Mister-Meeseeks · Oct 13, 2026 · Oct 13, 2026
diff --git a/research/jp8-clean-work/CLAIM.json b/research/jp8-clean-work/CLAIM.json
@@ -0,0 +1,49 @@
+{
+  "track": "complex supplier",
+  "assignment_seed": 33,
+  "seed_source": "Python secrets.randbelow(100), operating-system randomness",
+  "base_commit": "7ee02b7c96d85f22b38d28bd7ed49bb8c1aeb261",
+  "selected_at_base": {
+    "source_pr": 741,
+    "source_head": "e516b1463606840a15f80aac1995d64ce0092e89",
+    "kappa": "1661202643272239/1000000000000000000"
+  },
+  "coordination": "No pinned issue exists (GraphQL checked immediately before this local claim); no reservation issue opened.",
+  "claimed_at_utc": "2026-10-13T06:39:00Z",
+  "expected_finish_utc": "2026-10-13T07:25:00Z",
+  "time_budget": "50 minutes from 06:35 UTC; orientation at most 10 minutes; final 12.5 minutes reserved for verification and handoff.",
+  "candidates": [
+    {
+      "task": "Retiming one fixed-order historical JP8 frame run",
+      "decision": "Not selected: overlaps the exact fixed-order negative results in PR776 and PR815."
+    },
+    {
+      "task": "A restored helper or target basis for the historical CK compensation panel",
+      "decision": "Not selected: PR807 and PR816 already investigate these mechanisms."
+    },
+    {
+      "task": "Specialize PR696 unit-coefficient lowering when private work is already proved zero",
+      "decision": "Selected. Seed 33 chooses JP8 (odd seed) and indexes an omission witness. Preserve zero work at every addition boundary; forget the invariant at every non-scalar event and frame change."
+    }
+  ],
+  "object": "Both complete scalar orientations of the frozen PR626 JP8 provider, building on PR696's copy/unit lowering",
+  "hypothesis": "Same-frame straight-line propagation proves some unit-macro work clears redundant, without changing a scalar macro, a recursive boundary, or its private-work postcondition.",
+  "domain": "All events in the two pinned JP8 local programs; no randomized search or expanded closed search.",
+  "stopping_rule": "One deterministic exhaustive pass per orientation; complete exact replay and controls, or report a concrete obstruction. No frontier search.",
+  "optimistic_budget": "Preliminary census: 4,614 eligible unit clears per real orientation, hence 9,228 real-component primitives saved beyond PR696 in each complex orientation. No extra source reads, copies, helpers, frames, restorations or children. Recompute the complete local invoice, retaining the adapter reserve.",
+  "deliverable": "Executable state-aware lowering, exact whole-program certificate and dirty-work/barrier controls, cost/proof note, source pins and attribution; no new kappa claim.",
+  "status": "completed",
+  "completed_at_utc": "2026-10-13T06:53:46+00:00",
+  "result": {
+    "classification": "finite circuit improvement; no new kappa",
+    "extra_real_primitives_saved_per_orientation": 9228,
+    "new_forward_real_primitives": 191152,
+    "new_backward_real_primitives": 191168,
+    "all_columns": "716 native + 9 arbitrary private",
+    "negative_controls": 15,
+    "package_tests": 4,
+    "result_link": "https://github.com/faw01/integer-mult-bounds/blob/research/jp8-clean-work-seed33/research/jp8-clean-work/README.md",
+    "receipt": "EXPECTED.json",
+    "validation": "VALIDATION.md"
+  }
+}
diff --git a/research/jp8-clean-work/EXPECTED.json b/research/jp8-clean-work/EXPECTED.json
@@ -0,0 +1,140 @@
+{
+  "controls": {
+    "coefficient_crosschecks": 574,
+    "positive_controls": [
+      "same_frame_nonwork_macros",
+      "both_unit_signs",
+      "all_opaque_boundaries"
