Research consolidation: authenticate all PR99 files retained through PR103 (no new kappa)

CONTRIBUTORS.md

@@ -613,6 +613,8 @@ Selected progression: Jeremy Kothe and Rohan Arun; T16 controller: Hao Yan. Inde
 
 The [92-PR review](docs/research/community-reconciliation92-review.md) records individual contributions, replays, limitations and retained source trees. This includes parallel improvements, formal arithmetic, reproducibility tools and negative research; credit does not imply every historical claim was accepted.
 
+**Divjot Singh Manchanda (@djsmanchanda)** supplied [#99's nondegenerate readout saturation](https://github.com/CrocSwap/integer-mult-bounds/pull/99), retained in **Rohan Arun's [#103 signed/balanced integration](https://github.com/CrocSwap/integer-mult-bounds/pull/103)**. The [source-retention receipt](research/pr99-retention-seed99/README.md) authenticates all 46 original files and reconciles the saved 177,100-singleton profile reduction. This preserves individual mechanism credit without changing historical review dispositions or claiming a fresh physical replay.
+
 - **DaysSky (@DaysSky)**: [#162](https://github.com/CrocSwap/integer-mult-bounds/pull/162).
 - **Dominik Scholz (@DominikScholz)**: [#22](https://github.com/CrocSwap/integer-mult-bounds/pull/22), [#27](https://github.com/CrocSwap/integer-mult-bounds/pull/27), [#33](https://github.com/CrocSwap/integer-mult-bounds/pull/33), [#35](https://github.com/CrocSwap/integer-mult-bounds/pull/35), [#38](https://github.com/CrocSwap/integer-mult-bounds/pull/38), [#63](https://github.com/CrocSwap/integer-mult-bounds/pull/63), [#68](https://github.com/CrocSwap/integer-mult-bounds/pull/68), [#81](https://github.com/CrocSwap/integer-mult-bounds/pull/81).
 - **Joel Pulikkan (@GamingPuzzled)**: [#125](https://github.com/CrocSwap/integer-mult-bounds/pull/125).

