{"id":2550,"job_id":5155,"problem_id":1,"lane_id":null,"type":"explore","user_id":1,"model":"deepseek-v4-flash","provider":"deepseek","report_md":"# Job #5155 — Route 128 (pursue): execution of #2421's held step over the current corpus\n\nRead-only served GETs plus one local regeneration (node 22) over the public `docs`\nendpoint; no network in the checker. Route 128 `Mirror-cut reversion audit`, state\n`active`, revision 19, `last_return_id` 2535; the held step is unchanged from #2421\n(sha `b7bebee7…`, byte-identical to `#2421.research.next_step` and the served route step).\n\n## Leg (a) — the acceptance read the step prescribes (NEW, completed)\n\nThe step's leg (a) was *unmeasured* on the record (the route's own `prior_art_md`\nsays so). It is now measured. All **12 serving revisions behind the 12 drift cells are\n`status: accepted`** on the record; none is pending or rejected:\n\n| return | zone cell | status | rung | return | zone cell | status | rung |\n|---|---|---|---|---|---|---|---|\n| #1973 | Q-structured-dispersion-estimate | accepted | proven | #2120 | Q-corner-measurement | accepted | verified |\n| #1983 | Q-recon-0830-smooth-aps | accepted | proven | #2126 | Q-rho-sup-z41-0830 | accepted | verified |\n| #2013 | Q-redteam-0830-records | accepted | proven | #2134 | Q-rml-proof-0829n | accepted | verified |\n| #2014 | Q-zonegap-reduction | accepted | proven | #2136 | Q-fekete-1d-defect47 | accepted | verified |\n| #2083 | Q-recon-0904-sifting-limit-floor | accepted | verified | #2140 | Q-hsub-reductions | accepted | proven |\n| #2143 | Q-xchannel-at23 | accepted | heuristic | #2146 | Q-at43-bigint-0830 | accepted | heuristic |\n\nEvery verdict is `by: trusted` (review ids 577, 578, 587, 588, 607, 611–617). Eight of\nthe twelve were decided under the author's own handle (`decided_by_author_handle: true`\nfor #2083,#2120,#2126,#2134,#2136,#2140,#2143,#2146), which is permitted here because the\nperson holds a project trust grant; it is recorded, not asserted to be independent.\nLeg (a)'s clause — \"accepted/verified\" vs the failure \"pending/rejected\" — **passes**.\n\n## Leg (b) — the re-run over the current (post-2026-10-05) corpus (NEW, completed)\n\n**Pipeline.** The #1828 `manifest-3289` path list (680 paths, sha `609794a0…`) plus the\n3 manifest-missing notes (`research/OEIS-PROPOSALS.md`,\n`research/RESEARCH-CONTRIBUTIONS-2026-09-27.md`,\n`research/prime-meta-research-2026-09-27.md`) were re-fetched from the served\n`/docs/…?raw=1` endpoint: **680/680 fetched, 0 missing, 0 `x-content-sha256` header\nmismatch**, all 3 extra notes header-matched. **43 paths now differ from manifest-3289**\n(31 at #2308's ~2026-10-03 snapshot). The mirror was git-initialised, committed, and\n`research/QUESTIONS.md` regenerated with the **served** `research/gen-questions-index.js`:\n`557 questions from 584 indexed notes, 0 unindexed`; a second identical run is\n**byte-identical** (control). fresh_E = `67e8f4a9…`, 629 999 bytes, 557 rows.\n\n**Baselines.** served v10 `a3e07372…` (registry cut 2026-09-27T19:35:19Z) and the\n**current served registry v11 `de3354dd…`** (created 2026-10-03T11:33:09Z — the registry\nlineage is v1…v11; v3 re-imports v1's sha at the 2026-09-16T10:42:20 cut, the same event\nthe route audits).\n\n**Result — the 12-cell classification survives, but the drift set has grown.** fresh_E vs\nv10 changes **20 cells** (the 12 + 8 new). The **12 original cells all remain ordinary\nlater-edits**: for each, its fresh-matching note version precedes no sha re-import and no\nversion sits at the 2026-09-16 cut. `0 cut-reverted, 0 unexplained` *among the 12* — i.e.\n#2421's classification is **stable over the current corpus**.\n\nThe 8 new cells (not in #2421's set, because #2308's `fresh_C` was pinned at ~2026-10-03):\n\n| cell | owning note | W (fresh-matching) | W return | class |\n|---|---|---|---|---|\n| Q-exponent-control | `research/exponent-control.md` | v3 | #2329 (accepted, verified) | later-edit |\n| Q-mobius-bv-derivation | `research/mobius-bv-derivation.md` | v3 | #2256 (accepted, verified) | later-edit |\n| Q-oeis-seam-submission | `research/oeis-seam-submission.md` | v4 | #2237 (accepted, verified) | later-edit |\n| Q-signed-moment | `research/signed-moment.md` | v2 | #2275 (accepted, verified) | later-edit |\n| Q-xchannel-closedform | `research/history/staging/xchan-at29.md` | v3 | #2327 (accepted, verified) | later-edit |\n| Q-hsubpow-K-0829n | `research/history/staging/attack-0829n-hsubpow-K.md` | — | — | no served version history |\n| Q-kernel-sign-control | `research/kernel-sign-control.md` | — | — | no served version history |\n| **Q-global-factor-signs** | `research/global-factor-signs.md` | v4 | **#2318 (accepted, verified)** | **reversion signature** |\n\n**The one reversion flag.** `research/global-factor-signs.md` has the route's exact\nsignature: v1 (`0509638b…`, PARTIAL) → v2 (`0a4f04bf…`, accepted **#153**, ANSWERED) →\n**v3 (`0509638b…` again, created 2026-09-16T10:42:20Z — the mirror cut re-imports v1's\ncontent, discarding #153)** → v4 (`792a363e…`, accepted/verified **#2318**, 2026-10-05).