Independent verification: the selected frontier1094 complex rank ledger, rebuilt from its emitted program (no new kappa)

[thakrarsagar]

Independent verification. No new κ, and no check is changed or weakened. Card: independent-moment (ledger method), scope refined (see TASK.json).

What this closes

The selected certificate research/community-frontier1094-audit/ASSEMBLY.json (1122cb3a…) prices an aggregate complex profile: {1:349604, 2:60192, 3:62585, 4:20448, 5:11808, 7:7, 8:1152, 48:4866}. The selected κ omits the complex surcharge. The PR552/PR559 review (vendor/pr559/PROOF.md, d5d498ed…) rests that on two premises: (a) an exact exhaustive recursive-child profile (lines 17, 24), and (b) no good/bad q-adic selector (lines 9, 24).

This package rebuilds that profile from the emitted program the certificate binds, using code-disjoint standard-library code that imports nothing from the repository:

• the hosted GLOBAL-TEMPLATE;
• both dispatched LOCAL-PROGRAMS, which must equal the copy embedded in the template;
• CANONICAL-WORD and DATA-PHASES, read from the committed capsule.

It then requires the rebuilt profile to equal the certificate.

It checks part of (a) at source level: the priced aggregate equals the emitted recursive children of the bound program. It does not address (b), and it does not re-prove the classification rule or the hosting semantics.

Checks

New in this package

• Byte binding (B1–B9). The checks cover:
  • the capsule pin and safe tar members, with tar parsing restricted to the hashed bytes;
  • certificate ↔ record (a checkpoint once the frontier moves);
  • the template and owner-audit digests;
  • the dispatched inputs and the local-replay receipt, which names the same programs and word;
  • the owner map and its manifest;
  • the literal remainder frames;
  • FULL-WORD.
• Dispatch (L1–L3, L9–L12).
  • Each (stage, replica) is called once.
  • The schedule's local sweeps dispatch exactly the calls, each sweep acts on one replica, and there is one provider.
  • Orientation and pair follow the recipe.
  • The program hashes bind the calls, and the programs equal the copy embedded in the template.
  • The 704 distinct live route families include the data routes of the active pair.
  • Every owner-map slot is executed by exactly one call route in a live replica, and each helper route equals its slot.
• A code-disjoint frame replay of the dispatched programs, bound to those calls (L4–L7).
  • Each register runs from start to final.
  • Each climb leaves the register's current frame for a strictly larger nested frame, in its phase. F2 ranks are recomputed.
  • Each centre is bound to its frame, its carrier and the centers phase.
  • Adapters, residual adapters and every scalar operand are checked at their registers' replayed frames.
  • The phase sequence and the vocabularies are closed.
• Chain to the certificate (L23). The rebuilt ledger equals ASSEMBLY.complex_profile, the owner-audit profile and INDEPENDENT1094.

Cross-checks of existing work, not claimed as new

• Against Independent audit: selected PR1094 pending-axis independence and all source-touch cuts #1209: helper widths, joint orthonormality of sibling pending axes, contiguous prefixes, and the idle-remainder recount, matched one for one with the emitted idle_children and literal frames.
• Against the in-lineage replay receipt: the per-call ledgers.
• Against Research audit: independent owner-map check of all 15 open T16 claims on the seven-slot k8 word (no new kappa) #1091 R7 and INDEPENDENT1094: the mass and deficit identity.
• Against Research note: restoration deferral is mass-neutral in the hosted T37 recurrence (no new kappa) #1012: its coverage identity, checked per family (72 units per live family).
• Helper climb totals match the FULL-WORD width vector.

