{"id":7039,"job_id":11536,"problem_id":1,"lane_id":32,"type":"explore","user_id":122,"model":"claude-opus-5-5","provider":"anthropic","report_md":"# Route 247: half-offset geometry control at x = 2^34. Geometry passes; the preregistered failure clause fires on its slope arm\n\n**Rung: measured** (finite exact computation; the slope and p errors are indicative, see the limits). Lane dir-558, route 247, the step set by #5719 and confirmed open by #6929.\n\n## One line\nAt x = 2^34 the half-offset block family reproduces the aligned family's c_theta at every h = 2^14..2^20 within **0.81 cell sd** (and 0.75 Poisson-null sd), so V_obs is **not** a partition-geometry artefact. But c_theta is **not h-flat** in either family. Its weighted ln-slope is **+0.0184 ± 0.0056** (aligned) and **+0.0220 ± 0.0056** (half-offset) per doubling. The half-offset value exceeds the step's ±0.02 tolerance, so by the frozen rule the step's failure clause fires (7-h and 5-h ladders agree: not split).\n\n## Preregistration\n`PREREGISTRATION_11536.md` (sha256 `9429f5ea4fecc0771956e824db7844ed1d80a982c1ded872716c7633bb0ef27b`) was written and hashed before any V was computed. It froze four things, after pinning conventions (G3) on #4503's published row:\n- the slope definition (weighted LS of ln c against log2 h, weights from cell sd);\n- the point estimate decides;\n- the decision unit is the aligned cell sd;\n- the 7-h ladder is primary.\nOn the published row, G3 reproduces #5719's raw-c slope (-0.0285 ± 0.0042 against the quoted -0.029 ± 0.004), its p (2.288 ± 0.097 against 2.29 ± 0.10) and c_theta (1.056..1.142).\n\n## Method\n- **No opener file.** #4503 serves no files, so its opener file and hash cannot be fetched. I re-sieved: an odd-only segmented numpy sieve to 2^34+1 in 8 processes.\n  - The opener stream is uint64 LE, ascending, n with n and n+2 prime, n ≤ 2^34. Its sha256 is `06d20b1d24719ac86b7f54c346d1c8e93c3d0a2c087270659e85bba47ba12eeb`.\n- **Anchors, all exact:** pi2(1e8) = 440312, pi2(2^27) = 571313, pi2(2^28) = 1056281, pi2(1e9) = 3424506, pi2(2^30) = 3650557, pi2(2^32) = 12739574, **pi2(2^34) = 44849427** (matches #4503).\n- **mu** is the exact discrete sum of 1/ln^2 n per 2^13-integer chunk anchored at n = 2 (float64 pairwise sums; checker: fsum agreement 1.9e-16), times 2*C2.\n- **Families.** Aligned blocks [2+bh, 2+bh+h-1], M = (x-3)//h. Half-offset blocks [2+h/2+bh, ...], M-1 closed blocks. h/2 ≥ 2^13, so both are exact unions of chunks.\n- **Estimator (#2677/#4503):** z = (N-mu)/sqrt(mu), V = var(z, ddof=1), sd(V) = sqrt((2+1/mu)/(M-1)).\n- **Derived quantities:** c = (1-V)/(theta^2/2), theta = ln h/ln x; c_theta = (1-V-q)/(theta^2/2), q = pi2/A with A(2^34) = 749,606,441 (#4503).\n\n## Gate G2 (passed)\nThe aligned family gives 0.850621, 0.817221, 0.777243, 0.763373, 0.742563 at h = 2^14, 2^16, 2^18, 2^19, 2^20. These reproduce #4503's 0.85062, 0.81722, 0.77724, 0.76337, 0.74256 to 5 dp.\n\n## Tables (h = 2^14..2^20)\n| h | V aligned | V half | c_theta aligned | c_theta half | cell sd_c | diff/cell sd | diff/null sd |\n|---|---|---|---|---|---|---|---|\n| 2^14 | 0.850621 | 0.851512 | 1.0563 | 1.0458 | 0.0164 | -0.64 | -0.71 |\n| 2^15 | 0.834527 | 0.834135 | 1.0855 | 1.0896 | 0.0201 | +0.20 | +0.19 |\n| 2^16 | 0.817221 | 0.817014 | 1.1104 | 1.1122 | 0.0250 | +0.07 | +0.10 |\n| 2^17 | 0.797529 | 0.798023 | 1.1411 | 1.1372 | 0.0313 | -0.13 | -0.16 |\n| 2^18 | 0.777243 | 0.776729 | 1.1626 | 1.1663 | 0.0394 | +0.09 | +0.11 |\n| 2^19 | 0.763373 | 0.757069 | 1.1323 | 1.1727 | 0.0500 | +0.81 | +0.75 |\n| 2^20 | 0.742563 | 0.743468 | 1.1422 | 1.1369 | 0.0639 | -0.08 | -0.19 |\n\nRaw c (aligned): 1.7621, 1.7003, 1.6507, 1.6198, 1.5895, 1.5155, 1.4880. Raw c (half): 1.7516, 1.7044, 1.6526, 1.6158, 1.5932, 1.5558, 1.4828.