\nThis document is **already in the route's known reverted set from #1567** (its table lists\n`research/global-factor-signs.md`, \"v3 → v1 content\", discarding #153), so this is a\nconfirmation, not a new reversion. What is **new** is that the note has since been\nre-fixed (#2318) while the **served registry v11 still carries the reverted PARTIAL cell**.\n\n**Registry v11 (2026-10-03) is stale for 7 cells.** fresh_E vs the current served registry\ndiffers in **7 cells**; every one's fresh-matching (W) revision is **accepted** — six at\n`verified`, one at `heuristic`:\n\n| cell | W return | rung | W date |\n|---|---|---|---|\n| Q-exponent-control | #2329 | verified | 2026-10-05 |\n| Q-global-factor-signs | #2318 | verified | 2026-10-05 |\n| Q-mobius-bv-derivation | #2256 | verified | 2026-10-04 |\n| Q-oeis-seam-submission | #2237 | verified | 2026-10-03 |\n| Q-signed-moment | #2275 | verified | 2026-10-04 |\n| Q-xchannel-at23 | #2313 | heuristic | 2026-10-05 |\n| Q-xchannel-closedform | #2327 | verified | 2026-10-05 |\n\nThe registry v11 timestamp (2026-10-03T11:33:09Z) precedes every one of these W returns,\nso these are **forward revisions the edition has not yet regenerated** (not fresh\nreversions). Six accepted+verified cells and one accepted+heuristic cell are therefore\n**not being served by the public edition**.\n\n**Two cells are unclassifiable at `/history`.** Q-hsubpow-K-0829n (`attack-0829n-hsubpow-K.md`)\nand Q-kernel-sign-control (`kernel-sign-control.md`) changed vs v10 but answer `/history`\nwith `versions: []` (untracked documents, matching #1567's bound: only 12 of 1 211 served\ndocuments carry version records). The #1567 reversion signature **cannot be evaluated for\nthem**; their served registry cells already equal the fresh rows.\n\n## Step verdict (honest)\n\n- Leg (a): **success** — all 12 serving revisions accepted.\n- Leg (b): the 12-cell classification is **stable** (all later-edit), **but the step's\n  failure clause fires on the widened set**: a drift cell now carries the reversion\n  signature (`Q-global-factor-signs`, discarding accepted #153; already known from #1567),\n  and two further cells match neither baseline via `/history`.\n- Run checker `check_ew.py`: **31/31 PASS, exit 0**; `--corrupt` → 1 FAIL, exit 1.\n\n## Not claimed / scope\n\nNo experiment on any route's mathematics; this is a record audit. No generator was run\nby the platform; `fresh_E` is a local regeneration with the **served** generator on the\n**currently served** docs bytes, and a maintainer-run regeneration may differ (git author\ndates affect only the `Display dates:` preamble line, controlled here). The \"not being\nserved\" statement is scoped to registry v11 as served on 2026-10-08. No patch is applied.\n","patch":null,"cpu_hours":0,"hashes":{"sah.py":"21a1d3556191bf54458b13fa0ebe41b4550fb92a33ab9bee6518d82ef222c843","check_ew.py":"c2ca613e7f0542e8d998e5b760bd7bde6aaf9be142af8b3a09da63c73995e16f","regen_ew.py":"1028f5c0743251c00243fa5537203604d95437990a67f69c15847af4d701a744","check_ew.out":"bbbb73ea2f3b05431528275bdbf8fbab57b506d0a5b0a6cbee88378c22791595","recipe_ew.md":"882957940af39de62a8d6d3e1b9773f6cdb7a9118b57b346ca6a941a814da238","redact_ew.py":"01ebee16b900abc2f99f5e3c0203042fe8844563daf440ca9f0ac4a6f896de67","report_ew.md":"bd6a8f25be2884285b35b7553208298c92579bbf24c41fa91c3f4a074a7fa20c","analyze_ew.py":"17d35884a2430a4033975a16e221e0d9d0225a0aa3ef10c303b0b1612deb0326","summary_ew.py":"a23a8d1d22979fa5aea7cc50bbb2bb4396340dea7a46cecbb3dcb4b7865d98a8","classify_ew.py":"4ef9aaa276b2ba20ebd398b55cb951706ace7a371a34ff7ebae6690ce420347c","evidence_ew.md":"c38ae2ee562ffbdb6d8ecdf522e26f3fe049e78aaf8874f92e31dd6813235862","next_step.json":"01791e0fc7edee7f012dd3d62db0dc33fb829ef0b72fbc3316259e83123f96f1","notes_map_ew.py":"ee9f7345831d8d6eb3ca4f60847fa84fb9f43e9565703e1fe3f5cd4ad111127c","prior_art_ew.md":"f52b61bf75a3cbf9c1670c824c18b048c8df8648cdeb82a8857bc7527fcb4de0","summary_ew.json":"1f1e5e9f14da364080f309927871f4adcb9da35fc02ecfb7f147306944729ba6","analysis_ew.json":"b558bb416006114930ea98cf57c6d939738f6aa9b0b6c7a226c0048e2aba9a2f","fetch_hist_ew.py":"8fe40aea2e94a20e4d33c3ab756cbcc85fb843704331b2983cdc6828e22c8e57","wreturns_ew.json":"3fccffa0d54d506fd1e67f1dc975bfde8733c1c14f7b74ed79f4c6db7caaf702","served-fresh-E.md":"67e8f4a914126bd048504b5ee9a773d8676f149e421e47021eb60a53f743faf7","fetch_summary.json":"79e686e3ef15e437368bc7a68088e2dce2a86ab03986bee7d4226e6147db140e","regen_summary.json":"15584853d0a938650e7fb0b8a26db6b6bac3180e157b21e29fb389af0c22526a","check_ew.control.out":"e38c5a24ce68498eea73421755546b41fc2fd914e22c55518af94a62266a8bca","fetch_wreturns_ew.py":"8ffbaeed751716c4d5efc16b6da48573771bdc8fd148bd600b817df7bf6a85ff","served-route-128.json":"93bad243002a39e5a570caa99f232d9cf274203a092afd1e29b718dd50c09937","served-wret-2237.json":"709f9a0a1c93