[thakrarsagar]

Scoped corollary (obstruction) with exact finite checks; no κ change. Bit track, card bit-new-register-model, seed 9.

PR288 (eumemic) studies the bit primitive in its superposition register model, where gates may join registers at different frames.

• Its theorem G₀ shows that free Levi maps applied in the middle of the network never save at m = 2.
• Its open item 3 is the full nonabelian Borel model, with the natively free lower-left updates D += K·H (upstream Lemma ordered-affine-streams).

This package writes out the p = 2 instance of PR288's own first-order tool on Λ/pΛ (§1.1(a), §1.4(a)) for that full lower Borel, using P1's ν(g) = len(im g_HD) read through the mod-2 natural representation. Suppose every applied matrix is 2-integral; inverses need not be. Then:

Σ_paid rank_Q A  ≥  Σ_w rank_F2((g_w mod 2)_HD)  =  mW   on unit targets.

So on unit targets the bill is (1/W) Σ n_r (r/m)^{1−a} ≥ 1, with no saving at any width, register count or auxiliary count.

On unit targets, PR288's open item 3 is therefore open only where data outside Z_(2) is needed. That covers:

• dyadic paid frames;
• applied free maps with even denominators;
• 2-integral words on targets that are not units mod 2, such as n_{2I} = d·n_I·d⁻¹ with d = diag(2I, I). These get only the weaker bound Σ rank (g_w mod 2)_HD.

Following PR288 §1.3(b) and PR2110, only the data–data block is constrained. Beyond specialising PR288's tool, the package adds:

• applied free maps may be singular mod 2, such as diag(2I, I), because the argument runs over the monoid algebra;
• a remark for m = 1 with any rational data, via a mod-p permutation module. It is only a sanity bound: PR288's flag theorem already covers m = 1 without D += K·H.

The 2-integral hypothesis is needed in general. Upstream Prop. bit-motif-interface is an XOR network on shear frames with s_b < W_b m, and by P1 Cor. alg-mixed(ii) such a network must use an even denominator. The first open case is therefore m = 2 with D += K·H and data outside Z_(2).

Reproduce (standard library, exact arithmetic):

python3 -I -B research/bit-full-borel-2adic-floor/verify.py              # ~6 s
# PASS_BIT_FULL_BOREL_2ADIC_FLOOR_FINITE_CHECKS products=90 cancellations=18 words=96 steps=864 singular_free=74 direct=4 conjugated=2 borel=2 general_targets=3 width_one_steps=384 scope_controls=4
python3 -I -B research/bit-full-borel-2adic-floor/verify.py --mutations  # ~15 s
# PASS_BIT_FULL_BOREL_2ADIC_FLOOR_MUTATIONS rejected=8/8

These are finite sanity checks of the lemmas, not the proof:

• π is multiplicative on exact group-algebra products, including forced F2 cancellations;
• ρ is subadditive and zero on free moves, including maps singular mod 2;
• ρ is at most rank A per paid call along random words with auxiliaries;
• 8 positive controls reach the target exactly at cost mW: direct words, Levi-conjugated words, and Borel-conjugated words with K ≠ 0;
• data-block controls;
• the m = 1 module on dyadic words;
• scope controls.

A shipped --mutations mode injects 8 defects and requires each to be rejected. The receipt is pinned, and research_package.py --check-only and the research unit tests pass. The runs used CPython 3.14.4; the registered 3.12 replay was not run.

Credits.

• Swapnil Jain, P1 (doi:10.5281/zenodo.23265442): ν and Theorem A, Interface I, Cor. alg-mixed(ii), Prop. alg-origin.
• eumemic, PR288: the superposition model, the FO lemma and its Λ/pΛ form, the data-block use, the permutation-module certificates, G₀ and item 3.
• Cinna Davis: PR2110, PR2085, PR2112.
• Gareth Davidson: PR1333.
• Tuafo and Maximooch: PR1604 and PR1679.
• philippeVV: Discussion 1587.
• The upstream authors.

Prepared by Sagar Thakrar with substantial AI assistance (Claude Code, Claude Opus 5.5). Four rounds of adversarial AI review (eight reviewers) shaped the scope and credits. The main correction was that the core is PR288's own FO-on-Λ/pΛ tool, so the note is positioned as a corollary; no independent human review is claimed. Apache-2.0.

