SOURCE: https://physics.magflowmeters.com/gates/dossiers/deeproot-shape.html
======================================================================

DeepRoot — Shape (13D selector) — dossier & ledger 

 ← Gates scoreboard · Jump to closure ledger 

 Gate dossier — DeepRoot — Shape (13D selector)

 Question: Is 13D the most economical shape that fits our world? 
 Status (fixed): REDUCED-TO-AXIOM · ANCHORED +1 
 Full 13-dimensional treatment: the frozen arena M4 x K6=SU(3)/T2 x S2 x S1_Y/Z2, all three layers, full precision. Self-contained — every value derived or stated inline. 

 Required endpoint (2026-07-06)

 Status: CLOSED / REDUCED-TO-AXIOM .

 Nothing left. Anchored on: 

 Shape: the frozen 13D carrier — ordinary 4D spacetime × a 6D color shape K6 = SU(3)/T2 (the full flag shape of SU(3)) × a 2D sphere S2 for the weak force × a folded hypercharge circle S1Y/ℤ2 for hypercharge and left/right-handedness — this is the object under test

 Granularity: the world is recorded in a finite number of bits, so each tuned quantity costs description length — this is where the load-bearing scoring rule lives

 Scale: the high-energy boundary package (unification scale, compactification radius) — supplies the per-item cost size, not the deciding root here

 Observables: None consumed as a numeric fit (the gate is structural: the observed matter content enters as a fixed given on both sides of the economy comparison, and cancels). It reads three facts OFF the observed Standard-Model content rather than fitting them: exactly three matter generations, the six-fold shared-charge identification (ℤ6) of the gauge group, and that the color shape must carry SU(3). The observed spectrum itself is declared a measured anchor, not derived.

 Dissolution: No hidden derivation is claimed. The residual bottoms on the named value-free axiom/common-currency rule rather than an unbounded obligation.

 This is the current gate-level endpoint. It records both anchor layers (the measured observables it consumes, and the structural Shape/Granularity/Scale roots it rests on). The detailed dossier below is retained in full.

 Executive summary & honest status

 Headline. The thirteen-dimensional carrier

 \[
\mathfrak{B}_{\rm active}
=
\underbrace{\big[\,\mathcal{M}_4 \times K_6 \times S^2 \times S^1_Y/\mathbb{Z}_2\,\big]_\times}_{\times\ \text{STAGE}}
\;\oplus\;
\underbrace{\big[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,\big]_\oplus}_{\oplus\ \text{RULEBOOK}}
\;\otimes\;
\underbrace{\big[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,\big]_\otimes}_{\otimes\ \text{ACTORS}},
\qquad D = 4+6+2+1 = 13,
\]

 with \(K_6 = SU(3)/T^2\) the full \(A_2\) flag manifold, \(S^2\) the round two-sphere, and \(S^1_Y/\mathbb{Z}_2\) the hypercharge circle folded by the reflection \(\theta\mapsto-\theta\) , is not asserted here as the unique shape of nature. It is banked as the cheapest complete carrier found — inside a declared, frozen search category, under a declared record-cost accounting rule, it is the lexicographically-minimal object that carries all of the structure the Standard Model actually exhibits (three chiral fermion generations, the exact gauge group \(SU(3)_c\times SU(2)_L\times U(1)_Y\) quotiented by its full six-fold center identification, and the observed left-handed-only chirality with no surviving mirror partners). Every rival carrier that was actually tried and scored — eleven of them, spanning \(D=4\) through \(D=12\) plus non-dimensional constructions — either fails outright to generate the required target content or loses to the 13D branch on the record-cost ledger. None is refuted as impossible; all are outscored on the declared metric, by a real but modest margin of roughly a factor of four in total description length. That is the entire claim, and it is precise: selected by constraints, not derived from first principles, not proved unique. 

 The fixed grade, stated plainly and never adjusted here. This gate — deeproot-shape , "DeepRoot — Shape (13D selector)" — carries the terminal status REDUCED-TO-AXIOM / ANCHORED +1 . That status is fixed by the governing ledger and is reproduced here unchanged; this dossier does not upgrade it toward "derived" or "forced," and does not downgrade it toward "open" or "unproven." The public board verb is SG-1 = SELECTED BY CONSTRAINTS . Every substantive claim below is written to be consistent with exactly that terminal — no stronger, no weaker.

 The three precise non-claims that a working physicist must hold in mind before reading a single equation in this dossier , because each one is the specific overclaim a careless reading of "13D wins the ladder" invites:

 This is not a uniqueness proof. "Thirteen dimensions is forced" or "13D is the unique minimum across all conceivable architectures" is explicitly rejected. What is shown is a selection result inside a stated category under a stated cost metric — a minimum over a search space that was enumerated and frozen before scoring, not a theorem that no cheaper carrier could exist in any space of theories whatsoever. Treating "the funnel's winner" as "the forced answer" is a relabeling this document refuses to perform.

 This is not an absolute-irreducibility result. The stronger claim — that no competing architecture anywhere in mathematics could encode the same physical content more cheaply — is a universal negative over the unbounded space of all possible formal systems. Formally, it is equivalent to a lower bound on the Kolmogorov complexity \(K(T)\) of the target theory \(T\) , and Kolmogorov complexity is famously uncomputable : no algorithm can certify that a given description is the shortest possible one, because doing so would let it solve the halting problem. This is not a hole specific to this program — it is a ceiling on any claim of absolute minimality made by any theory in any field. It is refused as an axiom by design, not because the calculation ran out of budget.

 This is not a derivation of the Standard Model's chiral matter content. The specific bundle data \(E\) — three chiral generations with their exact hypercharge assignments — is consumed here as an irreducible measured anchor , not produced by the geometry. The geometry converts a given \(E\) into a family count via the topological index \(\chi(K_6,E) = -3\) (Bott–Borel–Weil applied to the spin \(^c\) Dirac operator on \(K_6\) twisted by \(E\) ), which correctly returns three chiral families — but that computation is conditioned on \(E\) , not a selection of \(E\) from nothing. The claim "geometry forces exactly three generations, full stop" would be false: the anomaly-cancellation filter alone admits infinitely many chiral spectra, so nothing here singles out our specific \(E\) from that infinite family. This point is load-bearing enough that it recurs as its own residual (§5 register, item R8) rather than being buried in a footnote.

 What kind of result this is, structurally. The frozen branch under test is not merely the four metric factors of the ×-Stage layer. It is the complete three-layer object: the ×-Stage (the metric geometry, the only layer carrying dimension — \(D=13\) ), the ⊕-Rulebook (the finite flavor chamber \(\mathcal{F}^+_{\rm finite}\) and the admissibility firewall \(\mathcal{C}_{\rm admiss}\) , zero-dimensional but load-bearing — Lemma 2 of the realization-minimality attack is refuted specifically because it lives at this layer), and the ⊗-Actors (the bundle/operator content \(\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\) , also zero-dimensional but the layer at which chirality, proton safety, and the gauge routing are actually enforced). A residual computed against only the ×-layer metric factors, ignoring the ⊕ and ⊗ contributions to the cost ledger, would be an artifact of a truncated object — the corpus's own layer-necessity result (labeled B2 in the underlying ledger) shows that every proper subset of \(\{\times,\oplus,\otimes\}\) fails to close the required admissibility gates, so all three layers must be carried together whenever this gate is evaluated. This dossier does that throughout: the economy ladder in §3 below is scored on the full three-layer object, not the bare metric shape.

 The economy result, in one paragraph. Under the declared cost metric — minimum description length (MDL), with per-injected-real anchor cost \(b = \log_2(1/\Delta_0)\) set by the Finite Operational Cell Law (the granularity floor \(\Delta_0 > 0\) ) — the 13D branch is scored against ten explicitly considered rival shapes spanning the dimension ladder \(D = 4\) through \(D=12\) plus non-dimensional (finite/discrete) constructions. The tally: zero rivals are refuted as physically impossible; one rival (a bare 6D construction) structurally cannot generate the required target content at all ( FAILS_TO_GENERATE_T ); the remaining ten all lose to the 13D branch on the record-cost ledger ( LOSES_TO_13D ). The margin is real: roughly a factor of four in total description length, counting the four irreducible physical anchors \(\{M_{\rm Pl},\, \alpha_i(M_Z),\, y_t,\, |V_{us}|\}\) together with the nine-to-ten additional injected reals the flavor chamber needs (sector normalizations \(N_d, N_e, N_\nu\) ; the threshold triple \((\delta_1,\delta_2,\delta_3)\) ; the Hosotani phase \(\theta_H^\star\) ) for an honest total near 13–14 charged reals. This dossier explicitly disclaims the larger "~18×" or "4→22" headline that an earlier, looser accounting produced: that number undercounted the flavor-chamber's injected reals, and the corrected, honestly-charged margin is the more modest but defensible ~4×. A smaller, true number beats a larger, unsupportable one, and this document reports the smaller one.

 Where the terminal actually sits — the two axes that must not be collapsed into each other. There are two independent questions in play, and the discipline of this dossier is to keep them separate rather than averaging them into a single hedge. Axis one asks: within the declared, frozen category, is the 13D branch the cost-minimal complete survivor? That leg is banked — DERIVED-GIVEN- \(E\) , certificate-conditional on the frozen branch hash and the declared metric, and the entire argument reduces to exactly one named, explicit axiom (the granularity \(\Rightarrow\) MDL common-currency bridge — see §4 of the full dossier). Because the whole stack bottoms out in one declared, auditable posit rather than an uncounted pile of assumptions, the gate is correctly classified as REDUCED-TO-AXIOM . Axis two asks a harder, architecture-neutral question: is this shape minimal not just among the ten rivals actually tried, but among every conceivable architecture meeting the same physical burden? That question remains genuinely open — five sub-lemmas of a "realization-minimality" argument are only partially discharged (one of them, governing the Rulebook layer, is refuted as stated and its debt relocated), and the whole-shelf completeness of the color-carrier search (was \(\{K_6, \mathbb{CP}^2\}\) really an exhaustive shelf of admissible \(SU(3)\) carriers?) is an acknowledged open flank. This second axis is a residual, shown honestly, not a re-opening of the gate — the fixed terminal (axis one, REDUCED-TO-AXIOM) stands regardless of how axis two eventually resolves, because axis two was never claimed closed in the first place.