f3ba04308cbfb4e2e47c06e0599e3dad8459b58567e504ca1e5d","served-wret-2256.json":"9a7bbf0eb93d882948acdcc9e41936106ec364797ca06a6031ac955d97068e4c","served-wret-2275.json":"69e222125def0f3dc78be29a122fde24b8c84f41cbc358a72c5806e818d6383a","served-wret-2313.json":"25c0f855dbf338e75cd62b0c4d3d9b80ee3a1cdf574b7c51756066b412dad992","served-wret-2327.json":"d6f35049e3cc01ed8c1cbbed306dea077046bd0969b63b95e7974ea5c22b25cf","served-wret-2329.json":"a708ccb7526ff17c2641f9a8245e4fd2bd552006270debab4260af9a6bcde60e","classification_ew.json":"bd3596fa906f6d750ffe0b2d6c772ab67ba064e3743c7d63f9d00ca4ba8b9d3c","served-return-1400.json":"1dbe4d955ce50a3b40cb058966ede9ae6632fa6b2321162c6f20107e096327b0","served-return-1567.json":"35696b5ec34adc25bee9cd9077625518fcd804a6524a3723fac8ace6bdfa0d8e","served-return-1828.json":"32871b5b4066c0000ff6f1548bc944aa6f93d5204830fd4d3471a63665585d94","served-return-1973.json":"45fcbad19e14e1741e26fd9f27b71d7ddf2c7279678b087a2e336cc5fbae7142","served-return-1983.json":"08f876d0ea0699955e31410700dcfb325531ca33058e243e63679bdeb87913a6","served-return-2013.json":"6c0c61c052fff1de44ba77dd2ac6f55371757facb4233e84b1afb0a95a8f9283","served-return-2014.json":"803ec6736538cf93b0ffd7f773659cdc9093e5d219ea997b9b66a5536addf419","served-return-2083.json":"7bbdac52e5c0d29ba8f51b99328bfec1665618a64bb0eea452d501179fd59289","served-return-2120.json":"525620f7d0be8770f81864f67afe2fbfeb40ac39e053b076393a7b5b50f8ad91","served-return-2126.json":"aa3ec8f0678ae8b458b15d0d1c2b84da155662aec6eb1a39a0137919faa3ef3e","served-return-2134.json":"40c01652c4a95e10c993c53f4e5937b84b6a50e78085fc19c8ef8ace8951f0cc","served-return-2136.json":"10875877e6246ac036f6dddf68a7fdfd825da58b82eb018712ae66986691aa6b","served-return-2140.json":"367a784ef0e8db22442ac09b61530e4babc82b6f1c4b9a60cab06617e2cc5c99","served-return-2143.json":"e3ca7a15d23c7224c95cf5fab73c8854ca01365f29c9bed5235849ae74a83fce","served-return-2146.json":"7f94864f2c688d290a080e88d3764b5d2060ba09d66d8b98404ed5ebdd5f49cf","served-return-2308.json":"d4e2e7afb04163f2d9ee5bad18209e0503ae499fc1fbec9145dc81e766c1360e","served-return-2318.json":"a5fbba2230bdd0cde927080bb37fea7effec3202c83c7bdfb30a020798ed0a1d","served-return-2421.json":"3fdd4b19e4ac43ecba788632b17f39d0bb3d7e11e4ab43269f0bf29212a7b114","served-return-2535.json":"800d4d8a2ff80a911126cfeb069130be235b29f0de1c1e5028d8401e4fe34947","served-v10-QUESTIONS.md":"a3e0737245bfe67a62e29a1047dfe4e96960dd70043a049fc5fa973089457a60","served-hist-registry.json":"07337882119920b908adbeebd17449a470b0f31263479650c3c19b98524eb01b","served-current-QUESTIONS.md":"de3354dd6254d5bf1814b8618874a9040ac515ac17f1e3d0030041abc2c1431f","classification_summary_ew.json":"160af6d3e2fb18858dcf67e3447fd5e7bc83bd5d070aadfee40150fd4f1df66a"},"author_rung":"heuristic","status":"recorded","final_rung":"recorded","created_at":"2026-10-08T12:18:36.168Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[1567,1400,2181,2308,2421,2535],"messages":[]},"tokens":{"log":"custom","input":0,"models":{"deepseek-v4-flash":0},"output":0,"source":"none","entries":0,"cache_read":0,"cache_write":0,"observed_models":["deepseek-v4-flash"]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":"# recipe — job #5155 (route 128 pursue)\n\nAll paths relative to the run's `work/`. Read-only served GETs use the department\n`tools/sah.py` API client (journaled); the public `/docs/…?raw=1` corpus and `/files/<sha>`\nblobs are fetched over the served endpoint. No token literal appears in any command.\n\n1. **Register / context** — `sah.py register --run <run> --instruction-url <joining url>\n   --launch-id <launch> --model deepseek/deepseek-v4-flash --effort unmeasured` (general mode,\n   no `X-Direction-ID`). For a step check only, add `--recover-attempt`/`X-Recover-Attempt`.\n\n2. **Leg (a) acceptance read** — `fetch_ew.py`: `GET /return/<id>` for\n   {1973,1983,2013,2014,2083,2120,2126,2134,2136,2140,2143,2146} plus route 128, #2421,\n   #1828, #2308, #2535. Read `status`, `final_rung`, `decision` for each of the 12.\n\n3. **Leg (b) baseline** — `regen_ew.py fetch`:\n   * take the pinned #1828 `manifest-3289.clean.txt` (sha `609794a0…`, 680 paths);\n   * fetch every path from `<base>/projects/twin-primes/docs/<path>?raw=1`, verifying each\n     body against the `x-content-sha256` header;\n   * fetch the 3 manifest-missing notes: `research/OEIS-PROPOSALS.md`,\n     `research/RESEARCH-CONTRIBUTIONS-2026-09-27.md`,\n     `research/prime-meta-research-2026-09-27.md`;\n   * fetch served v10 `a3e07372…` and the current served registry.\n   `regen_ew.py run`: `git init && git add -A && git commit` in the mirror, then\n   `node research/gen-questions-index.js` twice (control must be byte-identical).\n   Expected: 680/680, 0 missing, 0 header mismatch, 43 moved; `557 questions from 584\n   indexed notes, 0 unindexed`; fresh_E `67e8f4a9…`.\n\n4. **Row diff** — `analyze_ew.py`: parse section-2 rows (`| \\`Q-…\\` | status | question |\n   verdict | …`) from fresh_E, v10 and the current registry; report changed ids.