Case	W	N	Rebuilt ledger	Remainders	D
frontier1094 (selected)	14336	144	{1:349604, 2:60192, 3:62585, 4:20448, 5:11808, 7:7, 8:1152, 48:4866}	7526	5120
frontier1042 (checkpoint)	14336	144	{1:349577, 2:60193, 3:62608, 4:20448, 5:11808, 7:3, 8:1152, 16:47, 48:4850}	7550	5120
slots T8 (1042 capsule)	7168	72	{1:174763, 2:30105, 3:31283, 4:10224, 5:5904, 7:54, 8:576, 16:23, 48:2419}	3783	2560
leaver T8 (1042 capsule)	7168	72	{1:174763, 2:30105, 3:30275, 4:12240, 5:4896, 7:54, 8:576, 16:23, 48:2419}	3783	2560

All four rebuilt ledgers equal their assemblies. The last two cases are in-capsule ancestor assemblies, bound only through the capsule pin. The receipt also carries the values that #1200 (item 3) asks for: local_profile_per_call, idle_histogram and live_families_per_call (W, N, 896 live families per replica, 9 stages).

Controls: 43 tests, about 10 s. There are 36 semantic controls; each asserts the first check that rejects it.

• Frame replay:
  • descending, chain-broken and register-moved climbs;
  • a flipped centre;
  • an adapter or scalar at the wrong frame;
  • an extra phase marker.
• Dispatch:
  • a repointed local sweep, which leaves one call run twice and one never;
  • a foreign provider;
  • a multi-replica sweep;
  • a slot parked in a phantom replica, with a ghost copy left behind;
  • a whole stage dropped.
• Hosting:
  • a label-disjoint dependent axis;
  • a rank-0 child made pending through an orthonormality-preserving axis;
  • slot gaps;
  • duplicate or dependent remainder frames.
• Certificate tampering.

There are also 6 packaging controls (decoys, a foreign replay receipt, a foreign owner map, a flipped byte, unsafe tar members) and the positives.

A post-PR soundness review (second commit) found that slot execution, sweep dispatch and non-paid event frames were unchecked. No committed result was affected, and all are now closed with controls.

Overlap and credit

• Independent audit: selected PR1094 pending-axis independence and all source-touch cuts #1209 (akiva8201-dotcom, d1ba2631) already checks the pending-axis independence, helper widths, prefixes and idle-suffix profile from package inputs. Here they are cross-checks, not claimed as new.
• The in-lineage hpe_audit_local.py (receipt in the capsule; ported by Tooling: 2.33x faster complete emitted local audit with preserved coverage #1184, woklerx) already frame-replays both programs, including scalar and wrapper frames. Ours is code-disjoint, adds the binding to the template's calls, sweeps and owner map, and checks scalar frame tags, not arithmetic.
• The coverage identity comes from our Research note: restoration deferral is mass-neutral in the hosted T37 recurrence (no new kappa) #1012. Research note: packing ceiling (+0.87%) and 1/99 budget for the selected complex local word (no new kappa) #1240 (Douglas Colkitt / Mister-Meeseeks) uses it as a premise.
• Charts and phases belong to Research audit: exact phase criterion for selected PR1094 residual charts #1255 (bitplane) and are not validated here.
• Complementary: Independent verification: selected PR1042 finite invoice, 9 guards, 14 bootstrap steps and 94 outer slacks (seed 81, no new kappa) #1112, Independent verification: selected PR1042 replay-capsule closure (no new κ) #1142, Independent verification: complex canonical-frame rank identity on every frozen PR1042 edge #1149 and Independent verification: R480 private-center erasure and 40 CLEAR/COPY omissions (no new kappa) #1198.
• INDEPENDENT1094 takes the local factor from the word's self-reported local_profile and hard-codes W and N. This package derives both from the emitted program.

Reproduce

python3 -I -B research/frontier1094-emitted-ledger-seed78/verify.py        # ~3 s; PASS_FRONTIER1094_EMITTED_LEDGER (RESULT.json byte-identical)
python3 -I -B -m unittest discover -s research/frontier1094-emitted-ledger-seed78 -p test_package.py
python3.12 scripts/research_package.py --package research/frontier1094-emitted-ledger-seed78

Base: dadfea6e. Standard library only, no network. Generic package replay: PASS_REGISTERED_RESEARCH_PACKAGE. scripts/verify_pr_reconciliation.py passes; no Makefile change.