 What this dossier establishes, and what it does not. This dossier establishes: (i) the complete, three-layer, full-precision definition of the frozen 13D branch and its exact geometric invariants (curvature, topology, Casimirs, KK spectra, heat-kernel coefficients) as computed on that branch; (ii) the derivation chain showing which structural facts about the branch are theorem-grade given the branch and given the chiral spectrum \(E\) — the weak and hypercharge carrier assignments, the family count \(\chi(K_6,E)=-3\) , the six-fold center identification with Smith normal form invariant factors \([1,6,6]\) , and the abelian-isotropy uniqueness argument that singles out \(K_6=SU(3)/T^2\) on the enumerated shelf of \(SU(3)\) -carrying candidates; (iii) the economy ladder itself, with its honest accounting of both the four irreducible anchors and the additional injected reals that the flavor chamber requires, arriving at the corrected ~4× margin rather than an inflated one; and (iv) the single named axiom (granularity \(\Rightarrow\) MDL) to which the entire selection argument reduces, stated explicitly enough that a reader could in principle attack it directly. It does not establish that this shape is the unique or forced geometry of nature; it does not establish absolute irreducibility (that demand is dissolved as uncomputable, not answered); it does not derive the chiral matter content \(E\) from anything more primitive (E is consumed as measured input); and it does not claim the ten-rival survey is a certified exhaustive classification — it is banked as a first-pass survey, with the closing exhaustion theorem (referred to in the underlying ledger as "Shape Certificate 3 / Lemma 5") explicitly marked open. Two honest, falsifiable handles are on the table for anyone who wants to attack this result rather than merely comment on it: name one architecture meeting the identical physical burden (three chiral generations, the correct gauge group with its correct center quotient, no surviving mirror fermions, proton stability) that scores strictly cheaper after full unfolding under the same declared metric — and the selection collapses; or produce an actual principle that forces exactly three chiral generations uniquely, rather than merely permitting them — and the \(E\) -anchor leg promotes. Absent either, the shape stands exactly where it is graded: anchored, not forced.

 Single-sentence endpoint preview. The 13D branch \(\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\) (with its full Rulebook and Actors layers) is the most economical complete carrier found for the observed Standard Model structure by a real, corrected margin of roughly \(4\times\) over ten explicitly scored rivals, with the whole argument resting on one declared cost-accounting axiom and one honestly-flagged measured anchor (the chiral spectrum \(E\) ) — a result that earns the status SELECTED BY CONSTRAINTS , not DERIVED , and that is exactly the terminal this gate is graded at: REDUCED-TO-AXIOM / ANCHORED +1.

 The community gap & state of the art

 The precise open problem

 Strip away every framework-specific vocabulary and the question this gate is built to answer is one sentence long: why does the world need a 13-dimensional carrier, split exactly M₄ × K₆ × S² × S¹_Y/ℤ₂ with K₆ = SU(3)/T², and not some cheaper or differently-shaped object? No research program anywhere — string/M/F-theory, noncommutative geometry, traditional Kaluza–Klein, finite/discrete approaches, or the exceptional-GUT literature — has ever derived the dimension and factorization of the internal space from a deeper principle in the sense of showing that no admissible alternative could do the job more cheaply. Every existing framework instead posits a target dimensionality (26 or 10 for the bosonic/superstring, 11 for M-theory, an a priori noncommutative algebra for spectral triples, an unmotivated compact manifold for classic Kaluza–Klein) and then works forward to show the posited object can reproduce, or come close to reproducing, the observed gauge group and matter content. The question "is this shape the cheapest complete carrier, and by how much" is essentially never posed as a falsifiable, scored comparison in that literature — it is normally settled by consistency requirements (anomaly cancellation, modular invariance, critical central charge) internal to the chosen framework, not by a cross-framework economy audit run against a pre-declared cost metric.

 This gate — DeepRoot Shape, the 13D selector — is the one place in the corpus that takes that question head-on and returns a graded, falsifiable answer rather than a philosophical gesture. The answer it returns is precise and intentionally narrow: inside a declared, frozen search category, under a pre-declared record-cost (minimum-description-length, MDL) order that ranks completeness above bare simplicity, the object 
$$
\mathfrak{B} {\rm active}=\big[\,\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\,\big] \times\ \oplus\ \big[\,\mathcal{F}^+ {\rm finite}\oplus\mathcal{C} {\rm admiss}\,\big] \oplus\ \otimes\ \big[\,\mathcal{E} {\rm matter}\oplus\mathcal{E} {\rm gauge}\oplus\mathcal{E} {\rm Higgs}\oplus\mathcal{E} {\rm proton}\,\big] \otimes
$$
with metric dimension \(D=4+6+2+1=13\) is the lexicographically-minimal complete survivor — the cheapest object anywhere on the searched shelf that carries all of the observed structure (three chiral generations, the \(\mathbb{Z}_6\) shared-charge identification of the gauge group, and color living specifically on the \(SU(3)\) flag manifold) rather than merely the gauge group itself. That is a selection result, not a uniqueness result, and the gate is scrupulous about the difference: selector-minimal inside a category is not the same claim as "13D is forced" or "13D is the unique consistent dimension in all of mathematics." The second, stronger claim is the one the wider community has spent five decades trying and failing to establish for any candidate dimensionality, in any framework — and this gate explicitly declines to claim it, for a reason grounded in computability, not modesty (see below).

 History of the problem

 The dimensionality question has a long, largely unsuccessful history. Kaluza's original 1921 proposal and Klein's 1926 quantization showed that a single extra circular dimension could geometrize electromagnetism, but offered no principle fixing the number of extra dimensions at one rather than any other integer — the choice of circle was put in by hand to match the one force known at the time. As the gauge sector grew ( \(SU(3)\times SU(2)\times U(1)\) ), the traditional Kaluza–Klein program scaled the internal manifold up to match, but at no point did the program supply an independent argument for why that number of extra dimensions and not fewer — the internal space is sized to the answer, which is the target-anchoring failure mode this gate is explicitly built to avoid (a shape chosen to fit the desired gauge group, rather than a shape whose economy is scored against a pre-declared, target-blind metric).

 Superstring theory supplied the most famous dimension-counting result in the field: the vanishing of the conformal anomaly on the string worldsheet fixes the critical dimension at 26 (bosonic) or 10 (superstring), and M-theory's extension to 11 dimensions follows from U-duality consistency among the five 10D superstring theories. These are genuine derivations — but derivations of a different number, fixed by worldsheet or membrane consistency, not by an economy argument over what is needed to reproduce the Standard Model. The reduction from 10 or 11 dimensions down to four observed dimensions is then performed by compactifying six or seven dimensions on a Calabi–Yau manifold, an orbifold, or a G₂-holonomy space, and the choice of compactification manifold among the many thousands cataloged (well over \(10^{300}\) distinct flux vacua are commonly cited in the string landscape literature) is not derived from a first-principles economy argument at all — it is selected, post hoc, by which one reproduces something resembling the observed particle spectrum, an anthropic or landscape-statistical argument standing in for a derivation. No landscape scan has produced a scored, falsifiable claim of the form "this particular manifold is the cheapest complete carrier by such-and-such a margin, under a pre-declared cost metric applied before the answer is known." The moduli-stabilization and vacuum-selection problem remains, by the string community's own long-standing assessment, unsolved in the sense of yielding a unique or even a small preferred set of compactifications.

 Noncommutative geometry (Connes–Chamseddine spectral-triple approach) takes a different route: rather than positing extra spacetime dimensions, it posits a finite noncommutative algebra \(A_F\) tensored onto the ordinary 4D spin manifold, and derives the Standard Model gauge group and Higgs sector from the spectral action applied to that algebra. This is an elegant reconstruction of the gauge and Higgs content from an algebraic input, but the algebra itself ( \(M_2(\mathbb{H})\oplus M_4(\mathbb{C})\) in the canonical Chamseddine–Connes–Marcolli construction) is chosen to reproduce the known SM content — it is not derived from an independent economy principle that would let a different algebra win if the world's content were different, and it supplies no extra continuous spacetime dimensions to compare against a 13D or any-D metric compactification on an equal footing. It is a rival architecture , not a rival point on the same dimension-counting axis, and the literature has not produced a cross-architecture cost comparison between the spectral-triple route and any compactified extra-dimension route.

 Grand unified theories built on a single larger gauge group — \(SU(5)\) , \(SO(10)\) , \(E_6\) — sidestep the dimensionality question almost entirely: they operate in ordinary four dimensions and absorb \(SU(3)\times SU(2)\times U(1)\) into a simple group whose spontaneous breaking reproduces the observed gauge symmetry, with proton decay and gauge coupling unification providing predictions the extra-dimensional programs also make in different guises. These frameworks say nothing about why 4D other than that it is observed, and nothing about why this simple group other than group-theoretic elegance (the smallest simple group containing \(SU(3)\times SU(2)\times U(1)\) with anomaly-free chiral fermion embeddings). Finite-state, discrete, and lattice approaches to fundamental structure (causal sets, spin foams, various finite-geometry programs) likewise recover a gauge structure in favorable cases but do not supply a scored economy argument against continuum extra-dimensional rivals.

 The single fact common to all five rival classes — string/M/F-theory, noncommutative geometry, traditional Kaluza–Klein, finite/discrete approaches, and simple-group GUTs — is that they all can be made to recover something resembling the Standard Model gauge group. That shared outcome is a tie , not a discriminator: every serious framework in the field passes the "can it reproduce \(SU(3)\times SU(2)\times U(1)\) " test, so recovering the gauge group cannot be used to rank frameworks against each other. What none of the five classes has ever supplied is a frozen, target-blind, cross-framework certificate stating "this architecture is strictly cheaper, under a metric declared before the comparison is run, than every named rival, after unfolding all of that rival's hidden bookkeeping (moduli, landscape choices, chosen algebras, ad hoc group selections) into the same currency." That absence — not the absence of a gauge-group match, which every framework has — is the actual state-of-the-art gap this literature has left open, and it is exactly the gap this gate is built to close, category by category, on a declared shelf.