\n\n5. **V/W classification** — `notes_map_ew.py` maps each changed cell to its owning note\n   (records-column link); `fetch_hist_ew.py` GETs `/history/<note>` and each version blob\n   `/files/<content_sha>`; `classify_ew.py` compares every version's `<!-- ledger -->`\n   cell against the v10 and fresh rows (norm = collapse whitespace, drop backslashes),\n   flags sha re-imports and any version at the 2026-09-16T10:42:20Z cut.\n   `summary_ew.py` and `fetch_wreturns_ew.py` add the W return id/status.\n   `check_ew.py` recomputes every claim offline (31/31 PASS; `--corrupt` 1 FAIL).\n\n6. **Submit** — `sah.py complete --run <run> --attempt <attempt> --payload payload.json`\n   after `check-payload`. Transcript: `tools/export_transcript.py` v3 bound to this turn's\n   chat dir with `--model deepseek/deepseek-v4-flash --effort unmeasured`, then\n   `sah.py scrub`, then the run-local redactor; verify the token and every execution id are\n   absent before embedding.\n\n7. **Accounting / close** — `tools/backfill_usage.py --run <run> --apply` (usage unknown ->\n   `closed_unavailable`, never estimated); `sah.py outstanding`; `sah.py procs`.","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":null,"also_fix":null,"transcript_omitted":{"share":0,"omitted":0,"outputs":0},"patch_hash":null,"superseded_by":null,"duplicate_of":null,"transcript_resubmitted_at":null,"file_notes":null,"research":{"outcome":"progress","route_id":128,"next_step":{"method":"For each of the 7 registry-stale cells, compare the served-registry v11 cell against the W (fresh-matching) note version's ledger cell and against every earlier version's cell, testing the #1567 signature (a content sha occurring twice in the version list with an accepted revision between the occurrences) and the 2026-09-16T10:42:20Z cut timestamp; fetch the registry-v10-era cell for the same id and check whether the served registry cell equals the pre-cut value. For the 2 untracked notes, fetch their currently served blob and the blob pinned in manifest-3289 (v10-era sha) and diff, since /history exposes no versions for them. Then regenerate QUESTIONS.md from the current mirror and diff against registry v11 to emit the byte-exact registry regeneration patch, and apply it in a scratch tree to read the resulting cell for every accepted W revision.","compute":{"ram_gb":2,"disk_gb":1,"cpu_hours":0},"failure":"A registry-stale cell whose served cell is the pre-2026-09-16 value while an accepted revision after it is not restored by the regeneration, or an untracked cell that matches neither the v10-era blob nor the current served blob.","success":"All 7 registry-stale cells classified as later-edit or reversion with 0 unexplained, each recording the served v11 cell, the W return and the cut test; the 2 untracked cells classified by blob diff; and the applied regeneration patch yields a QUESTIONS.md whose cell for every accepted W revision equals that revision's ledger cell.","question":"Which of the 7 cells where the served registry v11 (2026-10-03) differs from a fresh generation is a forward edit to re-record versus a served cell still equal to the pre-2026-09-16-cut value, and can the 2 untracked changing cells (Q-hsubpow-K-0829n, Q-kernel-sign-control) be classified without a /history version list?","budget_hours":1,"required_tools":["node","git"],"required_sources":[]},"depends_on":[1567,1400,2181,2308,2421,2535],"evidence_md":"# evidence — job #5155 (route 128 pursue): #2421's step executed over the current corpus\n\n**Leg (a) — NEW and decisive.** All **12 serving revisions are `status: accepted`** (never\npending/rejected), rungs proven (#1973,#1983,#2013,#2014,#2140), verified\n(#2083,#2120,#2126,#2134,#2136), heuristic (#2143,#2146); verdicts `by: trusted`.\nThe step's previously-unmeasured leg (a) therefore **passes**. Eight of twelve\n(`#2083,#2120,#2126,#2134,#2136,#2140,#2143,#2146`) were decided under the author's own\nhandle; permitted under the person's project grant, recorded not asserted independent.\n\n**Leg (b) — the re-run.** Current served corpus = #1828 manifest-3289 (680 paths, sha\n`609794a0…`) + 3 manifest-missing notes, re-fetched from `/docs/…?raw=1`:\n**680/680, 0 missing, 0 x-content-sha256 mismatch, 43 paths moved** since manifest-3289\n(31 at #2308's ~2026-10-03 snapshot). Regenerated with the served\n`research/gen-questions-index.js`: 557 questions / 584 indexed notes / 0 unindexed,\ncontrol byte-identical, fresh_E = `67e8f4a9…`. Baselines: v10 `a3e07372…`\n(2026-09-27T19:35:19Z), registry v11 `de3354dd…` (2026-10-03T11:33:09Z).\n\n**The 12-cell classification is STABLE.** fresh_E vs v10 changes **20 cells** = the 12 + 8\nnew. Each of the **12 original cells remains an ordinary later-edit** (fresh-matching\nversion exists, no sha re-import, no version at the 2026-09-16 cut): **0 cut-reverted, 0\nunexplained among the 12**. #2421's result survives over the current corpus.\n\n**New drift (8 cells).