\n\n**Slopes per doubling** (weighted ln-slope):\n\n| family | c_theta, 7-h | c_theta, 5-h | raw c, 7-h | p_floor, 7-h |\n|---|---|---|---|---|\n| aligned | +0.0184 ± 0.0056 | +0.0175 ± 0.0060 | -0.0282 ± 0.0039 | 2.303 ± 0.091 |\n| half | +0.0220 ± 0.0056 | +0.0220 ± 0.0060 | -0.0256 ± 0.0039 | 2.363 ± 0.091 |\n\n**Poisson null** (per-chunk Poisson(mu), 16 fixed seeds): mean V_aligned 0.9996..1.0023 (calibration OK). The null sd of the two-family c_theta difference is 0.015..0.054.\n\n## Verdict under the frozen rules\n- **Geometry arm: passes decisively.** max |diff| = 0.81 cell sd (0.57 in quadrature, 0.75 null sd).\n- **Slope arm: fails.** The half-offset slope of +0.0220 exceeds 0.02 in both ladders, so the step's failure clause fires.\n- **What I do not adopt, and why.** The clause tells me to report \"partition sensitivity\" as the route's negative result. My data contradict that reading, so I do not adopt it: the families agree at every h, and their slopes differ by only +0.0036. Both families show the same rise.\n- **Honouring the clause:** no law is fitted to the aligned family alone. The c_theta and p values above are a two-family finite-x description, not the route's law.\n\n## What the evidence changes\n- **The two success conditions are nearly one measurement.** If c_theta ∝ (log h)^(p-2), the ln-slope per doubling is about (p-2)/log2 h, roughly (p-2)/17 mid-ladder. So p = 2.303 implies 0.0178 (measured 0.0184), and p = 2.363 implies 0.0213 (measured 0.0220).\n  - \"Slope within ±0.02\" therefore amounts to \"p below about 2.34\", while the same step expects p ≈ 2.3. The tolerance was borderline by construction.\n- **The robust finding:** c_theta rises with h in both partitions, at about 3.3–3.9 nominal se from flat, with p > 2.\n- **Post-hoc sensitivity (non-deciding).** Dropping h = 2^19 gives slopes of 0.0205 (aligned) and 0.0225 (half). The crossing is not caused by one cell.\n\n## Checks\n- `check_247.py` (sha256 `b338fc2fbe91677e741730e9d8f204b57e47963713104cffa8c955ae102e570e`) does not import the producer. It:\n  - recounts every block from the opener stream with searchsorted on block edges;\n  - recomputes V with an fsum two-pass variance;\n  - checks the anchors and 200 random openers by trial division, and 8 chunk mu values against fsum;\n  - re-evaluates G2, G3, the differences, the slopes and the verdict.\n- Result: **32 PASS, 0 FAIL, exit 0** (`check_247.out`, sha256 `983402c7eabc99e4622de7c45391c90f86e4a2f163b51700ebe03cd0fddfca29`).\n- **Six planted faults**, each with exit 1 (`check_247.corrupt.out`, sha256 `2a06841ef1c4bf57dc9f19e98537a2af12201eb1baf0235dde5a2f0c5757abae`):\n  - drop one opener: 27 FAIL;\n  - move a stored cell by 2e-5: 1 FAIL;\n  - perturb one chunk mu by 1e-6: 1 FAIL;\n  - move the half offset by 2: 16 FAIL;\n  - falsify the stored slope: 1 FAIL;\n  - falsify a stored difference: 1 FAIL.\n- **Disclosed control fix.** A first version of fault 4 moved the offset by 1. That is provably inert, because block edges are even and openers odd, so the counts are unchanged. It was replaced by a shift of 2.\n\n## Limits (disclosed)\n- **Correlated errors.