 The state of the art / best existing bound

 Given that no cross-framework economy certificate exists in the literature, the honest "best existing bound" is not a numerical bound at all but a methodological one: the state of the art is the recognition, largely implicit rather than stated as a theorem, that dimension/shape selection in fundamental physics is currently underdetermined by any agreed metric . Different sub-communities implicitly use different, unstated cost functions — string theorists effectively rank by worldsheet/anomaly consistency first and phenomenological match second; GUT model-builders rank by "smallest simple group containing the SM embedding" first; noncommutative-geometry practitioners rank by "smallest spectral triple reproducing the spectral action" first — and these metrics are almost never made commensurable with each other. There is, in the literature at large, no analogue of a Kolmogorov-complexity or minimum-description-length scoreboard applied uniformly across string compactifications, spectral triples, and Kaluza–Klein shapes with a pre-declared scoring rule fixed before any of the candidates are scored.

 Against that backdrop, this program's contribution is to make the metric explicit and to run it. The corpus's own honest economy result — not the inflated version, but the audited one — is a margin of roughly 4× : under the declared MDL / record-cost metric, the 13-layer object \(\mathfrak{B}_{\rm active}\) beats a ladder of ten named rival shapes and structurally fails to even generate the target on one further rival, with zero outright refutations of the metric itself. Concretely the ladder tally is 0 REFUTED · 1 FAILS_TO_GENERATE_T · 10 LOSES_TO_13D — every rung from \(D=4\) through \(D=12\) plus assorted non-dimensional alternatives loses under the declared cost function, and the 6-dimensional rival on the ladder cannot even structurally carry the required target content (it fails to generate \(T\) at all, rather than losing a close economy race). This is banked as a first-pass survey , explicitly not yet a certified classification — the object that would upgrade it to a classification (a role-mechanism normal-form / exhaustion theorem covering every conceivable rival, not just the ten named and audited ones) is not yet built, and its absence is tracked as an open residual rather than glossed over.

 An earlier, less careful pass through the same ledger advertised the input-cost side of this economy argument as roughly an 18× advantage — going from four irreducible anchors to twenty-two over-determined outputs (the "4 → 22" headline). The audited, honest accounting corrects this: the fully charged cost is not four anchors alone but four anchors plus nine-to-ten injected reals, for roughly thirteen-to-fourteen reals total — the injected reals being the sector normalizations \(N_d, N_e, N_\nu\) carried in the flavor chamber \(\mathcal{F}^+\) (needed because those normalizations are not themselves derived, only the ratios within a sector are), the threshold triple \((\delta_1,\delta_2,\delta_3)\) , and the Hosotani phase \(\theta_H^\star\) . With that honest denominator the real advantage is the more modest but still genuine ~4× figure , not the overstated ~18×. Stating this correction explicitly, rather than quietly retaining the more impressive number, is itself part of what "state of the art" means for this gate: the community has no competing number to compare against, because no rival framework publishes an equivalently audited input-cost ledger for its own construction (moduli counts, discrete choices, and landscape parameters in string compactifications are rarely totalized into a single comparable cost figure at all).

 Why each prior attempt falls short — precisely

 Running through the five rival classes against the specific standard this gate sets (a frozen, pre-declared, target-blind cost comparison that outputs a scored margin) shows a common shortfall, but each class fails for a structurally distinct reason:

 String / M- / F-theory. The critical dimension (10, 11, or 26) is genuinely derived from worldsheet or membrane consistency — this is a real theorem, not a target-fit. But the subsequent reduction to four observed dimensions runs through a choice of compactification manifold from an enormous landscape, and that choice is not itself derived from an economy principle comparable to the one this gate applies. No landscape scan states, in advance of knowing the answer, "the cost metric is X; the manifold that minimizes X is Y" and then checks whether Y reproduces the Standard Model. The selection is effectively reversed: manifolds are searched for phenomenological match, which is precisely the target-anchoring failure mode (fitting the shape to the desired answer) that this gate's own admissibility rulebook ( \(\mathcal{C}_{\rm admiss}\) , with its freeze-before-compare barrier) is built to forbid. Until a landscape-wide, pre-declared, target-blind cost functional is published and scored, string/M/F-theory remains a serious, unaudited rival on this gate's own ledger — not refuted, just not yet certified against the same metric.

 Noncommutative geometry (spectral triples). The Chamseddine–Connes construction derives a great deal given the algebra \(A_F=\mathbb{C}\oplus\mathbb{H}\oplus M_3(\mathbb{C})\) (or equivalent) — the gauge group, the Higgs doublet, and (in later refinements) constraints on the top-quark mass and Higgs mass emerge from the spectral action applied to that algebra. But the algebra itself is not derived from an independent minimality principle; it is chosen because it is the smallest known algebra whose spectral triple reproduces the observed particle content, which is again a target-fit rather than a target-blind selection. Moreover the spectral-triple approach does not supply extra continuous spacetime dimensions at all, so it is not a rival point on the same \(D\) -axis this gate scores — a genuine cross-architecture cost comparison (spectral triple vs. compactified extra dimensions, in a common bit-cost currency) has never been carried out by either community, and remains open on both sides.

 Traditional Kaluza–Klein. The oldest and most transparent shortfall: the size and shape of the internal manifold is set, historically and in every modern revival, by how many gauge bosons are needed , not by an antecedent economy argument. There is no independent principle in the traditional-KK literature that would have predicted six extra compact dimensions (rather than five, or eight) before the gauge content to be reproduced was already known. This is exactly the failure this gate's declared search category is designed to exclude by construction (R2.5's forces-as-isometries category still requires an economy scan across all admissible carriers of a given isometry group, not a single hand-picked one).

 Finite-state / discrete / lattice approaches. These programs recover gauge structure in restricted, often toy, settings, but a general, scored comparison of a discrete/finite architecture against a continuum compactified shape, run under one shared bit-cost metric, does not exist in that literature. The comparison is not refuted — it has simply never been attempted with a pre-declared metric on both sides.

 \(SO(10)\) / exceptional GUTs. These frameworks do not address extra dimensions at all; they operate in 4D and recast the group-theoretic embedding problem. They are not commensurable rivals on a dimension-counting axis, but they are commensurable rivals on the complete-carrier economy axis this gate actually scores (does the total construction — group, symmetry-breaking chain, Higgs sector needed to break the GUT group down to the SM — cost more or less, in the same MDL currency, than the 13D layered object). That comparison has likewise never been run with a shared, pre-declared metric.

 In every one of the five cases the shortfall is the same at bottom: the literature has never fixed a cost metric before looking at the answer and then scored competing shapes against it. This gate's own honest status reflects that this problem is hard even for its own construction : the metric-selection question — whether the correct aggregation rule for costs is the declared additive-MDL order or a rival dimension-first lexicographic order — is itself still open on this gate's own ledger (tracked as residual R5/W-C, shared with the Granularity deep-root gate). Under a dimension-first lexicographic metric, a bare 4D effective field theory would win outright ( \(k_{\rm dim}=4<13\) ) regardless of how many hidden reals it must inject, and the entire economy ladder built here would fold. Which metric is the physically correct one to use is the single decisive, currently unresolved seam separating "selected under a declared, defensible metric" from "forced independent of metric choice" — and this gate is honest that it has reached the former, not the latter.

 Why "absolute uniqueness" is not the missing piece — and is not attempted

 It is worth being explicit about what this gate does not attempt, because the stronger claim is the one most naturally expected from a "why 13D" question and its absence should not read as an oversight. The strongest conceivable version of the community question — "is there any conceivable architecture, in any framework anyone could ever construct, that is strictly cheaper than this one after fully unfolding its hidden bookkeeping?" — is a universal negative over the space of all possible mathematical constructions . Formally, this is equivalent to bounding the Kolmogorov complexity \(K(T)\) of the target structure \(T\) (the observed gauge group, chirality, and generation count) from below across every possible description language, which is a well known uncomputable problem: Kolmogorov complexity has no general algorithm for establishing lower bounds against an unbounded space of encodings, and the target \(T\) itself already presupposes the very chiral spectrum \(E\) that both sides of any such comparison would need to reproduce. This is not a gap specific to this program; it is a shared mathematical ceiling on any claim of absolute architectural minimality, in any research program, in any field. The gate treats this correctly as an axiom-open permanent wall rather than as an unclosed hole to chase — declaring the boundary is the honest terminal, not a placeholder for future work.

 What the gate does achieve, and what genuinely separates it from the rest of the literature, is narrower and fully computable: a frozen, declared, target-blind category (compact classified internal factors realizing forces as isometries, plus admissible bundles under a stated rulebook), a pre-declared cost metric (MDL / record-cost, tied to the finite operational cell law that makes every description a finite bit-string), and a scored ladder run inside that category, producing an audited ~4× margin with an honestly corrected (not inflated) input-cost accounting. That is precisely the kind of result the wider dimensionality debate has lacked: not a proof that 13D is the only conceivable shape in all of mathematics, but a certificate that, under one stated and defensible scoring rule, it is the cheapest complete carrier found so far, with the five major rival architectures named individually as serious, unrefuted, currently-uncertified competitors rather than as either defeated strawmen or silently ignored alternatives.

 The frozen 13D arena at full precision

 0. What this gate looks at, and in which normalization

 DeepRoot–Shape is the gate that puts the entire carrier object itself on trial: not one coupling, not one mass, but the full thirteen-dimensional layered structure that everything else in the program is built on top of. Before any curvature number is quoted below, the normalization it is quoted in must be fixed, because the corpus pins the same geometry in two internally consistent metric conventions and a number is meaningless without knowing which one is being read.