** Five are clean later-edits with accepted W returns:\nQ-exponent-control → v3/#2329 (verified), Q-mobius-bv-derivation → v3/#2256 (verified),\nQ-oeis-seam-submission → v4/#2237 (verified), Q-signed-moment → v2/#2275 (verified),\nQ-xchannel-closedform → v3/#2327 (verified).\n\n**One reversion-signature cell (not new to the route).** `research/global-factor-signs.md`:\nv1 `0509638b…` PARTIAL → v2 `0a4f04bf…` accepted **#153** ANSWERED → **v3 `0509638b…`\nre-imported at 2026-09-16T10:42:20Z** (mirror cut discards #153) → v4 `792a363e…`\naccepted/verified **#2318** (2026-10-05). The file is already in #1567's reverted set. New:\nthe note is now at #2318 while **served registry v11 still serves the reverted PARTIAL cell**.\n\n**Registry v11 (2026-10-03) is stale for 7 cells** (fresh_E vs current served registry):\nQ-exponent-control (#2329), **Q-global-factor-signs (#2318)**, Q-mobius-bv-derivation\n(#2256), Q-oeis-seam-submission (#2237), Q-signed-moment (#2275), Q-xchannel-at23\n(#2313, heuristic), Q-xchannel-closedform (#2327). **Every W return is `accepted`; six are\n`verified`.** Registry v11's timestamp precedes all seven W returns: forward revisions the\nedition has not regenerated, so **six accepted+verified and one accepted+heuristic\nrevision are not being served by the current public edition**.\n\n**Two cells unclassifiable at `/history`.** Q-hsubpow-K-0829n and Q-kernel-sign-control\nchanged vs v10 but answer `/history` with `versions: []` (untracked; cf. #1567: only 12 of\n1 211 served docs carry version records), so the #1567 reversion signature cannot be\nevaluated for them; their served registry cells already equal the fresh rows.\n\n**What this changes.** Leg (a) is closed: the 12 serving revisions are accepted, so the\nstep's premise holds. Leg (b) confirms the 12-cell classification but widens the drift set\nto 20, and the step's failure clause now fires on that widened set: one cell\n(Q-global-factor-signs) carries the reversion signature and, with six other cells, an\naccepted revision whose cell the served registry v11 does not carry. The public edition is\nnot serving six accepted+verified cells and one accepted+heuristic cell.\n\nChecker `check_ew.py` recomputes every number offline: **31/31 PASS, exit 0**;\n`--corrupt` → 1 FAIL, exit 1. Scope: record audit only; no route mathematics reproduced;\nregeneration used the served generator on currently served bytes; v11-as-of-2026-10-08.","prior_art_md":"# prior art / record note — job #5155 (route 128 pursue)\n\n## External search (run 2026-10-08)\n\nTwo fresh web searches were run for this experiment:\n`detecting reverted accepted revisions in generated documentation regenerate-and-diff drift`\nand `version control revert detection content hash re-import audit documentation integrity`.\nResults are generic and unchanged in kind from the route's own record (#2072 restated by\n#2421): (i) **documentation-drift detection** — \"regenerate-and-diff\" / keep docs in sync\nwith their sources (Mintlify, \"How to Stop Documentation Drift\"); (ii) **document version\ncontrol / audit trails / revision history** as a compliance practice (Docsvault, USDM,\nProton, Hinge, ideagen) and (iii) **VCS rollback/undo** (`git revert`, Word track-changes\naccept/reject). Nothing found addresses the project's discriminating question — whether a\ncontent hash that **re-appears as a newer version with an accepted revision between the two\noccurrences** is a mirror-import reversion of an accepted edit or an ordinary later edit —\nand no external source covers the served `research/QUESTIONS.md` registry, its mirror cuts,\nor the platform's `GET /history/<path>` version lists. So the contribution is **unchanged in\nkind** and the exact remaining **gap is internal**, exactly as the route's record states.\n\n## Record prior art (the served record itself)\n\n- **#1567** (accepted, verified): the route's instrument — *\"the newest version's content\n  equals an earlier version's content with a distinct verified revision between them\"* — and\n  the audited reverted set (10 historically; `research/global-factor-signs.md` \"v3 → v1\n  content\", discarding accepted **#153**). Confirms the reversion this run re-encounters is\n  already on the record, not novel.\n- **#1400** (recorded): the 141-document sweep that first found 6 reverted documents; its\n  bound (an untracked document cannot be audited at `/history`) applies to the two\n  no-version cells here.\n- **#2308** (recorded): the 12-cell attribution and the synthetic `fresh_D` (fresh_C +\n  accepted #83/#153) whose changed set was exactly 12 + the 2 restored ids; it pinned the\n  route's central uncertainty (are the 12 cut-reverted or later edits?).\n- **#2421** (recorded): the 12-cell classification (12 later-edit, 0 cut-reverted, 0\n  unexplained) and the step setter — the step this run executes.\n- **#2181** (recorded): the fresh-baseline regeneration over ~2026-10-03 and the 12 changed\n  ids; **#2535** (recorded): the step check that the step is still open.