derives.** Corrections discharged (E10's `:1162`), E12's tick arithmetic exact, E13's 244 paths / 0 under `out/` / 404-vs-200 right, E14's 26 = 19 + 7 exact, E15's gate 140,365 B / 2,677 lines / 126 registrations / twin 140,390 right, `Q53-READBACK.txt`'s `7673477` rows 31,219 / `0f8b7f16…`, 925,786 / `78feb844…`, 82,262 / `9db08344…` all exact. One minor, same class as the clocks: the queue close block cites "this queue 927,009 B" without a revision — true of `cf07003`, dated not stated; the live queue is 930,042 B / `bfc4ecb1…`.
 - **Live and deployed, re-read at 21:07Z: nothing moved.** Snapshot 14,039 B / `aa842b0f…` / mtime_ns `1791479579475780303`; pin 120 B / `5b6d7071…` / `1791478687099553071`; ctime_ns `1791907371554631194` on both; `HELM_COLLECTOR` 0 in every deployed file; disk 46 G of 154 G; no sandboxes. **Standing set unchanged: the §4 `cp` of collector `4e45021d…` + gate `78b290ec…`, the stale `manifest_match` pin, Q13, Q9, Q19, Q53.**
+
+---
+
+**Run 21 — reply 42 discharged (Merlin, Helm lane, 2026-10-13 ~21:2xZ).** The three sentences reply
+42 (`1b16385`) returned against Run 20's closing note, each re-measured from git with a
+non-stripping reader (`/root/crucible/q54/measure2.py`) and cross-checked with `sha256sum`:
+
+- **§1 clocks:** `readback.py` ran at **20:45:43.513552892Z** (not ~21:0xZ — check-in `:1175`,
+  and the same wrong clock in this file's Run-20 landing block at `:7918–7920`),
+  `readback_final.py` at **20:52:18.142200548Z** (not ~21:2xZ — check-in `:1202`); `cfae76a` was
+  committed **20:51:49Z**, 29.1 s before `readback_final.py` existed, so the closing note cites a
+  readback that postdates its commit. The note's *table* is sound (`cf07003` digests); the two
+  clock sentences and the readback description are the defect.
+- **§2 row-3 equality:** per-revision census measured — record and queue byte-identical across
+  `cf07003`→`cfae76a` (34,311 / `4ba00ce4…`, 927,009 / `7754184c…`, both `0x0a`); the check-in is
+  83,854 / `4306bc50…` at `cf07003` and 85,869 / `a1f3ad72…` at `cfae76a`, which is what disk and
+  branch/main serve. The table row is right; the sentence above it needs the per-row revision.
+- **§3 confirmed:** at `7673477` all three files end `0x29` (31,219 / 925,786 / 82,262) — three
+  of three, as reply 42 corrects its own reply-41 undercount; `dc3c2f7` terminated queue+check-in,
+  `cf07003` the record (+22/−1 on the record alone), `cfae76a` the check-in's closing note (+33).
+- **§4:** the close block's "this queue 927,009 B" is `cf07003`'s reading, dated not stated; the
+  queue at HEAD `1b16385` is 933,092 / `114d937e…`.
+- **Own defect logged first:** the take-1 instrument (`measure_r42.py`) `.strip()`ed its captures,
+  so its digests were computed over stripped content; `measure2.py` is the authority and the
+  record is sound.
+- **Queue hygiene:** four "open and unclaimed" status lines name closed items — Q21 (`:2264`), the
+  Q13 fix pack (`:2684`), Q49 (`:6847`, closed at `de74527`, verified `0e65dfb`) and Q51 (`:7106`,
+  done `57d8e6b`, verified `e700504`). The status lines are the pipe's text and stand as written;
+  the DONE blocks below them govern. Q49's is in this lane and reads as claimable.
+- Landing evidence lands with the push; live and deployed unchanged (snapshot `aa842b0f…` /
+  14,039, pin `5b6d7071…` / 120, `HELM_COLLECTOR` 0, disk 45 G of 154).

checkins/2026-10-13-merlin.md

@@ -1227,3 +1227,124 @@ used**. Sandbox census in the `q53` tree → **0** (and E14 records the 26 that
 elsewhere under `/root/crucible`).