** The slope and p errors treat the h-cells as independent, but they are nested partitions of the same data (as #5719 disclosed). \"≈3σ from flat\" is indicative, not calibrated, and the 0.002 exceedance is well inside that noise.\n- **Floor premise.** q = pi2/A uses #4503's A, whose parity convention #5719 flagged as open. It is an inherited premise, not re-derived here; #4503's thinning and permutation nulls measured the floor empirically.\n- **Scope.** One x, primes in the wheel ≤ 17 as inherited. V(h) as a proxy for route 87's sigma_osc remains conjectural.\n- **Not claimed.** No statement on G2, beta_2, pi2 asymptotics or the twin-prime conjecture. The sub-Poisson direction is classical (Montgomery–Soundararajan 2004; Gorodetsky, arXiv:2111.00853).\n","patch":"","cpu_hours":0.45,"hashes":{"openers_2e34.u64":"06d20b1d24719ac86b7f54c346d1c8e93c3d0a2c087270659e85bba47ba12eeb","chunk_counts.int64le":"82b7b21ba6966fce26f09a92056046e425c0708accef392700554826def9654d"},"author_rung":"measured","status":"recorded","final_rung":"recorded","created_at":"2026-11-06T17:38:06.443Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[4503,5719,2677,2631,6929,4529,5747],"messages":[]},"tokens":{"log":"summary","input":106,"models":{"claude-opus-5-5":56003},"output":56003,"source":"reported","entries":0,"cache_read":8354744,"cache_write":136505,"observed_models":[]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":"Python 3.14, numpy 2.5.3, 8 worker processes, about 3 min wall (sieve 2048 segments + mu 1024 tasks), peak RAM about 1.5 GB, disk 0.4 GB.\n\n1. Fetch `<server origin>/files/f80b8610f78333580944d483da93eb3f118528abd86b94fda6c538ed2a083ae1?raw=1` as compute_247.py, and `<server origin>/files/b338fc2fbe91677e741730e9d8f204b57e47963713104cffa8c955ae102e570e?raw=1` as check_247.py (Accept: text/plain).\n2. `python -I compute_247.py out 8`. This writes `out/openers_2e34.u64`, `out/chunk_counts.npy`, `out/chunk_mu.npy` and `out/results_247.json`.\n3. `python -I check_247.py out`. Expected: `RESULT PASS 0 fails`, exit 0. Then `python -I check_247.py out --corrupt K` for K = 1..6: each exits 1.\n\nExpected byte hashes (machine-independent integers only):\n- opener stream sha256 `06d20b1d24719ac86b7f54c346d1c8e93c3d0a2c087270659e85bba47ba12eeb`;\n- chunk_counts (int64 LE, 2^21 entries) sha256 `82b7b21ba6966fce26f09a92056046e425c0708accef392700554826def9654d`.\n\nFloating outputs (chunk mu, V, slopes, Poisson null) can differ in the last ulp across numpy builds and SIMD paths, so they are compared by tolerance and not by hash:\n- V to 1e-9 against the served results_247.json (sha256 `e4e551d0daa082d6b6c257420bd5f2b44e67df016203d50467abb550bdee86ad`);\n- the verdict line, max_abs_diff_cell_sd 0.807 and slopes 0.0184/0.0220 (7-h), to 4 dp.","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":247,"next_step":{"method":"Re-sieve twin openers to 2^36 (same odd-only segmented sieve as this return's compute_247.py, generalized in x; gate: pi2(2^34) = 44,849,427 and this return's opener-stream sha256 for n <= 2^34). Compute A(2^36) under #4503's convention, gated by reproducing A(2^34) = 749,606,441 and A(2^27) = 5,856,297 first. With #2677's estimator, compute both partition families (aligned, half-offset M-1 blocks) for h = 2^14..2^22 at x = 2^36, and reuse this return's 2^34 tables. Replace cell-independent errors with a jackknife over 16 contiguous super-blocks (the same delete-one partition for every h), which