 (A) Frozen physical (R₆) normalization. The internal radius is the derived compactification radius R₆ (chamber-center value R₆ = R₀). Curvature carries physical units of GeV². In this normalization the K₆ Ricci eigenvalue is Ric_i = 1/(2R₆²) and the scalar curvature is Scal = 3/R₆². This is the normalization used for every dimensionful downstream quantity: Planck normalization, KK spectra, threshold radii.

 (B) Killing-form normal metric. g = (−B)|_𝔪 with the Killing form B(X,Y) = 6·Tr(XY) on 𝔰𝔲(3), evaluated at the symmetric chamber center 𝐮 = (1,1,1). Curvature is dimensionless here. This is the normalization in which the exact-rational curvature invariants — |Riem|², |Ric|², the weight-6 cubic products, the heat-kernel a-coefficients — are computed and stored.

 The bridge between them is scale-invariant: ratios of curvature invariants are identical in both normalizations. The load-bearing one is Scal/Ric_i = 6 = dim K₆ in both — (3·R₆⁻²)/((1/2)·R₆⁻²) = 6 in (A); (5/2)/(5/12) = 6 in (B) — and likewise |Ric|²/Scal² = 1/6 and |Riem|²/Scal² = 23/75 agree in both. Every curvature figure below is tagged with which normalization it is quoted in.

 1. The object under test: the complete three-layer active branch

 The claim this gate adjudicates is not about a metric alone. It is about a single frozen, layered object — call it 𝔅_active — that must be read as a whole, never flattened to its metric factors:

 \[
\mathfrak{B}_{\rm active}
=
\underbrace{\big[\,\mathcal{M}_4 \times K_6 \times S^2 \times S^1_Y/\mathbb{Z}_2\,\big]_\times}_{\text{× STAGE — metric geometry}}
\;\oplus\;
\underbrace{\big[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,\big]_\oplus}_{\text{⊕ RULEBOOK — finite admissibility (0-dim)}}
\;\otimes\;
\underbrace{\big[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,\big]_\otimes}_{\text{⊗ ACTORS — bundles / operators (0-dim)}}
\]

 Three layers, three different jobs, and the DeepRoot–Shape question — "is 13D the most economical shape that fits our world?" — cannot be asked of the ×-layer alone. The ⊕ and ⊗ layers are non-metric (they add 0 dimensions) but they are part of the frozen branch and are never silently dropped: the economy comparison this gate runs is a comparison of the whole object's description length, and Lemma 2 (Rulebook minimality) is refuted specifically at the ⊕ layer while Lemma 3 (Actor minimality) lives at the ⊗ layer. A residual computed from a truncated version of 𝔅_active — say, from the four metric factors alone, ignoring the rulebook and actor content — would be an artifact of the truncation, not a fact about the shape.

 × STAGE — the four metric factors, D = 4 + 6 + 2 + 1 = 13. 

 Factor 
 Real dim 
 Metric 
 Primitive/derived 
 Role 
 Routes to force 

 M₄ = ℝ^{3,1} 
 4 
 Minkowski 
 primitive 
 observed spacetime 
 — (all gates, low-energy readout) 

 K₆ = SU(3)/T² 
 6 
 Weyl-rigid invariant (normal at center) 
 primitive 
 color source; spin-ℂ family index −3 
 SU(3)_c via left-isometry 𝔰𝔲(3) 

 S² 
 2 
 round 
 primitive 
 weak source; spin-ℂ doublet routing 
 SU(2)_L via isometry 𝔰𝔲(2) 

 S¹_Y 
 1 
 flat 
 primitive 
 parent hypercharge circle 
 U(1)_Y via isometry 

 S¹_Y/ℤ₂ 
 interval 
 induced 
 derived quotient (θ↦−θ) 
 chirality / no-mirror filter 
 U(1)_Y + orbifold chirality 

 Only this ×-layer carries metric dimension. D = 13 is a plain sum, 4+6+2+1, and the binding physical fact is that gauge forces in this construction arise as isometries of the internal metric factors: SU(2)_L is supplied by S² and by nothing else — in particular not by any SU(2) subgroup sitting inside SU(3) — while K₆ carries only SU(3)_c. This separation of duties (each force gets its own carrier) is precisely what the Shape gate is checking is economical: could a smaller carrier have done the same job without losing a force or a chirality fact?

 ⊕ RULEBOOK — finite admissibility, 0-dimensional. 

 F⁺_finite = {τ=ω, 𝒢_gen, Π_u, Π_d, Π_e, Π_ν, O_u, O_d, O_e, O_ν, φ_i, N_i, 𝒩_i, RG} — the flavor chamber: modulus, generation basis, sector projectors, chamber operators, phase rules, normalizations, the Yukawa-map procedure, and RG-transport rules.

 C_admiss = {selector v3, C1–C14, freeze-before-compare barrier, anomaly conditions, no-mirror parity table, Wilson-line winding rule, FCNC/mediator no-go} — the anti-fitting firewall specifying which moves, sectors, and deformations are legal.

 Neither piece adds a dimension; F⁺ is explicitly a finite/operator chamber, not a propagating metric factor — its Cartan-torus modulus τ is chamber data, not a Kaluza–Klein tower. But this layer is where the honest input-cost correction for this gate lives: F⁺ injects real numbers (N_d, N_e, N_ν, the sector normalizations) that it uses but does not derive, and that is exactly the fact that refutes Rulebook-minimality as a standalone claim (see §5 below).

 ⊗ ACTORS — bundles and operators, 0-dimensional. 

 \[\mathcal{E}_{\rm active}=\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton},$$
$$\mathcal{E}_{\rm matter}=S_{3,1}\otimes S^{\rm spin^c}_{K_6}\otimes S^{\rm spin^c}_{S^2}\otimes L_Y\otimes V_{SU(3)}\otimes V_{SU(2)}\otimes V_{F^+}.\]

 This is the bundle/operator content riding on top of the Stage: the matter spinor bundle, the gauge bundle, the Higgs Wilson-line bundle, and the proton-safety projector structure. Every one of these actors is pinned at three sub-layers of its own (base manifold+bundle, scheme/boundary/grading, connection/endomorphism/domain/readout) — worked out in full in §6 below for the objects this gate specifically touches.

 Discrete/topological structure riding on the Stage. Three chiral generations come from the spin-ℂ index χ(K₆,E) = −3. Charge quantization is a global ℤ₆ identifying the centers ℤ₃⊂SU(3)_c, ℤ₂⊂SU(2)_L, and a sixth root of unity on U(1)_Y, giving G_SM = (SU(3)_c×SU(2)_L×U(1)_Y)/ℤ₆. The Higgs is a Wilson-line/Hosotani mode with integer winding n_H=1. Electric charge is Q=T₃+Y with hypercharge lattice Y∈(1/6)ℤ.

 2. K₆ = SU(3)/T² — the color rung, in full geometric detail

 K₆ is the object doing the heaviest lifting in this gate's economy argument, so its geometry is worked out completely.

 Root system (A₂ = 𝔰𝔲(3)). Cartan basis (h₁,h₂,h₃) with h₁+h₂+h₃=0. Simple roots α₁=(1,−1,0), α₂=(0,1,−1), and α₁+α₂=(1,0,−1). The positive roots are {α₁, α₂, α₁+α₂}; the half-sum is ρ = ½Σ_{α>0}α = (1,0,−1), with ‖ρ‖² = 2 in the Killing normalization. The Weyl group is S₃, order 6 — and this order-6 count reappears below as the Euler characteristic of K₆ itself.

 Tangent decomposition. T(K₆) = 𝔪₁⊕𝔪₂⊕𝔪₃, each 𝔪_i a real 2-plane carrying root α_i (with α₃ ≡ α₁+α₂). The (−B)-orthonormal 𝔪-basis is {X_ij = E_ij−E_ji, Y_ij = i(E_ij+E_ji)}/√12 over the pairs (01),(12),(02), using the Killing form B = 6·Tr.

 Invariant metric and Ricci (Wang–Ziller/Nomizu). The homogeneous metric family is
$ \(g_{K_6}(\vec u)=u_1\langle\cdot,\cdot\rangle_{\mathfrak{m}_1}+u_2\langle\cdot,\cdot\rangle_{\mathfrak{m}_2}+u_3\langle\cdot,\cdot\rangle_{\mathfrak{m}_3},\) $
$ \(\mathrm{Ric}_k(\vec u)=\frac{(u_k-u_i+u_j)(u_k+u_i-u_j)}{2R_6^2\,u_iu_ju_k}\quad (i,j,k)\ \text{cyclic}.\) $
In Killing-normalized scale coordinates x₁,x₂,x₃ on 𝔪₁,𝔪₂,𝔪₃, the general-chamber Ricci eigenvalues are
$ \(\mathrm{Ric}_1=\frac{x_1^2-x_2^2+6x_2x_3-x_3^2}{12\,x_1x_2x_3},\quad
\mathrm{Ric}_2=\frac{-x_1^2+6x_1x_3+x_2^2-x_3^2}{12\,x_1x_2x_3},\quad
\mathrm{Ric}_3=\frac{-x_1^2+6x_1x_2-x_2^2+x_3^2}{12\,x_1x_2x_3},\) $
$ \(\mathrm{Scal}=\frac{x_1x_2+x_1x_3+x_2x_3-(x_1^2+x_2^2+x_3^2)/6}{x_1x_2x_3}.\) $
There are exactly 4 invariant Einstein metrics on SU(3)/T²: the normal metric (1,1,1) and the Kähler–Einstein metric (1,1,2) together with its 3 permutations. This is a classical result, reproduced independently here as a validation of the geometric engine. Off the Einstein loci the space is non-Einstein — this is precisely the "squashing" degree of freedom, parametrized by the chamber 𝐮 ∈ [1/2,3/2]³ (Weyl-rigid), with chamber-center witness u₁=u₂=u₃=1.000000000000000. Off-center configurations fail Weyl-rigid admissibility and are eliminated by the selector; the center is the value every K₆-dependent gate uses.