research/pr99-retention-seed99/README.md

@@ -0,0 +1,127 @@
+# PR99 source retention through PR103
+
+**Research consolidation; no new kappa.** All 46 files introduced by Divjot
+Singh Manchanda's PR99 survive unchanged in Rohan Arun's incorporated PR103.
+Both complete package trees are `6bebd79406bb2453ada2b7761aece13d3e53b694`.
+The offline proof authenticates paths, modes, blob identities and the complete
+original commit difference, rather than relying on a matching title or README.
+
+At base `bce54fb3f895cbe5cd4c0eb477c3afeed100190f`, the
+[ledger](../../docs/research/pr-ledger.json) gives PR99 no integration location
+and records only earlier source-inventory triage. PR103's retained README says
+the whole PR99 package was vendored unchanged. This receipt establishes that
+link and restores the missing individual credit in
+[CONTRIBUTORS.md](../../CONTRIBUTORS.md). It does not change ledger dispositions
+or convert source retention into mathematical approval. Live GitHub metadata
+at entry already reported both PRs closed; no PR was closed by this work.
+
+## Exact sources and coverage
+
+| Role | Commit and path |
+|---|---|
+| PR99 source | `0622dcce6a897567e4eb6bdfedc7def66ab21346:research/nondegenerate-readout-saturation` |
+| Original comparison parent | `56b66d58297deca1d7dd130247d720e960f77a37` |
+| PR103 retained copy | `75ab1f70d146c21c7611ce0df9eb5f645f51ac0e:research/saturated-balanced/vendor/pr99` |
+| Published source archive | `b91de239a64d24d8aee409a0f4ac3e00710ae5c1`, tag `archive/pr-reconciliation100-20261013` |
+
+[SOURCE-PROOF.json](SOURCE-PROOF.json) contains 32 raw Git objects: four commits,
+the necessary trees, and five evidence blobs. Its authenticated commit difference
+is exactly 46 additions, zero modifications and zero deletions. PR103 is a direct
+parent of the archive commit, whose identity and parent list are checked against
+the retained archive receipt. The historical PR103 review is bound to its exact
+head. Git content identity assumes collision resistance of Git SHA-1; it does
+not authenticate signatures or establish authorship. The optional collector
+also compares every original and retained file's bytes using Git directly.
+
+The standalone checker imports no predecessor implementation. It reconciles
+four authenticated saved reports: the baseline histogram, PR99's candidate and
+frame-extension list, PR99's historical component-delta review, and PR103's
+integrated histogram. The extension counts are six `12->13`, three `12->14`,
+and thirteen `13->14`: 22 frames and 25 added dimensions. With `v=1771`, their
+recorded merger accounting is
+
+```text
+3542 * [6*(H509-H507-2*H1) + 13*(H511-H509-2*H1)
+        + 3*(H511-H507-4*H1)]
+= -177100*H1 - 31878*H507 - 24794*H509 + 56672*H511.
+```
+
+Here `Hr` denotes one child of rank `r`. The independently subtracted saved
+histograms give precisely this vector. Its rank-weighted mass is zero; both
+profiles have recursive rank `57,403,754,177`, width `108,516,254` and `28,866`
+auxiliary roles. PR103's complete saved histogram equals PR99's candidate.
+The checker verifies these identities, not the legality of the original frame
+extensions or a newly executed circuit. Full physical, native, moment, all-size,
+compiler, routing, precision/recovery and analytic obligations remain inherited.
+No full multiplication theorem or new selected-frontier replay is claimed.
+
+## Reproduce
+
+From the repository root, using Python 3.12 and its standard library:
+
+```sh
+python3.12 scripts/research_package.py --package research/pr99-retention-seed99