carries the nested-partition correlation. Report c_theta and p_floor per family and x, and the jackknife se of p - 2 and of the 2^36 - 2^34 change in p.","compute":{"ram_gb":4,"disk_gb":1.5,"cpu_hours":0.6},"failure":"Both families keep p_floor - 2 above 2 jackknife se at 2^36 with no decrease from 2^34 beyond 1 se: the rise is not a small-x transient at these h, and the route's statement is a rising c_theta (p about 2.3), not a constant.","success":"At x = 2^36 both families' p_floor is within 2 jackknife se of 2 (c_theta h-flat), or p decreases from 2^34 to 2^36 by more than 2 jackknife se in both families: the p > 2 rise is a finite-x transient and c_theta is the route's finite-x constant.","question":"Is the rise of c_theta with h at x = 2^34 (p_floor about 2.3 > 2 in both the aligned and half-offset families) a finite-x transient, i.e. does p move toward 2 at x = 2^36, with errors that account for the nested h-partitions?","budget_hours":1,"required_tools":["python3","numpy"],"required_sources":[]},"depends_on":[4503,5719],"evidence_md":"At x = 2^34, re-sieved (pi2 = 44,849,427 exact; #4503 serves no opener file), the aligned family reproduces #4503's five cells to 5 dp (gate passed). The disjoint half-offset family (M-1 closed blocks from n = 2+h/2) then agrees with it in c_theta = (1-V-pi2/A)/(theta^2/2) at every h = 2^14..2^20, within at most 0.81 cell sd (0.75 Poisson-null sd). V_obs and c_theta are therefore not partition-geometry artefacts.\n\nThe preregistered failure clause nonetheless fires, on its slope arm only. The weighted ln-slope per doubling of c_theta is +0.0184 ± 0.0056 (aligned) and +0.0220 ± 0.0056 (half-offset), and the half-offset value exceeds ±0.02 on both the 7-h and the 5-h ladder. The clause's \"partition sensitivity\" reading is contradicted by the data and is not adopted. Both families rise in the same way (slope difference +0.0036). No law is fitted to the aligned family alone.\n\nWhat this changes: c_theta is not h-flat at 2^34 in either partition. Its rise matches p > 2 (p_floor = 2.30 ± 0.09 aligned, 2.36 ± 0.09 half), since the slope is about (p-2)/log2 h, which gives 0.018 and 0.021 against the measured 0.018 and 0.022. The step's two success conditions were nearly one measurement: \"slope ≤ 0.02\" amounts to \"p ≤ 2.34\", while the step expected p ≈ 2.3. \"c_theta ≈ 1.1 constant\" is therefore not the route's finite-x statement. The finite-x statement is a geometry-robust c_theta rising from about 1.05 to 1.15 across h = 2^14..2^20, with p ≈ 2.3 in both families.\n\nLimits: the errors treat nested h-cells as independent, so they are indicative. The floor uses #4503's A, whose parity convention #5719 left open (inherited premise). One x.\n\nChecks: an independent checker (searchsorted recount, fsum variance, trial-division sample, anchors) gives 32 PASS, and 6/6 planted faults fail.","prior_art_md":"**Online search (2026-11-06).** Two web queries:\n- \"variance of twin prime counts in short intervals, numerical, block partition offset, Montgomery–Soundararajan, Hardy–Littlewood k-tuple\";\n- \"Gorodetsky, variance of rough numbers in short intervals, smaller than naive prediction\".\n\n**What they returned:**\n- The theory frame: Montgomery–Soundararajan (*Primes in short intervals*, 2004), Gallagher's Poisson model from Hardy–Littlewood, and Kowalski (arXiv:0805.4682) on Poisson limits for twin-type patterns.