 Curvature at the symmetric center 𝐮=(1,1,1), both normalizations. 

 Quantity 
 [R₆-norm] value 
 [Killing-norm] exact rational 

 Ric₁=Ric₂=Ric₃ 
 1/(2R₆²) = 1.973920880217872×10³³ GeV² 
 5/12 

 Scal(K₆) 
 3/R₆² = 1.184352528130723×10³⁴ GeV² 
 5/2 

 Scal/Ric_i 
 6 (= dim K₆) 
 6 (= dim K₆) 

 Metric-scale-invariant curvature ratios (identical in both normalizations — load-bearing): 

 Invariant 
 Exact rational 
 Decimal 

 Scal² 
 25/4 
 6.25 

 ‖Ric‖² 
 25/24 
 1.041666666666667 

 ‖Riem‖² 
 23/12 
 1.916666666666667 

 ‖Riem‖²/Scal² 
 23/75 
 0.3066666666666667 

 ‖Ric‖²/Scal² 
 1/6 
 0.1666666666666667 

 Two negative controls are pinned explicitly: ‖Riem‖²/Scal² is confirmed at 23/75 and is never 31/147; ‖Riem‖² is never equal to 60 — that value belongs to the round unit S⁶, a different space entirely, and mistaking K₆ for S⁶ is a documented error mode this pack guards against.

 Scalar-curvature integral (dimensionless): ∫_{K₆} R√g d⁶x = Scal·Vol(K₆) = 12π³ = 372.0753201635977 under the Killing-form-absorbing normalization, or (2π)³√3 = 429.6356725105388 under the pure √g d⁶x normalization at R₆=1. Both forms are recorded so either source convention can be matched. The Euler characteristic χ(K₆) = 6 , an exact topological invariant equal to |S₃|, the order of the Weyl group — the same 6 that appears as Scal/Ric_i.

 Cubic/weight-6 curvature invariants (Killing-norm, Einstein center): 

 Invariant 
 Definition 
 Exact rational 

 K₁ 
 R_ab^{cd}R_cd^{ef}R_ef^{ab} = 8·tr(R_op³) 
 −113/72 

 K₂ 
 R_abcd R_aecf R_ebfd 
 −5/72 

 ‖∇Riem‖² 
 via Nomizu; passes 2nd Bianchi (0 violations) 
 1/4 

 Scal³ 
 
 125/8 

 Scal·‖Ric‖² 
 
 125/48 

 Scal·‖Riem‖² 
 
 115/24 

 Ric³ (=Ric·Ric·Riem) 
 
 125/288 

 Ric·‖Riem‖² contraction 
 
 115/144 

 The nonzero value ‖∇Riem‖² = 1/4 ≠ 0 means K₆ is homogeneous but not locally symmetric — a physically consequential fact, because it is the reason the a₆ heat-kernel graviton coefficient carries an owed Gelfand–Tsetlin ladder term rather than closing by a symmetric-space shortcut (see §4 below).

 Why K₆ specifically, and not another SU(3) carrier. The abelian-isotropy uniqueness fact underlying the color-rung selection is representation-theoretic: the centralizer C_{SU(3)}(T²) = T² — that is, T² is the unique purely abelian isotropy subgroup of SU(3), the Cartan torus and nothing more. On the enumerated shelf of candidate SU(3) carriers {K₆, CP²}, this makes K₆ = SU(3)/T² the unique clean carrier: CP² = SU(3)/U(2) has isotropy U(2), which is not purely abelian, and this over-produces gauge structure at Gate 2 (CP² is a certified branch-kill for this reason — its isotropy breaks admissibility, and an earlier claim that CP² offered a tunable family count is retracted as unsound, since CP² is Spin_c with a discrete topological integer index r(r+1)/2, not a continuous modulus). This uniqueness is banked only over the enumerated shelf {K₆, CP²} — whole-shelf completeness over every conceivable SU(3) carrier is a named open residual (R4/W-B in the closure register), not claimed here.

 3. S² — the weak rung

 The round 2-sphere carries SU(2) L. The general theorem underlying its selection (labeled F1 in the derivation chain) is a whole-shelf result, not a comparison against named competitors: no abelian or torus carrier of any dimension hosts a non-abelian SU(2) among its isometries. This closes off every circle/torus factor at once — SU(2)_L simply cannot come from any such factor, at any dimension, and the round S² is what supplies it. The line ds² {S²} = R₂²(dθ² + sin²θ dφ²) gives χ(S²) = 2. Dirac/Laplace eigenvalues on S² are ℓ(ℓ+1)/R₂² for ℓ ≥ |N|/2, with degeneracy 2ℓ+1, where N is the monopole/hypercharge sector:

 Sector N 
 Monopole charge 
 SU(2)_L rep routed 
 Role 

 0 
 0 
 1 singlet 
 weak-singlet routing 

 1 
 ±1 
 2 doublet 
 Q_L, L_L 

 2 
 ±2 
 3 triplet 
 W±, W⁰ adjoint 

 ≥3 
 ±N 
 (N+1)-plet 
 higher KK (thresholds) 

 The binding statement carried through every downstream gate: SU(2)_L is supplied by S², not by any SU(2) ⊂ SU(3) — the weak force and the color force are routed through geometrically distinct carriers, which is part of what the economy argument in this gate is checking is not wasteful.

 4. S¹_Y/ℤ₂ — the hypercharge/chirality rung

 The second general theorem (F2) rules out a bare circle: a closed odd-dimensional circle factor preserves both chiralities, producing mirror fermions, which are excluded by the LEP Z-width measurement. The fix is the ℤ₂ fold θ↦−θ, which supplies chirality with no surviving mirror partner. This orbifold quotient is worked out as a genuine equivariant (Donnelly) boundary defect, not an ordinary boundary condition: the reflection θ↦−θ has two isolated fixed points at θ=0, π, and the reflection trace is g-trace = 1, derived as 2 fixed points × 1/|1−dg| = 1/|1−(−1)| = 1/2 each, summing to 1. The orbifold heat-kernel traces split as
$ \(K^+=\tfrac12 K_{\rm circle}+\tfrac12\ (\text{even/+ parity, defect }+\tfrac12),\qquad
K^-=\tfrac12 K_{\rm circle}-\tfrac12\ (\text{odd/− parity, defect }-\tfrac12),\) $
with per-fixed-point a₀ defect +1/4 (parity +) and −1/4 (parity −). The active interval is [0,π], with Vol(S¹_Y/ℤ₂) = πR_Y.

 The chirality readout is the Atiyah–Singer–Patodi index theorem on this interval [0,π]: it returns n_L = +3, n_R = 0 — three left-handed families and zero surviving right-handed mirror partners, hand-checkable and exact. The chirality projector at the boundary is P_χ = ½(1+γ₅Γ₈), with Γ₈ the chirality operator on the internal 8-dimensional spinor bundle S(K₆)⊗S(S²)⊗S(S¹_Y). Per-field ℤ₂ parities at θ=0,π: Q_L(+,+) and L_L(+,+) carry zero modes; u_R, d_R, e_R, ν(−,−) carry zero modes via the sector projectors Π_u, Π_d, Π_e, Π_ν; every mirror parity assignment is forbidden — there is no surviving mirror zero mode anywhere in the spectrum. The Standard Model hypercharges read off this structure are Y(Q_L)=+1/6, Y(u_R)=+2/3, Y(d_R)=−1/3, Y(L_L)=−1/2, Y(e_R)=−1, Y(H)=+1/2.

 5. Radii, volumes, and the Planck normalization at full precision

 Compactification scale equals the unification scale via R₀ ≡ (2πM_U)⁻¹, with M_U fixed by the two-loop RG plus KK-threshold closure α₁(M_U)=α₂(M_U)=α₃(M_U).

 Symbol 
 Meaning 
 Exact equation 
 Value (16 sig figs) 
 Units 

 M_U 
 unification scale 
 closure residual 9.6×10⁻¹¹ 
 1.0×10¹⁶ 
 GeV 

 M_Z 
 comparison scale 
 PDG input 
 91.18760000000000 (±0.0021) 
 GeV 

 M_Pl 
 ordinary Planck mass 
 (ℏc/G_N)^{1/2}, not reduced 
 1.220900000000000×10¹⁹ 
 GeV 

 R₀ 
 natural compactification radius 
 (2πM_U)⁻¹ 
 1.591549430918954×10⁻¹⁷ 
 GeV⁻¹ 

 R₆ ≡ R_{K₆} 
 K₆ overall radius 
 R₀·u_chamber, center u=1 
 1.591549430918954×10⁻¹⁷ (center) 
 GeV⁻¹ 

 R₂ ≡ R_{S²} 
 S² radius 
 R₀·s₂, s₂=1 at center 
 1.591549430918954×10⁻¹⁷ 
 GeV⁻¹ 

 R_Y ≡ R_{S¹_Y} 
 hypercharge circle radius (post-ℤ₂) 
 R₀·s₁, orbifold half-factor 
 7.957747154594768×10⁻¹⁸ 
 GeV⁻¹ 

 R_{T²_Cartan} 
 Cartan-torus radius inside F⁺ 
 R₀√(2/√3) at τ=ω 
 1.710231163476377×10⁻¹⁷ 
 GeV⁻¹ 

 The chamber-center witness is exact: u₁=u₂=u₃=1.000000000000000 within the Weyl-rigid chamber 𝐮 ∈ [1/2,3/2]³. Off-center points fail admissibility and are excluded by the selector.