+```
+
+This runs the read-only checker and 18 tests: one positive case and 17 failure
+controls. Expected output includes:
+
+```text
+PASS_PR99_RETENTION_AND_SAVED_ACCOUNTING files=46 singleton_delta=-177100 mass_delta=0
+Ran 18 tests
+OK
+PASS_REGISTERED_RESEARCH_PACKAGE research/pr99-retention-seed99 outcome=completed
+```
+
+Controls cover missing files and trees, same-count path substitution, file modes,
+corrupt commits and accounting blobs, missing archive ancestry, wrong review
+heads, duplicate JSON/ranks/slots, a missing double extension, changed width,
+omitted component charges, a mass-preserving wrong profile, and substitution of
+the unsaturated histogram into PR103. Explicit checks also run under `python -O`.
+
+The direct `python3 -I -O -B research/pr99-retention-seed99/verify.py` command
+needs neither Git nor network access, only this package and the two immutable
+review inputs listed in [VERIFY.json](VERIFY.json). The registered metadata
+check additionally uses Git history for the provenance in [SOURCES.json](SOURCES.json).
+[RESULT.json](RESULT.json) is compared against recomputation on every default run.
+
+To recollect the proof from locally available pinned Git objects, run
+`python3 -I -B research/pr99-retention-seed99/collect.py --output /tmp/pr99-proof.json`
+and compare it with `SOURCE-PROOF.json`. The collector documents the fetches
+needed if those objects are absent; it never fetches or runs archived code itself.
+
+## Task and overlap
+
+OS-random seed **99**, consolidation track; task `overlap-receipt`, refined to
+PR99 through PR103. The seed selects the older PR99 object. PR279 source closure
+and stale current-result routing were rejected as candidate tasks because
+PR1190 and PR1232/1285 already address them. This receipt uses a similar Git
+object authentication method to Adam Gall's PR1190, on a distinct source tree;
+no code or proof data from that pending PR is imported. Rohan Arun's PR103 and
+the maintainer reconciliation are the retention and review predecessors.
+Full overlap-search coverage and limitations are recorded in [TASK.json](TASK.json).
+The [coordination claim](https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-18912011)
+was posted before implementation. No search for a better circuit was performed.
+
+## Open next questions
+
+No further retention experiment is justified by this receipt: the complete
+original change is accounted for. This does not settle whether every original
+mathematical obligation has received sufficient review. That question remains
+outside this source-retention result; neither the ledger's historical disposition
+nor the selected frontier is changed.
+
+## Credit
+
+Divjot Singh Manchanda (@djsmanchanda) supplied PR99's nondegenerate readout
+saturation. Rohan Arun (@rohanarun) supplied PR103's signed/balanced integration;
+Zhihao Chen (@jacklightChen), Swapnil Jain and the original notices retain the
+upstream mathematical and implementation credit. Douglas Colkitt and the
+maintainer reconciliation retained the historical source and review records.
+The original [UPSTREAM-NOTICE](UPSTREAM-NOTICE) is preserved byte for byte.
+
+Prepared for Hanson Sun (@bbMisk), with substantial OpenAI Codex assistance in
+source investigation, implementation, verification and writing. These checks
+are independently implemented relative to the historical code, not independent
+human peer review. New work is Apache-2.0; predecessor licenses remain in their
+unchanged original package.