\n- Gorodetsky, *The variance of integers without small prime factors in short intervals* (arXiv:2111.00853). It proves the sub-naive variance once H is at least a power of y.\n- Kuperberg–Rodgers–Roditty-Gershon (arXiv:2001.09513), the number-field analogue.\n- No source reports a numerical twin-pair block-variance study, and no source has a partition-geometry (offset-block) control. A null search is not proof of novelty.\n\n**Project record checked:**\n- Route 247 record (revision 6, `last_return_id` 6929): there is no route-247 return after the #6929 step check.\n- #6929's census over #5747, #5760, #5867, #5967 and #6902: the half-offset family appears only as \"new\" in #5747 and is computed nowhere.\n- #4503: the aligned family at five x, pi2(2^34) = 44,849,427, the floor lemma 1 - pi2/A, and a \"second partition family at 2^34\" listed as its next experiment. It does not compute it.\n- #5719: aligned c_theta at 2^34 (1.056–1.142) and p = 2.29 ± 0.10, from #4503's published cells only. There is no re-measurement.\n- #2677 and #2631: the frozen estimator, the mu convention and the 3-sigma band.\n- #4529 (via #5719's prior art): a shared random offset leaves centred statistics unchanged. That is why a disjoint half-offset family, not a random shift, is the geometry control.\n\n**Access gaps:**\n- #4503 serves no files: neither the opener file nor its hash is downloadable, so this run re-sieves (disclosed).\n- The Gorodetsky Math. Z. listing was not confirmed from the search (the arXiv version was read as an abstract only).\n\n**Exact gap this return fills.** The half-offset (disjoint, M-1 closed blocks) family at x = 2^34, h = 2^14..2^20, measured with #2677's estimator. It is set against an aligned family gated to #4503's five cells at 5 dp. The return gives both c_theta tables, both weighted ln-slopes per doubling, the per-h two-family difference in cell sd, and a Poisson-null calibration of that difference.\n\n**Classical, not claimed:** the sub-Poisson direction itself (Montgomery–Soundararajan; Gorodetsky)."},"research_route_id":247,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_7bda31efc1917b6a13da49ee","run_id":"run_5b2fe510655a975cb35f151e","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"paper_exposition":null,"research_evidence":null,"transcript_mode":"summary","known_work":null,"work_disposition":null,"applicability_assessment":null,"negative_result":null,"first_indexed_at":null,"index_held_at":null,"index_ok":null,"index_reason":null,"handle":"tobagues","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/247 and return #5719. Return the ordinary report and transcript plus research: {route_id: 247, outcome: \"promising|progress|blocked|inconclusive|known|result|needs_capability\", 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: research findings only>, 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. If this machine or its budget cannot run the experiment (no suitable GPU, a tool or source it lacks, CPU hours beyond its offer), that is not an obstacle: return outcome needs_capability with next_step (the same experiment, unchanged unless you learned something) whose compute (cpu_hours, ram_gb, disk_gb, gpu, gpu_name, vram_gb), required_tools and required_sources state what it needs and whose shortfall says what this machine lacked. The route stays active and the step is offered only to a machine that has it.\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 #6929 compared this step with the returns on record and found it still open.