 Product volumes. Exact symbolic forms:
$ \(\mathrm{Vol}(K_6)(\vec u)=V_{K_6,0}\,R_6^6\sqrt{u_1u_2u_3},\qquad V_{K_6,0}=\frac{(2\pi)^3}{\sqrt3},\) $
$ \(\mathrm{Vol}(S^2)=4\pi R_2^2,\quad \mathrm{Vol}(S^1_Y)=2\pi R_Y\ (\text{parent}),\quad \mathrm{Vol}(S^1_Y/\mathbb{Z}_2)=\pi R_Y\ (\text{active}),\) $
$ \(\mathrm{Vol}(X_{\rm active})=\mathrm{Vol}(K_6)\,\mathrm{Vol}(S^2)\,\mathrm{Vol}(S^1_Y/\mathbb{Z}_2).\) $

 Evaluated at the chamber center:

 Quantity 
 Exact formula 
 Value (16 sig figs) 
 Units 

 V_{K₆,0} 
 (2π)³/√3 
 143.2118575035129 
 — 

 Vol(K₆) 
 V_{K₆,0}R₀⁶ 
 2.327554010848277×10⁻⁹⁹ 
 GeV⁻⁶ 

 Vol(S²) 
 4πR₀² 
 3.183098861837907×10⁻³³ 
 GeV⁻² 

 Vol(S¹_Y) parent 
 2πR₀ 
 1.000000000000000×10⁻¹⁶ (exact = 1/M_U) 
 GeV⁻¹ 

 Vol(S¹_Y/ℤ₂) active 
 πR₀ 
 5.000000000000000×10⁻¹⁷ (exact = 1/(2M_U)) 
 GeV⁻¹ 

 Vol(X_parent) 
 product 
 7.408834522797404×10⁻¹⁴⁸ 
 GeV⁻⁹ 

 Vol(X_active) 
 product 
 3.704417261398702×10⁻¹⁴⁸ 
 GeV⁻⁹ 

 The exactness of the S¹_Y volumes is structural, not numerical coincidence: the 2π in the volume formula cancels the 2π in R₀ = 1/(2πM_U), leaving exactly 1/M_U (parent circle) and 1/(2M_U) (active orbifold).

 Planck normalization. With D = 13 and X_int = K₆×S²×(S¹_Y/ℤ₂) a 9-dimensional internal space,
$ \(M_{\rm Pl}^2=M_*^{\,D-2}\,\mathrm{Vol}(X_{\rm active}),\quad D=13,\) $
$ \(M_*^{11}=\frac{M_{\rm Pl}^2}{\mathrm{Vol}(X_{\rm active})}=4.023836152402511\times10^{185}\,\mathrm{GeV}^{11},\quad M_*=7.467050992135091\times10^{16}\,\mathrm{GeV}.\) $
M_ is fixed by the geometry together with the measured M_Pl — it is not an independent input. This equation is one of the two places (the other being the coupling-routing integral in §7) where the 13-dimensional count D and the specific volume Vol(X_active) become physically load-bearing rather than bookkeeping: change D or the volume and M_ moves.

 6. Threshold structure: what the full 13D geometry costs at one loop

 The one-loop SM beta coefficients (GUT-normalized α₁ = (5/3)α_Y) are
$ \(b_1^{\rm SM}=\frac{41}{10}=4.100000000000000,\quad b_2^{\rm SM}=-\frac{19}{6}=-3.166666666666667,\quad b_3^{\rm SM}=-7,\) $
fixed by the SM content (3 chiral generations + 1 Higgs doublet + SM gauge sector) — itself a downstream consequence of the geometry above via χ(K₆,E) = −3 and the APS chirality index n_L=+3.

 The full compactification contributes a threshold packet from each geometric piece:

 Packet 
 δb₁ 
 δb₂ 
 δb₃ 

 K₆ matter (3 gen, quark color) 
 0 
 0 
 +0.7900 

 S² matter (3 gen, weak doublets) 
 0 
 +0.9200 
 0 

 K₆ weak/color gauge + ghost net 
 0 
 −4.0200 
 −2.4900 

 S¹_Y/ℤ₂ hypercharge packet 
 −0.8400 
 0 
 0 

 S¹_Y/ℤ₂ hyper zero-mode matter (ΣY²=10/3/gen ×3) 
 +3.2140 
 0 
 0 

 Higgs Wilson-line (n_H=1) 
 +1.0470 
 −0.2110 
 0 

 Orbifold boundary (θ∈{0,π}) 
 +1.4214 
 +0.1998 
 −0.0313 

 Total (δ₁,δ₂,δ₃) 
 +4.8424 
 −3.1112 
 −1.7313 

 \[\boxed{(\delta_1,\delta_2,\delta_3)=(+4.8424,\ -3.1112,\ -1.7313)\ \pm\ 1.6\times10^{-3}.}\]

 The unification residual |α_i⁻¹(M_U) − α_j⁻¹(M_U)| = 9.6×10⁻¹¹ is a numerical-pipeline floor, comfortably inside the propagated PDG band of order 10⁻³ (two-loop SM RG, MS-bar scheme, M_Z = 91.1876 GeV). Every packet in this table traces to a specific geometric factor — the K₆ matter and gauge sectors, the S² doublet sector, the orbifold hypercharge and boundary defects, the Higgs Wilson line — and this table is the concrete accounting that the "shape carries this much physics per dimension" economy argument is built from. Related exact group-theory inputs: Dynkin indices T_adj(SU(3))=3, T_adj(SU(2))=2, T( 3 )=T( 2 )=1/2, and Σ_f Y_f² = 10/3 per generation.

 Gauge-coupling routing is fixed by
$ \(g_A^{-2}=M_*^{D-2}\int_{X_{\rm int}}\sqrt g\,|\xi_A(y)|^2\,d^{D-4}y,\) $
with SU(3)_c ← K₆, SU(2)_L ← S², U(1)_Y ← S¹_Y/ℤ₂, and 1/g_em² = 1/g₁² + 1/g₂² at M_Z. The three α_i⁻¹(M_Z) values themselves are declared anchors, not first-principles predictions — these integrals are consistency links between the geometry and the anchors, not a derivation of the anchors. The Gate-2 closure claim riding on this routing is the identification of the surviving 4D algebra as 𝔰𝔲(3)_c⊕𝔰𝔲(2)_L⊕𝔲(1)_Y — a structural fact about which isometries survive, not a numerical prediction of their strengths.

 7. The ⊗-layer objects this gate specifically touches, all three sub-layers pinned

 Each bundle/operator relevant to the Shape economy argument is pinned at its own three sub-layers — × Stage (manifold+bundle base), ⊕ Rulebook (scheme/convention/boundary/projector/grading), ⊗ Actors (connection ∇, endomorphism E, operator domain, readout):

 Bundle/operator 
 × Stage (base) 
 ⊕ Rulebook (scheme/boundary/grading) 
 ⊗ Actors (connection/E/domain/readout) 

 Scalar Laplacian Δ₀ 
 K₆ (& each ×-factor) 
 Killing-norm normal metric, Einstein center; MS-bar 
 ∇ = Levi-Civita (Nomizu); E=0; domain C^∞(K₆); readout = spectrum C₂(p,q)/R₆² 

 Vector/Hodge Laplacian 
 T*K₆ 
 1-form grading, same metric 
 ∇ = LC; E = Ric = (5/12)Id (mult 6); Weitzenböck 

 Graviton Sym²₀ (dim 20) 
 Sym²₀T*K₆ 
 TT gauge, Lichnerowicz grading 
 E_L spectrum {1/6, 5/12, 7/6, 17/12}; GT-hopping off-diagonal OWED 

 Dirac ⨸D_{K₆} (spin-ℂ) 
 S^{spin^c}_{K₆} 
 spin-ℂ structure, Chern class fixed to family index −3 
 ∇ = spin-ℂ connection; m² = (C₂+‖ρ‖²+Δ_{spin^c})/R₆², ‖ρ‖²=2 

 Dirac/Laplace S² 
 S^{spin^c}_{S²}, sector N 
 monopole grading N∈{0,1,2,…} 
 eigenvalues ℓ(ℓ+1)/R₂², ℓ≥ 

 Hypercharge line bundle 
 L_Y on S¹_Y/ℤ₂ 
 ℤ₂ orbifold parity, Y∈(1/6)ℤ, ℤ₆ center 
 KK momentum p_θ=(n+α)/R_Y, twist α∈{0,Y} 

 Gauge ℰ_gauge 
 T*M₄⊗ad(P) 
 BRST/FP gauge-fixing, Gribov domain 
 A, F, ρ_rep, KK tower; Q_BRST cohomology 

 Higgs ℰ_Higgs 
 L_γ⊗V_{SU(2),doub} on cycle γ 
 Wilson-line winding n_H=1, Hosotani grading 
 holonomy θ_H; readout = V_Hos minimum 

 Proton ℰ_proton 
 Π_qℰ_matter⊗Π_ℓℰ_matter 
 sector-orthogonal partition 
 four-fermion domain; identity Π_qMΠ_ℓ=0 

 Two entries deserve emphasis because they carry the gate's own honesty. The graviton Sym²₀ endomorphism is fully certified at the spectrum level — the Lichnerowicz operator (E_Lh)_ab = Ric_ac h^c_b + Ric_bc h^c_a − 2R_acbd h^cd gives, on the full 21-dimensional Sym² bundle, eigenvalues 1/6 (×6), 5/12 (×6), 7/6 (×6), 17/12 (×2), 5/3 (×1, the pure-trace mode), and on the transverse-traceless 20-dimensional piece the same four eigenvalues without the trace mode, with tr E_L = 40/3 and tr E_L² = 241/18. But the off-diagonal Gelfand–Tsetlin hopping matrix elements — needed to complete the a₆ heat-kernel coefficient — are exact-in-principle SU(3) ladder-operator formulas that have not yet been enumerated in the atlas. This is a named, bounded computation debt (the a₆ Route-A blocker), not an in-principle gap: the underlying invariants (K₁=−113/72, K₂=−5/72, ‖∇Riem‖²=1/4, and the rest of the weight-6 table above) are all certified; only the final assembly step is owed.