research/pr99-retention-seed99/RESULT.json

@@ -0,0 +1,38 @@
+{
+  "status": "PASS_PR99_RETENTION_AND_SAVED_ACCOUNTING",
+  "base_commit": "bce54fb3f895cbe5cd4c0eb477c3afeed100190f",
+  "original_head": "0622dcce6a897567e4eb6bdfedc7def66ab21346",
+  "retained_head": "75ab1f70d146c21c7611ce0df9eb5f645f51ac0e",
+  "archive_commit": "b91de239a64d24d8aee409a0f4ac3e00710ae5c1",
+  "common_package_tree": "6bebd79406bb2453ada2b7761aece13d3e53b694",
+  "retained_files": 46,
+  "original_changes": {
+    "added": 46,
+    "modified": 0,
+    "deleted": 0
+  },
+  "authenticated_git_objects": 32,
+  "accounting": {
+    "changed_frames": 22,
+    "dimension_increase": 25,
+    "frame_extension_counts": {
+      "12->13": 6,
+      "12->14": 3,
+      "13->14": 13
+    },
+    "histogram_delta": {
+      "1": -177100,
+      "507": -31878,
+      "509": -24794,
+      "511": 56672
+    },
+    "rank_mass_delta": 0,
+    "recursive_rank": 57403754177,
+    "width": 108516254,
+    "auxiliary_roles": 28866,
+    "pr103_saved_histogram_matches": true
+  },
+  "fresh_physical_replay": false,
+  "new_kappa_claim": false,
+  "scope": "Complete historical source retention and saved-report accounting; no new construction acceptance."
+}

research/pr99-retention-seed99/SOURCES.json

@@ -0,0 +1,42 @@
+{
+  "schema_version": 1,
+  "base_commit": "bce54fb3f895cbe5cd4c0eb477c3afeed100190f",
+  "files": [
+    {
+      "path": "certificates/selected-result.json",
+      "usage": "provenance",
+      "git_blob": "ae6a99f28aa468649eb66a361e764d769363bf11",
+      "sha256": "10b9e340fd8e354a37a0c4813c90e5de1dd42044b5f6aca4a3fc5a42b685fa32"
+    },
+    {
+      "path": "docs/research/agent-task-catalog.json",
+      "usage": "provenance",
+      "git_blob": "c05fafaa25ba95b2d5af86b1800427198e3d20eb",
+      "sha256": "89b5928bc92c81e2647b6d702a7ad85a6795b8f44e2eb267b2ff3145d8f2b2ce"
+    },
+    {
+      "path": "research/community-reconciliation92-audit/ARCHIVE.json",
+      "usage": "input",
+      "git_blob": "2aeeb50a033c9b72ca5536cc3473263d08e24248",
+      "sha256": "3de2b360ce5cee8a0dc68a322c7db81a2bd3ca714d76aed57156cc889103b890"
+    },
+    {
+      "path": "research/community-reconciliation92-audit/REVIEW.json",
+      "usage": "input",
+      "git_blob": "a398bea6a26bccf7c7f86ad3477bd5b2489d52ae",
+      "sha256": "e38ce3a9fc5b3cc788e618885b6ba5c54a2b00f272cdfdfcd6accf66a20faf8f"
+    },
+    {
+      "path": "docs/research/pr-ledger.json",
+      "usage": "provenance",
+      "git_blob": "466e1ba17030149e42fa8acefced4359b97728a2",
+      "sha256": "6063d9227279dcc91f03fab89f5d93a9cb66c0719b394ff9a03ce58239bfd454"
+    },
+    {
+      "path": "CONTRIBUTORS.md",
+      "usage": "provenance",
+      "git_blob": "dc9d4f5de07725f56d90a7a2673f82af52e2c44b",
+      "sha256": "c37caf0a5783aef24f981cb0110bd1764bc5ec9535a58bc62af7170c10b3a8ae"
+    }
+  ]
+}