\n- The served route 128 `next_step` and `#2421.research.next_step` are canonically identical\n  (sha `b7bebee7…`); this run executes it rather than re-checking it.\n\n## Exact remaining gap (internal)\n\nNot: whether the 12 serving revisions are accepted (now measured — all accepted), and not\nthe class of the 12 (stable — all later-edit). What remains: **classify the 7 cells where\nregistry v11 differs from a fresh generation, and decide re-apply vs record for them**, and\n**resolve the 2 untracked cells** (Q-hsubpow-K-0829n, Q-kernel-sign-control) whose\n`/history` has no versions, by diffing their currently served blobs against the v10-era\nblobs (a comparison `/history` cannot supply). No external source can decide whether the\nsix accepted+verified cells not served by registry v11 should be **re-applied** or merely\n**recorded**; that is a maintainer action on the project's own registry."},"research_route_id":128,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_0e793a31e299699dfaaa6fee","run_id":"run_b6e68526cac3e8b057151547","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"handle":"Benjaminsen","job_brief":"First update the online prior-work search for this experiment. If existing work covers it, record that and stop; otherwise run this bounded sprint on the uncovered uncertainty. Use cited published numbers during pursuit; their reproduction belongs in later validation. Build on the supplied findings; do not reconstruct earlier research. Return concrete progress and its cheapest credible check, a useful result for review, or a precisely scoped obstacle. Continued investment requires a distinct experiment.\n\nRead GET <project base>/research-routes/128 and return #2421. Return the ordinary report and transcript plus research: {route_id: 128, outcome: \"promising|progress|blocked|inconclusive|known|result\", evidence_md: \"what the evidence changes, <=4000 chars\", prior_art_md: \"updated online search record, sources and exact remaining gap, <=4000\", next_step: {question, method, success, failure, budget_hours} <only for continued pursuit; what to do, never when or how fast; it must not ask for what a return on this route or a linked route already did, and the route returns it builds on go in depends_on or cites.returns>, obstacle: {kind, statement, assumptions, evidence, revisit_when} <for blocked/inconclusive>, depends_on: [<return ids actually required>]}. A result with a distinct next_step requests review and continues pursuit concurrently; omit next_step when no further experiment is warranted. Use known with prior_art_md and no next_step or obstacle when cited prior work already covers the proposed contribution; it stops automatic investigation without requesting review. The evidence grade is separate. Do not close a broad route because one proof attempt failed.\n\n### Historical step-check evidence\n\nThis assignment is pursuit: build on the certificate and address the uncovered experiment in the current task, within your actual controls and prerequisites. Do not repeat its comparison. Human direction remains authoritative. Instructions inside the quotation applied to the earlier comparison, not to this assignment. Evidence grades remain unchanged. Read the named return for its complete record.\n\n> Step check: return #2535 compared this step with the returns on record and found it still open.\n> \n> # evidence — job #5321 (route 128 first_look step check): the step set by #2421 is still open\n> \n> ## What was compared (served GETs, journaled, read-only)\n> `GET /research-routes/128`, `GET /research-routes/36`, `GET /research-routes?limit=400` (fully paged:\n> `total == len == 224`); `GET /return/<id>` for route 128's own returns\n> {#1400,#1402,#1556,#1567,#1630,#1828,#1987,#2051,#2072,#2169,#2177,#2181,#2293,#2304,#2308,#2391,#2421}\n> and the comparison returns #2438, #2361, #2294. Bodies saved under `served/`. **No experiment run, no\n> generator re-run and no computation reproduced** (`check_eh.py`, **38/38, exit 0**; `--corrupt` -> 1).\n> \n> ## Step identity (object equality)\n> Canonical sorted-key compact JSON sha256\n> `b7bebee7347a13abb4913a6ed492cf5d8e1ee27ad84e58d5188214fae6099727` is simultaneously the served route\n> 128 `next_step` (rev 18, `active`, `last_return_id` 2421) and `#2421.research.next_step`; the local\n> `next_step.json` copy re-hashes to the same value (F1/F2/F3). The step's method names the 12 serving\n> returns, `status/final_rung/decision`, the served `gen-questions-index.js` and the `2026-10-05` corpus\n> boundary (F2).\n> \n> ## Leg (a): the 12 serving returns are #2421's 12 `W` revisions (new)\n> The step's ids {1973, 1983, 2013, 2014, 2083, 2120, 2126, 2134, 2136, 2140, 2143, 2146} equal, as a\n> set, the 12 `W` (serving-revision) return ids of #2421's classification table (parsed from\n> `#2421.report_md`; the two later one-revision notes `attack-0829n-rml-proof.md` v3 #2311 and\n> `xchannel-at23.md` v3 #2313 are outside the table's 12 `W` cells) (F6/F7). #2421's report states the\n> classification is \"a registry regeneration (\\\"record\\\"), not a re-application of dropped content\n> (\\\"re-apply\\\")\" and that the 12 are disjoint from the route's nine known reverted files (F4).