+    ],
+    "primitive_crosschecks": 13728,
+    "rejected": [
+      "dirty_first_unit",
+      "dirty_after_adapter",
+      "dirty_after_center",
+      "dirty_after_climb",
+      "dirty_after_phase",
+      "dirty_after_residual_adapter",
+      "dirty_after_frame_change",
+      "private_source_alias",
+      "private_target_alias",
+      "private_copy_alias",
+      "unknown_scalar",
+      "unknown_event",
+      "wrong_unit_sign",
+      "restored_endpoint_excess_prefix",
+      "nonunit_cleanup_omitted"
+    ]
+  },
+  "local_invoice": {
+    "adapter_primitive_reserve": 11334342205440,
+    "centers": 8,
+    "compiler": 25934336,
+    "helpers": 492,
+    "local_adapters": 437040,
+    "m": 73,
+    "maximum_scalar_macros": 22693,
+    "native_columns": 716,
+    "new_local_primitive_bound": 11334342396608,
+    "original_local_primitive_bound": 11334342462720,
+    "paid_children": 3606,
+    "pr696_local_primitive_bound": 11334342405836,
+    "scope": "Finite JP8 local invoice only; no regenerated global binding or exponent claim."
+  },
+  "programs": {
+    "backward": {
+      "actual_prefix_l1_peak": "1469388997/2",
+      "arbitrary_private_columns": 9,
+      "charged_payload_bits": 192,
+      "emitted_stream_sha256": "22b4b6e7538ff4df4c40102d502415ebbf94d3aef9ae692233f5a752cd4bf47c",
+      "events": 27909,
+      "extra_saved_real_primitives": 9228,
+      "extra_savings_by_phase": {
+        "A": 2596,
+        "B": 186,
+        "cleanup_tail": 78,
+        "full_frame_cleanup": 5968,
+        "zero_frame_precondition": 400
+      },
+      "logical_columns": 716,
+      "new_real_primitives": 191168,
+      "omission_witness_sha256": "3f1ead207ea60e2990b9d927ccd0526e12b65e31e909f522f0968e1dea39c729",
+      "original_real_primitives": 257280,
+      "paid_profile": {
+        "1": 2512,
+        "2": 400,
+        "3": 448,
+        "4": 160,
+        "5": 70,
+        "6": 8,
+        "8": 8
+      },
+      "pr696_real_primitives": 200396,
+      "primitive_tags": {
+        "add": 29176,
+        "clear": 35242,
+        "copy": 31992,
+        "double": 32576,
+        "halve": 39438,
+        "negate": 692,
+        "subtract": 22052
+      },
+      "removed_clears_per_real_component": 4614,
+      "retained_positive_prefix_l1": "16753412757380464429924108070201913461/8388608",
+      "retained_prefix_bits": 101,
+      "saved_vs_original": 66112,
+      "scalar_macros": 22693,
+      "seed33_omission_sample": [
+        102,
+        101
+      ]
+    },
+    "forward": {
+      "actual_prefix_l1_peak": "503316465",
+      "arbitrary_private_columns": 9,
+      "charged_payload_bits": 192,
+      "emitted_stream_sha256": "fda999850d25ebfb368e892eab82e3219496594b468cad93dc2a27716e369d4f",
+      "events": 27909,
+      "extra_saved_real_primitives": 9228,
+      "extra_savings_by_phase": {
+        "A": 2596,
+        "B": 186,
+        "cleanup_tail": 78,
+        "full_frame_cleanup": 5968,
+        "zero_frame_precondition": 400
+      },
+      "logical_columns": 716,
+      "new_real_primitives": 191152,
+      "omission_witness_sha256": "3f1ead207ea60e2990b9d927ccd0526e12b65e31e909f522f0968e1dea39c729",
+      "original_real_primitives": 257232,
+      "paid_profile": {
+        "1": 2512,
+        "2": 400,
+        "3": 448,
+        "4": 160,
+        "5": 70,
+        "6": 8,
+        "8": 8
+      },
+      "pr696_real_primitives": 200380,
+      "primitive_tags": {
+        "add": 29050,
+        "clear": 35226,
+        "copy": 32008,
+        "double": 32576,
+        "halve": 39438,
+        "negate": 692,
+        "subtract": 22162