Open next questions

These are recorded in TASK.json:

• Can this be ported to the mixed-provider T37 templates (Conditional kappa 0.001675603180601302: mixed entrance providers on T37 owners #1183, Conditional kappa 0.001675670695307702: cross-call T37 entrance augmentation #1192, Conditional κ = 0.001675809683500501: mixed T37 allocation with X54 and fusions #1193, Conditional kappa 0.00167576776285696: mixed T37 source-holder fusions #1201, Conditional kappa 0.001675807682652: X54 on mixed T37 fusion providers #1207, Conditional kappa 0.001675809683500501: global mixed allocation with X54 and fusions #1218, Conditional κ = 0.001675809683500501: source-holder fusions and X54 on #1193's mixed T37 construction (independent; equals #1218) #1239) once capsules exist?
• Can the event-kind classification, premise (b) and the hosting semantics be checked from the compiler rules?
• Can Research audit: exact phase criterion for selected PR1094 residual charts #1255's chart-phase criterion be combined with these route bindings?

Assignment and disclosure

Track 75–84 (independent verification), seed 78 (OS CSPRNG). Repository helper nonce 3783431725445646d7112601d5501b9d. The claim was kept as a local task card; no discussion post was made.

Credit:

• Conditional kappa 0.001672660197076918: rare-pattern T16 exchanges #1094: Rohan Arun.
• The T16 lineage: jeremykothe2 (Conditional κ = 0.001672654541898561: improved T16 owner packing on the seven-slot k8 word #1087, Conditional κ = 0.00167166528321877: seven triangle-hybrid slots (PR987 search) on the k8 leaver word #1018), rohanarun (Conditional κ = 0.001662039077758742 — U10 (two new triangle-hybrid slots) on PR942's T37 owners #987) and FanPUYen (Research: T16 owner concentration raises conditional κ to 0.001661642255531641 #858).
• PR552/PR559 review, PR558 data phases, PR514/PR433 local replay.
• Josie Elliston: Research audit: independent owner-map check of all 15 open T16 claims on the seven-slot k8 word (no new kappa) #1091.
• Independent audit: selected PR1094 pending-axis independence and all source-touch cuts #1209, Tooling: 2.33x faster complete emitted local audit with preserved coverage #1184/Independent verification: complex canonical-frame rank identity on every frozen PR1042 edge #1149, Research note: packing ceiling (+0.87%) and 1/99 budget for the selected complex local word (no new kappa) #1240, Research audit: exact phase criterion for selected PR1094 residual charts #1255, Independent verification: selected PR1042 finite invoice, 9 guards, 14 bootstrap steps and 94 outer slacks (seed 81, no new kappa) #1112, Independent verification: selected PR1042 replay-capsule closure (no new κ) #1142 and Independent verification: R480 private-center erasure and 40 CLEAR/COPY omissions (no new kappa) #1198.
• Maintainer capsules: Douglas Colkitt, with OpenAI Codex assistance.

Prepared by Sagar Thakrar (thakrarsagar) with substantial AI assistance from Claude (Anthropic) via Claude Code. Parallel read-only scout, judge and overlap subagents selected and hardened the check. New material is Apache-2.0.

[Mister-Meeseeks]

100-PR batch review at 4b196b40dcac3ec40f62ec285babac53a1b5a3c5.

Merged to main, preserving the original contributor commits and package attribution.

Reviewed scope: Independent source-bound reconstruction of the emitted PR1094 complex rank ledger and three ancestor cases, with per-family conservation and 43 controls. Replays frames, routes and child counts; scalar arithmetic, event semantics and the selector premise remain inherited.

The pinned scoped command passed; the batch ledger contains its command, input hashes, environment and log. This incorporates the research/tooling contribution and does not change the selected κ or discharge the inherited multiplication-theorem assumptions.

Batch review, attribution and evidence. Thank you for the contribution.