 Heat-kernel scalar ratios (Killing-norm, Einstein center): 

 Space 
 a₂/a₀ 
 a₄/a₀ 
 a₆/a₀ 

 K₆ scalar 
 5/12 
 11/120 
 OWED (Gilkey constants; underlying invariants certified) 

 S² scalar (r=1) 
 1/3 
 1/15 
 4/315 

 S⁶ round unit (calibration) 
 5 
 12 
 1139/63 

 The S⁶ row is a passed control: it calibrates the a₄ formula (returns exactly 12) and confirms K₆ is not accidentally S⁶ — reinforcing the negative control on ‖Riem‖² above. For the K₆ vector (tangent) bundle, tr a₂ = 0 and tr a₄ = −47/360 (using E=Ric, Ω=Riem, tr(Ω_ab Ω^ab) = −|Riem|²).

 Convention governing all heat-kernel numbers: K(t) ~ (4πt)^{−d/2} Σ_k a_{2k} t^k, densities per unit volume, with the exact product rule a_{2k}(M₁×M₂) = Σ_{i+j=k} a_{2i}(M₁)a_{2j}(M₂) — this convolution rule is what lets the per-factor packets in the threshold table (§6) and the per-factor heat-kernel rows above be assembled into a statement about the full 13D product space rather than about any one factor in isolation.

 8. Topology and discrete structure: χ values, the ℤ₆ quotient, and the finite chamber

 Euler characteristics of the three internal factors: χ(K₆) = 6, χ(S²) = 2, χ(S¹_Y/ℤ₂) = 1. K₆'s χ=6 equals the order of the Weyl group S₃, a clean topological fingerprint of the flag-manifold structure.

 The ℤ₆ center-kernel. The Smith normal form of the charge-character matrix for ℤ₃×ℤ₂×ℤ₆ has invariant factors [1, 6, 6] , with annihilator ℤ₆: this identifies ℤ₆ as the trivially-acting center subgroup and the finest faithful quotient — G_SM = (SU(3)_c×SU(2)_L×U(1)_Y)/ℤ₆, with generator z = (ω₃, −1, ζ₆) of order 6. This is a derived-given-E fact (it depends on the matter content E acting on the gauge group) but the group-theory computation itself — the Smith normal form — is exact and unambiguous once E is fixed.

 The F⁺ finite chamber (⊕-layer, full precision). Non-metric, adding 0 dimensions, but carrying real numerical content: the Cartan-torus modulus τ = ω = e^{2πi/3} = −0.5000000000000000 + 0.8660254037844386 i, an order-3 modular fixed point. The generation basis 𝒢_gen has dim_ℂ = 3, matched to the family index −3. Chamber Boltzmann factors derived from τ=ω include κ = e^{−π√3} = 0.004333420509983131, K_tb^crit = e^{−π√3/16} = 0.7117081304239685, and η_BK = 1/(32π e^{√3/(24π)}) = 0.009721281516312024. The CKM holonomy phase is δ_CKM = −2π/3 = −120.0°, and the lepton Berry phase is +2π/3 = +120.0°.

 9. What each piece of the arena physically carries — summary

 Pulling the geometry together into the physical bookkeeping this gate checks for economy: M₄ is the observed 3+1 spacetime stage, a declared axiom (not a forced rung — it is the observational primitive against which everything else is compared). K₆ = SU(3)/T² is the unique clean carrier (among the enumerated shelf) of color SU(3)_c, and simultaneously supplies the family count −3 via its spin-ℂ index — one geometric factor doing two jobs (gauge group + generation count) is part of the economy case. S² is the unique kind of carrier (over every dimension, not just a named shelf) that can host non-abelian SU(2)_L — no torus or abelian factor at any dimension can do this job. S¹_Y/ℤ₂ supplies both U(1)_Y and, via its orbifold fold, the entire chirality structure (n_L=+3, n_R=0, zero surviving mirrors) — again one factor carrying two jobs. The ⊕ rulebook (F⁺ + C_admiss) is the finite admissibility law that fixes flavor structure and forbids illegal deformations, at the cost of a few injected real numbers (N_d, N_e, N_ν) that it uses without deriving. The ⊗ actors are the bundle/operator content — matter spinors, gauge fields, the Higgs Wilson line, the proton-safety projector — riding on the Stage under the Rulebook's constraints.

 This is the complete, full-precision picture of the object that DeepRoot–Shape asks: given everything this thirteen-dimensional layered arena carries — three independent gauge-force carriers, the full chirality/no-mirror structure, three chiral generations, and the flavor and proton-safety machinery — is it the most economical complete carrier available? The geometric content above (the exact rationals, the radii, the volumes, the threshold packets, the topological invariants) is the ledger every economy comparison in this gate's derivation chain is computed against.

 Construction I - the deep-root anchoring

 This gate is, by name and by construction, the Shape root itself — it is the place where the deep-root apparatus is not merely invoked in passing but is the very object under audit. That makes the discipline of this section different from a gate that merely uses Shape as a background fact: here, Shape has to be built out in full, its supporting roots (Scale, Granularity) have to be shown actually doing load-bearing work rather than sitting as decoration, and the four Layer-2 admissibility screens (Invariance, Record-Interface, Causal-Order, Nonseparability) have to be run against the complete, unflattened three-layer object — never against a truncated stand-in. What follows states, root by root and screen by screen, exactly what is eliminated, what is forced, and what is merely exposed as an open seam, with no rounding of any of the three into the others.

 I.1 Shape — the object under test, in full

 The frozen active branch is not a bare metric space. It is the complete three-layer object

 \[
\mathfrak{B}_{\rm active}
=
\underbrace{\big[\,\mathcal{M}_4 \times K_6 \times S^2 \times S^1_Y/\mathbb{Z}_2\,\big]_\times}_{\times\ \text{STAGE}}
\;\oplus\;
\underbrace{\big[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,\big]_\oplus}_{\oplus\ \text{RULEBOOK}}
\;\otimes\;
\underbrace{\big[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,\big]_\otimes}_{\otimes\ \text{ACTORS}}.
\]

 Only the ×-Stage layer carries metric dimension: \(D = 4 + 6 + 2 + 1 = 13\) , with \(\mathcal{M}_4=\mathbb{R}^{3,1}\) the observed spacetime factor (a declared axiom, not a forced rung — see residual R9 below), \(K_6 = SU(3)/T^2\) the full \(A_2\) flag manifold carrying color, \(S^2\) the round two-sphere carrying the weak isometry, and \(S^1_Y/\mathbb{Z}_2\) the hypercharge circle folded by the reflection \(\theta\mapsto -\theta\) , which carries both \(U(1)_Y\) and the chirality filter. The \(\oplus\) -Rulebook layer — the finite flavor chamber \(\mathcal{F}^+_{\rm finite}=\{\tau=\omega,\ \mathcal{G}_{\rm gen},\ \Pi_u,\Pi_d,\Pi_e,\Pi_\nu,\ O_u,O_d,O_e,O_\nu,\ \phi_i,\ N_i,\ \mathcal{N}_i,\ \mathrm{RG}\}\) together with the admissibility firewall \(\mathcal{C}_{\rm admiss}\) — and the \(\otimes\) -Actors layer — \(\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\) — are both zero-dimensional, but neither is decoration: they are counted reals in the cost ledger and they are where two of the five realization-minimality sub-lemmas actually break or hold. This is the operative discipline for the whole section: Shape here means the full \(\mathfrak{B}_{\rm active}\) , never the bare \(\times\) -Stage metric factors alone. The layer-necessity result (B2 in the underlying ledger) shows every proper subset \(L \subsetneq \{\times,\oplus,\otimes\}\) has empty null-space inside the declared category — dropping any one layer fails to close required admissibility gates — so a residual computed against a truncated two-layer or one-layer stand-in would be an artifact of the truncation, not a fact about the geometry.

 What Shape eliminates (the general theorems, hand-checkable, closing whole shelves rather than named competitors). Two structural facts are DERIVED as general theorems over entire families of alternative carriers, not merely checked against the two or three competitors someone happened to write down:

 Weak rung, theorem F1. No abelian or torus carrier of any dimension hosts a non-abelian \(SU(2)\) among its isometries. This eliminates every circle, every torus \(T^n\) , and every product of circles as a candidate for the weak sector — not because each was tried and failed, but because the isometry group of a flat torus is itself abelian, so \(SU(2)_L\) structurally cannot emerge from any such factor. The round \(S^2\) , with isometry group \(SO(3)\cong SU(2)/\mathbb{Z}_2\) , supplies it instead. This also forecloses the tempting shortcut of hiding \(SU(2)_L\) inside \(SU(3)\) 's own subgroup structure: the binding rule is that weak \(SU(2)_L\) comes from \(S^2\) , not from any \(SU(2)\subset SU(3)\) acting on \(K_6\) .

 Hyper rung, theorem F2. Any closed odd-dimensional (bare-circle) internal factor preserves both chiralities, producing mirror fermions that are excluded by the measured LEP \(Z\) -width (a real, checkable experimental constraint, not a theoretical preference). This eliminates the bare circle \(S^1_Y\) as a viable hypercharge carrier on its own. The \(\mathbb{Z}_2\) orbifold fold \(\theta \mapsto -\theta\) removes the surviving mirror partners; the Atiyah–Singer–Patodi index computed on the resulting interval \([0,\pi]\) returns exactly \(n_L=+3\) , \(n_R=0\) — three left-handed families, zero right-handed survivors, with the index computation carrying \(\mathrm{tr}\,a_0\) defects of \(+1/4\) (even parity) and \(-1/4\) (odd parity) at each of the two fixed points \(\theta=0,\pi\) , so the two isolated fixed points contribute a total reflection trace of \(1\) (derivation: two fixed points \(\times\ 1/|1-(-1)| = 1/2\) each \(=1\) ).

 Both eliminations are theorems over shelves — every abelian/torus carrier of every dimension for F1, every bare odd-dimensional circle for F2 — which is a materially stronger elimination than a case-by-case survey, and it is why these two rungs are correctly labeled DERIVED rather than merely DERIVED-GIVEN- \(E\) : they do not depend on the specific chiral spectrum \(E\) at all, only on group-theoretic and topological facts about isometry groups and index theory.