+      },
+      "removed_clears_per_real_component": 4614,
+      "retained_positive_prefix_l1": "4536094588619580433782972610869617/281474976710656",
+      "retained_prefix_bits": 64,
+      "saved_vs_original": 66080,
+      "scalar_macros": 22693,
+      "seed33_omission_sample": [
+        102,
+        101
+      ]
+    }
+  },
+  "status": "PASS_JP8_CLEAN_WORK"
+}
diff --git a/research/jp8-clean-work/OVERLAP.json b/research/jp8-clean-work/OVERLAP.json
@@ -0,0 +1,132 @@
+{
+  "checked_at_utc": "2026-10-13T06:47:32Z",
+  "open_pr_count": 630,
+  "pinned_sprint_issue": null,
+  "scope": "Live PR metadata, relevant ledger/negative-results review, plus source inspection of PR755. Not an independent audit of these other PRs.",
+  "related_work": [
+    {
+      "pr": 696,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/696",
+      "observed_head": "5a3f983715027c4fd83f50bc9001d4b18463bed1",
+      "title": "Research: lower selected complex scalar primitives by 22\u201325% (no new kappa)",
+      "author": "thakrarsagar",
+      "relationship": "Direct code dependency: copy/unit macro lowering. This adds a conservative known-zero state precondition, preserving the same addition postcondition.",
+      "consumed_head": "5a3f983715027c4fd83f50bc9001d4b18463bed1"
+    },
+    {
+      "pr": 755,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/755",
+      "observed_head": "a1dd9f819102c40e0b5e4a75215cf69ee864f480",
+      "title": "Research: JP8 coefficient reuse across all scalar phases (no new kappa)",
+      "author": "taizaki69",
+      "relationship": "Source inspected at the pinned head: reuse_key returns None for unit coefficients; lower_scalar retains their work clear. Shares nonunit coefficient values, unlike this zero-state specialization.",
+      "inspected_file": "research/complex-coefficient-reuse/lower.py",
+      "inspected_file_sha256": "29d403338c7901b6cb9996f11c3499d6b0a7240a3a0cb05d7905731cf833bb56"
+    },
+    {
+      "pr": 722,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/722",
+      "observed_head": "a0766ab0fab379e1997fadc3ca157578d57986c1",
+      "title": "Research: share JP8 cleanup coefficient work, saving 19,688 primitives (no new kappa)",
+      "author": "ellipsi2",
+      "relationship": "Same-coefficient cleanup fanout batches; changes private clearing boundaries. No dependency or additive composition asserted."
+    },
+    {
+      "pr": 739,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/739",
+      "observed_head": "abb9825202dba6f9bc95c7df8d12d86088fdff1a",
+      "title": "Research construction: rescale JP8 coefficient work across magnitudes (no new kappa)",
+      "author": "namredips",
+      "relationship": "Power-of-two coefficient rescaling across cleanup runs; no dependency or additive composition asserted."
+    },
+    {
+      "pr": 787,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/787",
+      "observed_head": "93ceb65cfbf339d7ca743614d4756a3af1440cd4",
+      "title": "Research: compose complex coefficient lowering and cleanup sharing",
+      "author": "bitplane",
+      "relationship": "Composition of copy/unit, signed digits and cleanup sharing; no dependency or additive composition asserted."
+    },
+    {
+      "pr": 793,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/793",
+      "observed_head": "f9383f324ee86793c5ae7b73f8e39d5827a8aa52",
+      "title": "Research: batch JP8 compensation, saving 80,150 / 81,464 primitives (no new kappa)",
+      "author": "Varun-Mittal-1001",
+      "relationship": "Reorders/batches the compensation panel; this contribution does neither."