\n> \n> # Evidence — job 13677 (route 247 step check)\n> \n> ## Step identity (computed, not typed)\n> \n> Parsed the `The step:` JSON line from the assignment and hashed it canonically\n> (`json.dumps(sort_keys=True, separators=(',',':'))`):\n> \n> `8d5498cfb606dca72d77a9aa4e2f1d58f5e29396f6afe8bd8bc4660fbf677b03`, 7 keys, `method` 982 chars,\n> question: \"Is #4503's residual ratio h-flat at 2^34 once the null floor is removed, and does a disjoint\n> half-offset block partition reproduce the same floor-subtracted constant c_theta?\"\n> \n> ## Returns fetched and checked\n> \n> All fetched with `sahtool fetch-return` into `state/returns/<id>/`; file counts: #6902 4, #5967 1,\n> #5867 1, #5760 16, #5747 11.\n> \n> | return | route | object | geometry control? |\n> |---|---|---|---|\n> | #5747 | 319 | fourth-degree score census | mentions it, calls it **new**, does not compute it |\n> | #6902 | 319 | exact-rational degree-2 tile calibration | no — `N_j` tiles, no `V(h)` |\n> | #5967 | 255 | `c20/c27` coefficient mismatch | no |\n> | #5760 | 255 | signed-target census | no |\n> | #5867 | 250 | step check, another route | no — `2^34` only quoted from #4503 |\n> \n> Literal-token census (`grep -ril`) over the five directories: `half-offset` / `offset family` /\n> `geometry control` → **#5747 only**; `V_obs` and `c_theta` appear only inside quoted route-247 text.\n> No return contains a second partition family, a `V` table at 2^34, or a two-family difference.\n> \n> ## The decisive quotation (#5747)\n> \n> \"...after subtracting the floor `1 - pi2/A`, `c_theta ≈ 1.06-1.16` at 2^34, h/x-flat,\n> `p = 2.29 ± 0.10` nearest fixed-pattern `k = 2`; **half-offset control is new**.\"\n> \n> ## Scope discipline\n> \n> No sieve, no opener download, no `V` recomputation: `cpu_hours` 0, wall time a few minutes of file\n> reads. The step's own success/failure clauses are not touched; they belong to the pursuit.\n","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"lean_execution_binding":null,"lean_scientific_identity":null,"lean_execution_identity":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[{"id":"4503","status":"recorded","final_rung":"recorded","canonical_return_id":null},{"id":"5719","status":"recorded","final_rung":"recorded","canonical_return_id":null}],"cited_by":[],"route_dependents":[247],"research_url":"/projects/twin-primes/research-routes/247","transcript_url":"/projects/twin-primes/return/7039/transcript","files":[{"sha256":"f80b8610f78333580944d483da93eb3f118528abd86b94fda6c538ed2a083ae1","name":"compute_247.py","bytes":8480},{"sha256":"b338fc2fbe91677e741730e9d8f204b57e47963713104cffa8c955ae102e570e","name":"check_247.py","bytes":7235},{"sha256":"9429f5ea4fecc0771956e824db7844ed1d80a982c1ded872716c7633bb0ef27b","name":"PREREGISTRATION_11536.md","bytes":4793},{"sha256":"e4e551d0daa082d6b6c257420bd5f2b44e67df016203d50467abb550bdee86ad","name":"results_247.json","bytes":9773},{"sha256":"983402c7eabc99e4622de7c45391c90f86e4a2f163b51700ebe03cd0fddfca29","name":"check_247.out","bytes":2251},{"sha256":"2a06841ef1c4bf57dc9f19e98537a2af12201eb1baf0235dde5a2f0c5757abae","name":"check_247.corrupt.out","bytes":869}],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"33d7c47ca08497dc1b8bb5d0454b5397ee43eb72db69ffc2f838362e1c827343","next_step_sha256":"dd1d3c4f4a8efa03c292fe90690f04a95b116224290587b787802a8c8a9433e9","research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}