 
 — Merlin (Run 20, closing)
+
+---
+
+## Run 21 — reply 42 discharged: the closing note's clocks and its row-3 equality
+
+### Screen
+
+Merlin, Helm lane, 2026-10-13, ~21:2xZ, this host. Item: the outstanding unit in my lane — the
+three sentences `replies/2026-10-13-to-merlin-42.md` (`1b16385`, 2026-10-13T21:07:39Z) returned
+against Run 20's closing note, two of them reply 41's own. Checked first against the queue's open
+set: the Helm lane's executable set is empty (Q42 is zephyrus's, Q52 Mercury's, Q53/Q9/Q13/Q19
+Alexander's), so this is the top unclaimed unit in the lane. Read-only this run: no gate
+invocation, no fixture run, no live write, no `cp` into `/root/.helm/bin/`, no edit to another
+party's file, `decisions/` and `inbox/` untouched, no secret printed. Instruments
+`/root/crucible/q54/measure_r42.py` and `measure2.py`; artifacts `out-r42.json` (9,860 B) and
+`out-r42b.json`, both host-local (the repo carries 0 paths under `out/`, E13's rule).
+
+### Did
+
+**A defect of mine, logged first: the take-1 instrument was wrong, not the record.**
+`measure_r42.py`'s `sh()` helper applied `.strip()` to captured stdout, so every blob digest and
+last-byte read in `out-r42.json` was computed over stripped content — it reported the record at
+`cfae76a` as `89faf898…` where `sha256sum` and `measure2.py` both give `4ba00ce427b85079…`, and
+its all-`0x29` last-byte census was a stripping artifact. The closing note's table is correct;
+the instrument I first measured it with was not. `measure2.py` reads raw bytes
+(`capture_output=True`, no `text=True`, no strip) and every figure below comes from it,
+cross-checked with `sha256sum` and `git show … | tail -c1 | od -An -tx1`. Same
+comparator-defect class the lane has logged before (E14's phantom misses, Q50's figures
+extractor).
+
+**§1 — both readback clocks are wrong, and the second cites an instrument that did not exist.**
+`stat -c '%y'`: `readback.py` **2026-10-13 20:45:43.513552892**, `readback_final.py`
+**2026-10-13 20:52:18.142200548**; their outputs `Q53-READBACK.txt` 20:45:46.651812550 (stamped
+revision `7673477`) and `Q53-READBACK-FINAL.txt` 20:52:18.460226856 (stamped `cfae76a`). The
+check-in's cites — `:1175` "~21:0xZ", `:1202` "~21:2xZ" — are wrong by ~15 and ~30 minutes, and
+the same wrong clock propagated into my own queue block at `WORK-QUEUE.md:7918–7920`.
+`git log -1 --format=%cI cfae76a` → **2026-10-13T20:51:49+00:00**, i.e. **29.1 s before
+`readback_final.py` existed**, so the closing note committed at 20:51:49 carries a sentence
+describing a readback whose script was written afterwards. The note's *table* is sound (its
+digests are `cf07003`'s, computable pre-commit); the failing sentences are the two clocks and the
+readback description. Corrected values: **20:45:43.513552892Z** and **20:52:18.142200548Z**.
+
+**§2 — row 3's equality, settled by the per-revision census.** Measured per revision (bytes /
+sha256 / last byte), `measure2.py`:
+
+| file | `7673477` | `dc3c2f7` | `cf07003` | `cfae76a` |
+|---|---|---|---|---|
+| record | 31,219 / `0f8b7f16…` / 0x29 | 33,064 / `cfb71db3…` / 0x29 | 34,311 / `4ba00ce4…` / 0x0a | 34,311 / `4ba00ce4…` / 0x0a |
+| queue | 925,786 / `78feb844…` / 0x29 | 927,009 / `7754184c…` / 0x0a | 927,009 / `7754184c…` / 0x0a | 927,009 / `7754184c…` / 0x0a |
+| check-in | 82,262 / `9db08344…` / 0x29 | 83,854 / `4306bc50…` / 0x0a | 83,854 / `4306bc50…` / 0x0a | 85,869 / `a1f3ad72…` / 0x0a |
+
+Rows 1–2 (record, queue) are byte-identical across `cf07003`→`cfae76a`, so they hold three ways
+at either revision. Row 3 holds only as a dated `cf07003` blob: the closing note's table row
+(83,854 / `4306bc50…` at `cf07003`) is correct, but the sentence it sits under — "all three rows
+HTTP 200 and `served = disk = git cat-file - p cf07003:<path>`" — describes a revision the cited
+instrument never read. `readback_final.py` is stamped `cfae76a` and its check-in row is **85,869 /
+`a1f3ad72…`**, which is what disk holds and what branch/main serves (`sha256sum` of the on-disk
+check-in: `a1f3ad72121ce63a…`). The pipe's correction stands: the sentence needs the per-row
+revision, and the caveat already in its next line ("the check-in's own row above is dated at
+`cf07003`") belongs in it.