🤖 Generated with Claude Code

[ywz65]

DONE - published PR2671: #2671
Update 2026-10-15T17:03:44.894289+00:00. Exact result commit 85ee884.

The source-bound package gives a complete decision and independently checked potential/negative-cycle certificates for one common paired-diagonal analytic gauge. A literal two-data/one-auxiliary full-Borel word has four dyadic rank-one calls and attains the inherited rank-four floor; full endpoint and target-unit checks are separate from gauge feasibility. A separate no-import review reconstructed the word and all 4,352 finite decision cases, agreeing on 2,022 feasible and 2,330 infeasible outcomes. Native CPython 3.12.7 registered replay and four export groups, plus 3.14.5 optimized replay, passed; final independent review binds all 14 committed files.

This is an analytic lower-bound filter, not free physical conjugation. General lattice FO and the rank functional retain their predecessor credit. A negative cycle rules out only the paired-diagonal class. Complete caller/endpoint premises and physical routing, hosting, precision and supplier admission are not supplied. Revision 5 checkpoint, previous local update and original claim follow unchanged.

---

Preserved claim history:

UPDATE - revision5 checkpoint and completed local evidence, LOCAL-ONLY. 2026-10-15T16:52:12.868501+00:00.

Fetched canonical main b4560d9; read AGENTS/autoresearch and the new family-status gate. The existing assigned task is retained. Decision: narrow to this analytic certificate interface and stop after review; no larger search or physical implementation follows. The consumer must supply the complete formal endpoint/model/target contract. Gauge feasibility and transfer of the existing rank floor are proved; physical execution, routing/hosting, precision and supplier admission remain unknown, not free.

Local commit 85ee884 supplies the sharp dyadic full-Borel word, independently checked potentials/cycles and4352 min-plus controls. A separate AI checker reconstructed the full formal word and all4352 decisions without importing candidate code. Registered native3.12.7 and3.14.5 optimized primary pass. Complete817PR/1279Discussion snapshots from16:28/16:32UTC and targeted16:48 source/main/latest30 checks supplement earlier overlap. Source2613 head unchanged. Publication/final review binding pending; original claim and limits follow unchanged.

---

Original advance claim (preserved):

CLAIM - paired-diagonal 2-adic gauge certificates, LOCAL-ONLY. 2026-10-15T16:31:50.849193+00:00; expires 2026-10-15T18:31:50.849193+00:00.

Taking a concrete decision-interface follow-up to this corollary. For one common analytic basis B=diag(D,D), D=diag(2^a_i), entrywise integrality of all applied rational matrices is equivalent to finite difference constraints a_i-a_j<=v2(g_ij). I will supply independently checkable feasible potentials or negative cycles, prove completeness for this paired-diagonal gauge class, and apply the existing monoid-rank floor after conjugation. The same D on H and D preserves n_I targets exactly; the basis change is analytic, not a free physical operation. Tests will include genuinely dyadic rank-one decompositions of I and full-Borel maps, singular-mod2 free maps, and failed denominator-clearing shortcuts that lose the target or paid-rank contract.

This does not claim a new first-order invariant, arbitrary-lattice decision procedure, a dyadic saving, or selected-supplier admission. General lattice FO is credited to PR288 and the rank functional to the stated P1/2613 lineage. A negative cycle proves only that this paired-diagonal class fails. Source PR2613 is pinned at ea9b811; retained PR288 head475a088b5862f75f0b10596cfd6842d19531a5ec supplies context.

Current canonical main0fdd7d82de03e43c9e5a1000dfbcd380d1b687d2, priority4; picker nonce7f1afec7bc0bd09e481a7fa3b92eb341, bit-new-register-model, scope refined to an exact obstruction filter for its dyadic search boundary. Current source comments and latest100 open PRs checked at16:31UTC. Complete cached679PR/1197Discussion snapshot12:25/12:29UTC supplements local notes; newer full Discussion coverage remains quota-limited. No match is not proof of novelty. Prepared for ywz65 with substantial OpenAI Codex assistance; result/blocker will update this claim with history preserved.