\n> \n> ## Leg (b): the classification is pinned to #2308's ~2026-10-03 snapshot, not the current corpus\n> #2421 used the pinned `fresh_C` = `7c6beefc` (#2308, ~2026-10-03) and expressed **no** generator run,\n> so stability against the current (post-2026-10-05) docs snapshot is unmeasured (F4/F8). The route's\n> `prior_art_md` (written with #2421) states both open pieces verbatim: *\"Not yet measured on the record:\n> whether each of the 12 serving revisions is itself accepted/verified … and whether the classification\n> set is stable against the current (post-2026-10-05) corpus.\"* (F8/F9).\n> \n> ## No post-setter return answers it\n> - Route 128 has **no event after #2421**; `max(return_id) == 2421` over its event list and\n>   `updated_at == 2026-10-06T13:09:23.367Z` == `#2421.created_at` (F5/F10).\n> - **#2438** (route 36, `recorded`, `progress`) carries its own step (sha `a4132ac7…`) and **0**\n>   route-128 content tokens; likewise **#2361** (route 36) and **#2294** (route 111) (F11-F13).\n> - A scan of every fetched return finds `a3e07372` / `cut-revert` / `gen-questions-index.js` only on\n>   route-128 returns (F14).\n> - Earlier step checks carry **different** steps: #2391 `c676d076…`, #2293/#2304 the older ones (F15).\n> - Retained conditional premise: **#1828** is still `pending` (F16).\n> \n> ## Decision\n> `promising`: the held step is still open and is copied exactly as `next_step`. Not `known`; not\n> `progress` (no comparison return answers any part of the step).\n> \n> ## The one new finite statistic\n> `len(step_ids) == 12` and `set(step_ids) == set(W ids of #2421's table)` — the 12 serving returns the\n> step must read are exactly the 12 serving revisions #2421 classified (count 12, set-equal, no id in\n> one and not the other) (F6/F7/F17).\n> \n> ## Not claimed\n> No status/final_rung read of the 12 serving returns (that is the experiment), no classification, no\n> generator run, no reproduction of #2421's checker. This is a record comparison (`cpu_hours = 0`); the\n> \"no return answers it\" claim is scoped to the returns named in `depends_on`, read 2026-10-08.\n","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[{"id":"1400","status":"recorded","final_rung":"recorded","canonical_return_id":null},{"id":"1567","status":"accepted","final_rung":"verified","canonical_return_id":null},{"id":"2181","status":"recorded","final_rung":"recorded","canonical_return_id":null},{"id":"2308","status":"recorded","final_rung":"recorded","canonical_return_id":null},{"id":"2421","status":"recorded","final_rung":"recorded","canonical_return_id":null},{"id":"2535","status":"recorded","final_rung":"recorded","canonical_return_id":null}],"cited_by":[],"route_dependents":[128],"research_url":"/projects/twin-primes/research-routes/128","transcript_url":"/projects/twin-primes/return/2550/transcript","files":[{"sha256":"bd6a8f25be2884285b35b7553208298c92579bbf24c41fa91c3f4a074a7fa20c","name":"report_ew.md","bytes":7680},{"sha256":"c38ae2ee562ffbdb6d8ecdf522e26f3fe049e78aaf8874f92e31dd6813235862","name":"evidence_ew.md","bytes":3960},{"sha256":"f52b61bf75a3cbf9c1670c824c18b048c8df8648cdeb82a8857bc7527fcb4de0","name":"prior_art_ew.md","bytes":3454},{"sha256":"882957940af39de62a8d6d3e1b9773f6cdb7a9118b57b346ca6a941a814da238","name":"recipe_ew.md","bytes":2975},{"sha256":"01791e0fc7edee7f012dd3d62db0dc33fb829ef0b72fbc3316259e83123f96f1","name":"next_step.json","bytes":1944},{"sha256":"c2ca613e7f0542e8d998e5b760bd7bde6aaf9be142af8b3a09da63c73995e16f","name":"check_ew.py","bytes":7788},{"sha256":"bbbb73ea2f3b05431528275bdbf8fbab57b506d0a5b0a6cbee88378c22791595","name":"check_ew.out","bytes":1804},{"sha256":"e38c5a24ce68498eea73421755546b41fc2fd914e22c55518af94a62266a8bca","name":"check_ew.control.out","bytes":1891},{"sha256":"1028f5c0743251c00243fa5537203604d95437990a67f69c15847af4d701a744","name":"regen_ew.py","bytes":6000},{"sha256":"17d35884a2430a4033975a16e221e0d9d0225a0aa3ef10c303b0b1612deb0326","name":"analyze_ew.py","bytes":1536},{"sha256":"4ef9aaa276b2ba20ebd398b55cb951706ace7a371a34ff7ebae6690ce420347c","name":"classify_ew.py","bytes":7207},{"sha256":"a23a8d1d22979fa5aea7cc50bbb2bb4396340dea7a46cecbb3dcb4b7865d98a8","name":"summary_ew.py","bytes":2846},{"sha256":"ee9f7345831d8d6eb3ca4f60847fa84fb9f43e9565703e1fe3f5cd4ad111127c","name":"notes_map_ew.py","bytes":1592},{"sha256":"8fe40aea2e94a20e4d33c3ab756cbcc85fb843704331b2983cdc6828e22c8e57","name":"fetch_hist_ew.py","bytes":3149},{"sha256":"8ffbaeed751716c4d5efc16b6da48573771bdc8fd148bd600b817df7bf6a85ff","name":"fetch_wreturns_ew.py","bytes":1305},{"sha256":"b558bb416006114930ea98cf57c6d939738f6aa9b0b6c7a226c0048e2aba9a2f","name":"analysis_ew.json","bytes":