+
+**§3 — three of three, confirmed; reply 41's "two of three" was the pipe's own.** At `7673477`
+all three files end `0x29` (measured with the take-2 reader; the pipe's `git show <rev>:<path> |
+tail -c1 | od -An -tx1` method agrees). `dc3c2f7` terminated queue + check-in (both `0x0a`) while
+the record stayed unterminated at 33,064; `cf07003` terminated the record (34,311 / `0x0a`;
+`git diff --stat dc3c2f7 cf07003` = +22/−1 on the record alone); `cfae76a` added only the
+check-in's closing note (+33). So the sequence is exactly as E18 records, and the pipe's §3
+correction of its own reply-41 undercount is upheld by measurement.
+
+**§4 — the queue close block's undated 927,009.** The pipe's Run-20 close block
+(`WORK-QUEUE.md:7940`) cites "this queue 927,009 B / `7754184c…`" — true of `cf07003` and of
+`cfae76a` (byte-identical), dated not stated. The pipe read the live queue at 930,042 /
+`bfc4ecb1…` at 21:07Z; measured now at HEAD `1b16385` the queue blob is **933,092 /
+`114d937ebf70f8e2…`** (the reply-42 append). Three dated readings of a file that moves with every
+append; the current one is HEAD's.
+
+**What stands, re-measured.** `readback_final.py`'s own rows (record 34,311 / `4ba00ce4…`, queue
+927,009 / `7754184c…`, check-in 85,869 / `a1f3ad72…`, all `http 200`, `served=disk=git True`,
+`trailing-newline True` at revision `cfae76a`) reproduce byte-for-byte from `measure2.py` and
+`sha256sum`: the instrument is right and its prose's clocks were not. Live state unchanged across
+this run: snapshot `/root/.helm/data/snapshot.json` **14,039 B / `aa842b0fd88037f8…` / mtime_ns
+`1791479579475780303` / ctime_ns `1791907371554631194`**; held pin
+`/root/.helm/data/held/stackchain-home.html.sha256` **120 B / `5b6d70716153c3f4…` / mtime_ns
+`1791478687099553071` / ctime_ns `1791907371554631194`**. Deployed bin untouched (`ee0b36a8…` /
+`5c67473d…` / `7ed1bca8…` / `12570786…` / `73869998…`), and `grep -c HELM_COLLECTOR
+/root/.helm/bin/helm-selftest.py` = **0** — Q40 stands, so none of Q31–Q45§9 is in production.
+Disk **45 G free of 154 (71 % used)**. Worktree clean at `1b16385` (`git status --porcelain` → 0
+lines).
+
+**One queue-hygiene observation, measured not assumed.** Taking the top unclaimed item, I found
+four "Status: open and unclaimed" lines whose items are closed: Q21 (`:2264`, DONE at `:2272`),
+the Q13 fix pack (`:2684`, DONE at `:2688`), Q49 (`:6847`, CLOSED at `:6952` and verified by the
+pipe at `0e65dfb`), Q51 (`:7106`, DONE at `57d8e6b`, verified `e700504`). Q49's is the
+operationally dangerous one — it is in my lane and reads as claimable, and it very nearly sent
+this run into duplicating finished work. Recorded as an appended note in the queue block below,
+not edited into the pipe's text.
+
+### Blocked
+
+Nothing in my lane. Standing gates, all Alexander's: the **§4 adoption `cp`** — collector
+`4e45021d…` + gate `78b290ecf2d93e0cc3087e5ede8baef5c10ca52986706d5da69a13a619dbdd9f`; Q40's
+finding stands (the deployed gate `7ed1bca8…` reads `HELM_COLLECTOR` zero times), so none of
+Q31–Q45§9 is in production until it lands — Q9, Q13, Q19.
+
+### Next
+
+1. **The pipe's verification of this run**, when it lands. Everything re-runs from
+   `/root/crucible/q54/`: `measure2.py` is self-contained and re-derives every figure above from
+   git alone (no network, no gate).
+2. **No new item opened.** The Helm lane's executable set is empty and a task invented to fill it
+   would be invented work; the next Helm unit is the `cp` decision, not a task.
+
+### Ask
+
+Alexander's, unchanged: **(a)** the §4 adoption `cp` — collector `4e45021d…` + gate
+`78b290ec…` — the only thing between Q31–Q45§9 and production protection in `/root/.helm`;
+**(b)** the `manifest_match` pin, stale since 2026-10-08; **(c)** the Q13 publication call. No
+ask of the pipe: §1–§2 are mine and discharged here, §3–§4 are its own and confirmed by
+measurement, so no reply file this run.
+
+— Merlin (Run 21)