+    },
+    {
+      "pr": 811,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/811",
+      "observed_head": "b5c244c916dd0b30957332edad7ae7ae22b0eb9d",
+      "title": "Research: JP8 additive coefficient routes save 2,784 beyond rescaling (no new kappa)",
+      "author": "Ryan-Haines",
+      "relationship": "Adds dyadic source reads between coefficient preparations; this contribution deletes only a provably redundant zero clear."
+    },
+    {
+      "pr": 830,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/830",
+      "observed_head": "6a0b8d55bb8026f8b6d708ae07740453b41c872f",
+      "title": "Research: transposed JP8 work saves 32,810 backward primitives (no new kappa)",
+      "author": "Ryan-Haines",
+      "relationship": "Transposes the additive coefficient routes to backward; no dependency or additive composition asserted."
+    },
+    {
+      "pr": 807,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/807",
+      "observed_head": "b1a5c533a22c2665600102d872f6a84e0664a70d",
+      "title": "Research: restored CK helper pairs save 35,872 primitives per orientation (no new \u03ba)",
+      "author": "davidawad",
+      "relationship": "Restored CK helper basis overlaps a rejected candidate, not this JP8 state specialization."
+    },
+    {
+      "pr": 816,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/816",
+      "observed_head": "133735a2cc2319c12ba5d786214b2217e083e7da",
+      "title": "Research: signed CK target forests save 66,520 primitives with an exact class floor (no new kappa)",
+      "author": "dd-hemant",
+      "relationship": "CK signed target forests overlap a rejected candidate, not this JP8 state specialization."
+    },
+    {
+      "pr": 776,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/776",
+      "observed_head": "8bd985fe2842c948a5a4a1554b8e090f2e371afa",
+      "title": "Scoped negative result: PR626 jp8 and ck A/B frames admit no next complex tick at fixed order (exact duals, no new \u03ba)",
+      "author": "ntedvs",
+      "relationship": "Fixed-order frame search negative result; not reopened."
+    },
+    {
+      "pr": 815,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/815",
+      "observed_head": "d8c8b40ab0a95b6d77389e6726d86dddb60fb53c",
+      "title": "Research: exact frame search certifies #741 and #626 providers frame-optimal (rediscovers #741; negative result)",
+      "author": "Rohansguliani",
+      "relationship": "Fixed-order frame search/certificates; not reopened."
+    },
+    {
+      "pr": 651,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/651",
+      "observed_head": "df5b9c41fcea749242af886646dfebb9d86319a7",
+      "title": "Research: the complex delivery layer is rigid \u2014 target hosting and raw-face sharing closed exactly",
+      "author": "higherbeing",
+      "relationship": "Delivery-layer rigidity; no rule change or renewed packet search here."
+    },
+    {
+      "pr": 837,
+      "url": "https://github.com/CrocSwap/integer-mult-bounds/pull/837",
+      "observed_head": "7b856815737ba3f4d69756a718b10758bcd7089d",
+      "title": "Research: bidirectional Horner saves 312 SW4 primitives per orientation (no new kappa)",
+      "author": "bitplane",
+      "relationship": "Concurrent bidirectional Horner synthesis inside each coefficient macro, retaining its private-work boundary. This submission specializes already-zero unit work across same-frame scalar macros; no savings composition is asserted."
+    }
+  ],
+  "finding": "No matching known-zero unit-clear specialization found. Stronger scalar-saving claims in other work are different constructions; their savings are not added here.",
+  "upstream_refresh": {
+    "head": "f567884aeb8851b6e794af1184119a3bfbe1e631",
+    "selected_source_pr": 799,
+    "selected_kappa": "1661634653669567/1000000000000000000",
+    "historical_dependencies_unchanged": 9
+  },
+  "submission_refresh_at_utc": "2026-10-13T06:55:40+00:00",
+  "submission_open_pr_count": 634
+}