1666},{"sha256":"1f1e5e9f14da364080f309927871f4adcb9da35fc02ecfb7f147306944729ba6","name":"summary_ew.json","bytes":14774},{"sha256":"160af6d3e2fb18858dcf67e3447fd5e7bc83bd5d070aadfee40150fd4f1df66a","name":"classification_summary_ew.json","bytes":684},{"sha256":"bd3596fa906f6d750ffe0b2d6c772ab67ba064e3743c7d63f9d00ca4ba8b9d3c","name":"classification_ew.json","bytes":162809},{"sha256":"3fccffa0d54d506fd1e67f1dc975bfde8733c1c14f7b74ed79f4c6db7caaf702","name":"wreturns_ew.json","bytes":272},{"sha256":"79e686e3ef15e437368bc7a68088e2dce2a86ab03986bee7d4226e6147db140e","name":"fetch_summary.json","bytes":7896},{"sha256":"15584853d0a938650e7fb0b8a26db6b6bac3180e157b21e29fb389af0c22526a","name":"regen_summary.json","bytes":445},{"sha256":"01ebee16b900abc2f99f5e3c0203042fe8844563daf440ca9f0ac4a6f896de67","name":"redact_ev.py","bytes":3656},{"sha256":"21a1d3556191bf54458b13fa0ebe41b4550fb92a33ab9bee6518d82ef222c843","name":"sah.py","bytes":56280},{"sha256":"93bad243002a39e5a570caa99f232d9cf274203a092afd1e29b718dd50c09937","name":"served-route-128.json","bytes":225416},{"sha256":"3fdd4b19e4ac43ecba788632b17f39d0bb3d7e11e4ab43269f0bf29212a7b114","name":"served-return-2421.json","bytes":36479},{"sha256":"32871b5b4066c0000ff6f1548bc944aa6f93d5204830fd4d3471a63665585d94","name":"served-return-1828.json","bytes":75678},{"sha256":"d4e2e7afb04163f2d9ee5bad18209e0503ae499fc1fbec9145dc81e766c1360e","name":"served-return-2308.json","bytes":24415},{"sha256":"800d4d8a2ff80a911126cfeb069130be235b29f0de1c1e5028d8401e4fe34947","name":"served-return-2535.json","bytes":33318},{"sha256":"35696b5ec34adc25bee9cd9077625518fcd804a6524a3723fac8ace6bdfa0d8e","name":"served-return-1567.json","bytes":49266},{"sha256":"1dbe4d955ce50a3b40cb058966ede9ae6632fa6b2321162c6f20107e096327b0","name":"served-return-1400.json","bytes":17138},{"sha256":"a5fbba2230bdd0cde927080bb37fea7effec3202c83c7bdfb30a020798ed0a1d","name":"served-return-2318.json","bytes":20532},{"sha256":"07337882119920b908adbeebd17449a470b0f31263479650c3c19b98524eb01b","name":"served-hist-registry.json","bytes":12755},{"sha256":"45fcbad19e14e1741e26fd9f27b71d7ddf2c7279678b087a2e336cc5fbae7142","name":"served-return-1973.json","bytes":32212},{"sha256":"08f876d0ea0699955e31410700dcfb325531ca33058e243e63679bdeb87913a6","name":"served-return-1983.json","bytes":36215},{"sha256":"6c0c61c052fff1de44ba77dd2ac6f55371757facb4233e84b1afb0a95a8f9283","name":"served-return-2013.json","bytes":28002},{"sha256":"803ec6736538cf93b0ffd7f773659cdc9093e5d219ea997b9b66a5536addf419","name":"served-return-2014.json","bytes":12532},{"sha256":"7bbdac52e5c0d29ba8f51b99328bfec1665618a64bb0eea452d501179fd59289","name":"served-return-2083.json","bytes":12664},{"sha256":"525620f7d0be8770f81864f67afe2fbfeb40ac39e053b076393a7b5b50f8ad91","name":"served-return-2120.json","bytes":43593},{"sha256":"aa3ec8f0678ae8b458b15d0d1c2b84da155662aec6eb1a39a0137919faa3ef3e","name":"served-return-2126.json","bytes":47416},{"sha256":"40c01652c4a95e10c993c53f4e5937b84b6a50e78085fc19c8ef8ace8951f0cc","name":"served-return-2134.json","bytes":34779},{"sha256":"10875877e6246ac036f6dddf68a7fdfd825da58b82eb018712ae66986691aa6b","name":"served-return-2136.json","bytes":33173},{"sha256":"367a784ef0e8db22442ac09b61530e4babc82b6f1c4b9a60cab06617e2cc5c99","name":"served-return-2140.json","bytes":29898},{"sha256":"e3ca7a15d23c7224c95cf5fab73c8854ca01365f29c9bed5235849ae74a83fce","name":"served-return-2143.json","bytes":25442},{"sha256":"7f94864f2c688d290a080e88d3764b5d2060ba09d66d8b98404ed5ebdd5f49cf","name":"served-return-2146.json","bytes":13097},{"sha256":"a708ccb7526ff17c2641f9a8245e4fd2bd552006270debab4260af9a6bcde60e","name":"served-wret-2329.json","bytes":34850},{"sha256":"9a7bbf0eb93d882948acdcc9e41936106ec364797ca06a6031ac955d97068e4c","name":"served-wret-2256.json","bytes":15698},{"sha256":"709f9a0a1c93f3ba04308cbfb4e2e47c06e0599e3dad8459b58567e504ca1e5d","name":"served-wret-2237.json","bytes":15282},{"sha256":"69e222125def0f3dc78be29a122fde24b8c84f41cbc358a72c5806e818d6383a","name":"served-wret-2275.json","bytes":30987},{"sha256":"25c0f855dbf338e75cd62b0c4d3d9b80ee3a1cdf574b7c51756066b412dad992","name":"served-wret-2313.json","bytes":22315},{"sha256":"d6f35049e3cc01ed8c1cbbed306dea077046bd0969b63b95e7974ea5c22b25cf","name":"served-wret-2327.json","bytes":26916},{"sha256":"a3e0737245bfe67a62e29a1047dfe4e96960dd70043a049fc5fa973089457a60","name":"QUESTIONS.md","bytes":620428},{"sha256":"de3354dd6254d5bf1814b8618874a9040ac515ac17f1e3d0030041abc2c1431f","name":"QUESTIONS.md","bytes":627860},{"sha256":"67e8f4a914126bd048504b5ee9a773d8676f149e421e47021eb60a53f743faf7","name":"served-fresh-E.md","bytes":629999},{"sha256":"029efc05e4b791b297f3cb254a24887e3d23b98b1ab4a6639d1f6dc7b69cc82f","name":"export_transcript.py","bytes":10230},{"sha256":"882ffb0331330cac78cccb35504bace1f7b5907de914316b3679ff772eba9318","name":"backfill_usage.py","bytes":7776}],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"duplicates":[],"cited_messages":[]}