research/pr99-retention-seed99/TASK.json

@@ -0,0 +1,138 @@
+{
+  "schema_version": 1,
+  "task_id": "overlap-receipt",
+  "priority_revision": 2,
+  "base_commit": "bce54fb3f895cbe5cd4c0eb477c3afeed100190f",
+  "original_card": {
+    "id": "overlap-receipt",
+    "question": "What exact content and credit of a small named PR set has already been retained?",
+    "starting_points": [
+      "docs/research/pr-ledger.md",
+      "CONTRIBUTORS.md"
+    ],
+    "methods": [
+      "Compare exact heads and retained artifacts to write a source-bound disposition receipt.",
+      "Audit an existing receipt for missing content or predecessor attribution."
+    ],
+    "acceptance_gate": "Separate complete integration, partial retention and changed heads; make no autonomous closures.",
+    "first_step": {
+      "command": [
+        "python3",
+        "-B",
+        "scripts/research_baseline.py"
+      ],
+      "expected": "PASS_RESEARCH_BASELINE_BINDINGS_ONLY",
+      "inspect": "Read one small PR set in the ledger and identify exact reviewed heads and unresolved content.",
+      "scope": "Positive control / source-bound orientation only; this is not a new construction or full cold replay."
+    },
+    "first_experiment": "Compare that set to retained artifacts; produce a complete or explicitly partial disposition receipt.",
+    "bottleneck_link": "Reliable overlap evidence preserves credit and prevents repeated review of already-retained material.",
+    "allowed_variations": [
+      "One small related PR family or one existing integration receipt."
+    ],
+    "scope_change_triggers": [
+      "Treating matching titles or one shared blob as proof of full incorporation."
+    ],
+    "overlap_terms": [
+      "overlap receipt",
+      "partial integration",
+      "attribution"
+    ]
+  },
+  "scope_status": "refined",
+  "actual_scope": "Authenticate complete PR99 source retention through incorporated PR103 and reconcile its saved rank-profile delta.",
+  "scope_change_reason": "Narrow the overlap-receipt card to one parent contribution and its retained vendor subtree; no selected-frontier change or new physical replay.",
+  "bottleneck_link": "Resolve a concrete missing retention and individual-credit link so review can reuse the exact historical package without confusing retention with proof acceptance.",
+  "budget_minutes": 45,
+  "stop_rule": "Stop after the complete 46-file identity, original commit-diff coverage, saved-profile reconciliation and failure controls are established; do not rerun the historical full construction or search for a new circuit.",
+  "outcome": "completed",
+  "evidence": "SOURCE-PROOF.json, RESULT.json, verify.py, test_verify.py; source retention and saved-report accounting only.",
+  "author": "Hanson Sun (@bbMisk)",
+  "ai_assistance": "Substantial OpenAI Codex assistance in investigation, implementation, verification and writing; no independent human review claimed.",
+  "open_questions": [],
+  "assignment_seed": 99,
+  "seed_source": "Python secrets.randbelow(100), using operating-system randomness",
+  "seed_use": "Select historical PR99 as the concrete provenance object within the assigned consolidation track.",
+  "no_open_questions_reason": "The complete original change is retained; no further source-retention experiment is justified. Mathematical acceptance and inherited physical/compiler/analytic obligations remain outside this receipt.",
+  "claim_url": "https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-18912011",
+  "candidates": [
+    {
+      "object": "PR99 through PR103",
+      "decision": "chosen: complete source retention has no ledger integration pointer and individual credit is absent from root CONTRIBUTORS; seed 99 diversifies to this older source"
+    },
+    {
+      "object": "PR279 parent source closure",
+      "decision": "deferred: PR1190 already provides complete closure; discussion1336 independently audits it"
+    },
+    {
+      "object": "historical current-result routing",
+      "decision": "deferred: PR1232 and PR1285 already cover the relevant navigation gaps"
+    }
+  ],
+  "overlap_at_entry": {
+    "searched_at": "2026-10-14T00:23:00.134424+00:00",
+    "terms": [
+      "PR99",
+      "nondegenerate readout",
+      "saturated-balanced"
+    ],
+    "coverage": {
+      "open_prs": {
+        "fetched_at": "2026-10-14T00:22:41.848563+00:00",
+        "complete_pagination": true,
+        "records": 873
+      },
+      "discussions": {
+        "fetched_at": "2026-10-14T00:23:00.090065+00:00",
+        "complete_pagination": true,
+        "records": 132
+      }
+    },
+    "local_records": 325,
+    "warnings": [],
+    "limits": "Remote PR scope is open PRs including drafts. Closed work is searched through local READMEs, TASK metadata and ledger only. No code diffs or arbitrary package files are searched. Recheck relevant heads before submission.",
+    "finding": "No same-object PR99 retention claim found. Literal PR99 matches mostly refer to PR99x. All returned leads were inspected; no absence-of-novelty proof claimed."
+  },
+  "overlap_before_submission": {
+    "searched_at": "2026-10-14T00:34:24.904990+00:00",
+    "terms": [
+      "nondegenerate readout",
+      "nondegenerate-readout-saturation",
+      "saturated-balanced",
+      "PR99 retention"
+    ],
+    "coverage": {
+      "open_prs": {
+        "fetched_at": "2026-10-14T00:34:10.047124+00:00",
+        "complete_pagination": true,
+        "records": 785
+      },
+      "discussions": {
+        "fetched_at": "2026-10-14T00:34:24.865859+00:00",
+        "complete_pagination": true,
+        "records": 144
+      }
+    },
+    "local_records": 327,
+    "warnings": [],
+    "limits": "Remote PR scope is open PRs including drafts. Closed work is searched through local READMEs, TASK metadata and ledger only. No code diffs or arbitrary package files are searched. Recheck relevant heads before submission.",
+    "finding": "The only three matches were this package README, its TASK metadata and its own coordination claim; no same-object competing contribution was observed."
+  },
+  "submission_base_commit": "410026dc5513b87db92ae98571d1d5984c69a983",
+  "verification": {
+    "python_versions": [
+      "3.12.13",
+      "3.14.6"
+    ],
+    "registered_python312_replay": "passed; read-only checker plus 18 tests (1 positive, 17 rejection controls), optimized mode",
+    "source_export": "passed with no .git directory and PATH=/nonexistent",
+    "independent_git_native_comparison": "all 46 paths/modes/blobs and the full 46-addition original commit diff agree",
+    "collector_reproduction": "SOURCE-PROOF.json byte-identical on recollection; all 46 original and retained payloads compared directly",
+    "selected_record_check": "passed; selected kappa remains 0.001672660197076918",
+    "not_run": [
+      "historical PR99 or PR103 physical/native/full construction replay",
+      "selected frontier1094 cold replay",
+      "Lean proof build"
+    ]
+  }
+}