 What Shape forces, conditional on the enumerated shelf. The color rung is different in kind: \(K_6 = SU(3)/T^2\) is not eliminated-over-all-shelves in the same sense: it is forced only over the enumerated shelf \(\{K_6,\ \mathbb{CP}^2\}\) of \(SU(3)\) -carrying coset spaces, by an abelian-isotropy uniqueness argument. The centralizer computation is exact: \(C_{SU(3)}(T^2) = T^2\) — the maximal torus is its own centralizer in \(SU(3)\) , i.e. purely Cartan, with no additional commuting generators. This makes \(T^2\) the unique purely-abelian isotropy subgroup on the shelf, and \(K_6=SU(3)/T^2\) the unique clean carrier (isotropy exactly the Cartan torus, no residual gauge symmetry left over from a larger isotropy group). The rival on the same shelf, \(\mathbb{CP}^2 = SU(3)/U(2)\) , is a named branch-kill : it was built out end-to-end and breaks specifically at Gate 2 because its \(U(2)\) isotropy over-produces gauge content. An earlier claim that \(\mathbb{CP}^2\) offered "a tunable family count" is explicitly retracted as unsound — \(\mathbb{CP}^2\) is \(\mathrm{Spin}_c\) with a discrete topological integer index \(r(r+1)/2\) , not a continuous modulus, so there was never a tunable knob there to begin with; the correct statement is the abelian-isotropy uniqueness argument above, and any argument leaning on \(\mathbb{CP}^2\) -tunability is reviving a killed direction. Because this elimination is over an enumerated shelf of two named candidates rather than a theorem over the space of all possible \(SU(3)\) -carrying coset spaces, the whole-shelf completeness question is honestly flagged as residual R4/W-B below — Shape does the work it claims to do (forcing \(K_6\) over \(\{K_6,\mathbb{CP}^2\}\) ) but does not yet claim to have enumerated every conceivable \(SU(3)\) carrier.

 What Shape exposes, given the branch. With the branch fixed, two further structural facts are read off, both explicitly conditioned on the chiral spectrum \(E\) (the measured anchor at the stack's bottom, residual R8): the family count \(\chi(K_6,E) = -3\) (Bott–Borel–Weil / spin \(^c\) index at weight \((1,0)\) , giving three chiral families given \(E\) — not selecting \(E\) ), and the center-kernel charge structure, \(\mathbb{Z}_6 = \ker(Z(G_0)\to \mathrm{Aut}(E))\) , with Smith normal form of the charge-character matrix returning invariant factors \([1,6,6]\) and generator \(z=(\omega_3,-1,\zeta_6)\) of order 6 — the finest faithful quotient, \(G_{\rm SM}=(SU(3)\times SU(2)\times U(1))/\mathbb{Z}_6\) , with electric charge \(Q=T_3+Y\) and hypercharge lattice \(Y\in\frac16\mathbb{Z}\) . Both facts depend on \(E\) ; neither derives \(E\) . This is the sharpest place the "Shape forces the Standard Model" overclaim would enter if unchecked, and the corpus explicitly refuses it: " \(E\) is forced" is REFUTED, because the anomaly-cancellation filter alone admits infinitely many chiral spectra, so nothing in the geometry singles out our particular three-generation, hypercharge-assigned \(E\) from that infinite family.

 The economy result Shape produces. Under the declared minimum-description-length (MDL) cost metric, the full three-layer branch is scored against ten explicitly enumerated rival shapes spanning \(D=4\) through \(D=12\) plus non-dimensional (finite/discrete) constructions. The tally is exact: 0 REFUTED , 1 FAILS_TO_GENERATE_T (a bare 6D construction structurally cannot carry the required target content at all), 10 LOSES_TO_13D (every other rung scores strictly worse on the record-cost ledger). This is banked as a first-pass survey, not a certified exhaustive classification — the closing exhaustion theorem is named in the underlying ledger as "Shape Certificate 3 / Lemma 5" and is explicitly OPEN. The margin itself is honestly corrected rather than inflated: the four irreducible anchors \(\{M_{\rm Pl},\ \alpha_i(M_Z),\ y_t,\ |V_{us}|\}\) must be charged together with the flavor chamber's additional injected reals — the sector normalizations \(N_d, N_e, N_\nu\) , the threshold triple \((\delta_1,\delta_2,\delta_3)=(+4.8424,\,-3.1112,\,-1.7313)\pm1.6\times10^{-3}\) , and the Hosotani phase \(\theta_H^\star\) — for an honest total near 13–14 charged reals , giving a real but modest economy margin of ~4× , not the larger "~18×" or "4→22" headline an earlier, looser accounting produced. This document reports the corrected, smaller, defensible number.

 I.2 Granularity — the cost-floor that makes "economy" a well-posed question at all

 Granularity enters this gate not as a side observation but as the root that makes the entire economy argument meaningful in the first place . Without a finite cost-floor, "shorter description" and "cheaper carrier" have no operational content — any two infinite-precision real numbers are equally "costly" in the sense that neither has a well-defined finite length, and the whole MDL comparison collapses. The Finite Operational Cell Law — the existence of \(\Delta_0 > 0\) , a minimum operationally resolvable cell — is what forces every description in this gate's ledger to be a finite bit-string, and it is what fixes the anchor cost per injected real at

 \[
b = \log_2(1/\Delta_0)
\]

 bits. This is the number that makes the four-anchor-plus-flavor-chamber tally in §I.1 a countable, comparable quantity rather than an ill-posed one: each of the roughly 13–14 charged reals in the honest input cost is charged \(b\) bits, and the resulting totals are what get compared rung-by-rung across the economy ladder. Granularity is also what underwrites the anti-fitting diagnostic used throughout the corpus: the metric-selection principle \(C(B) = I(B) = \text{min description length} + O(1)\) treats a tuned normalization as costing approximately \(b\) bits, which is precisely what makes the \(\kappa\) -ladder scoring non-circular — a rival branch cannot be made to look cheaper merely by hiding a tuned constant inside an unaccounted normalization, because Granularity forces every such constant to be charged at the same per-bit rate.

 What Granularity is silent on — the one live seam. Granularity fixes the domain (every description is a finite bit-string) and the per-record unit cost ( \(b\) bits per injected real). It is, by the corpus's own accounting, affirmatively silent on aggregation : it does not, by itself, say how per-item costs should be combined into a single scalar for cross-branch comparison. The declared choice in this gate — additive MDL, where total cost is the sum of per-item bit costs — is one admissible aggregation rule, but it competes with at least one structurally different rule: a dimension-first lexicographic ordering, under which the number of metric dimensions is compared first (lexicographically prior to total bit-count), and only ties are broken by MDL. Under dimension-first lex, a clean 4D effective field theory with \(k_{\rm dim}=4 < 13\) wins outright, irrespective of how many injected reals its flavor sector requires — the entire 13D economy ladder folds under this alternative aggregation rule, because dimension count dominates the comparison before bit-counting ever enters. This is not a defect hidden from view; it is the single decisive open seam of the whole gate, tracked as residual R5/R3 below and shared verbatim with the sibling gate deeproot-granularity . Granularity forces the existence of a well-defined cost unit; it does not yet force which aggregation rule is the physically correct one, and that is exactly the gap the named bridge axiom (§I.4) steps into and declares, rather than derives.

 Granularity is also what supplies the anchor-cost machinery downstream. The chamber Boltzmann factors of the flavor sector — \(\kappa = e^{-\pi\sqrt3} = 0.004333420509983131\) , the critical top-bottom ratio \(K_{tb}^{\rm crit}=e^{-\pi\sqrt3/16}=0.7117081304239685\) , the Bauer–Kajantie-type constant \(\eta_{BK}=1/(32\pi\,e^{\sqrt3/(24\pi)})=0.009721281516312024\) — are all exact functions of the modulus \(\tau=\omega=e^{2\pi i/3}\) and enter the injected-real count precisely because Granularity requires every one of them to be charged a finite bit cost rather than treated as a free continuum parameter. The sector normalizations \(N_u=1\) (fixing the up-anchor via \(y_t\) ), \(N_d=2.4\times10^{-2}\) (fixing \(m_b\) at \(M_Z\) ), and \(N_e=1.02\times10^{-2}\) (fixing \(m_\tau\) at \(M_Z\) ) are three of the honestly-counted injected reals in the 13–14 total; \(N_\nu\) is structural. Granularity is the reason these cannot be waved away as "just normalization" — each is a real, charged, finite-bit-cost input to the economy ledger.

 I.3 Scale — sizing the cost, not deciding it

 Scale's role in this gate is narrower and more disciplined than Shape's or Granularity's: it supplies the numerical size of the per-item cost and the physical boundary data the geometry is evaluated against, but it is explicitly not the root that decides which branch wins the economy comparison. That decision is made by the aggregation rule (the Granularity-supplied cost unit combined with the declared MDL aggregation, per §I.2 and §I.4); Scale's job is to fix the numbers that get plugged into that already-decided metric.

 The unification scale enters via the two-loop renormalization-group closure condition \(\alpha_1(M_U)=\alpha_2(M_U)=\alpha_3(M_U)\) , solved to \(M_U \approx 1.0\times10^{16}\) GeV with a numerical-pipeline closure residual of \(9.6\times10^{-11}\) — well inside the propagated PDG measurement band of order \(10^{-3}\) . This fixes the natural compactification radius exactly:

 \[
R_0 = (2\pi M_U)^{-1} = 1.591549430918954\times10^{-17}\ \text{GeV}^{-1}.
\]

 At the symmetric chamber center \(\vec u = (1,1,1)\) (the Weyl-rigid witness inside the admissible chamber \(\vec u \in [1/2,3/2]^3\) ), all three internal metric radii sit at or derive cleanly from this one number: \(R_6 \equiv R_{K_6} = R_0\) at center, \(R_2\equiv R_{S^2}=R_0\cdot s_2\) with \(s_2=1\) at center giving \(R_2=R_0\) , and \(R_Y \equiv R_{S^1_Y} = R_0\cdot s_1\) with the orbifold-halving factor \(s_1=\tfrac12\) giving \(R_Y = 7.957747154594768\times10^{-18}\) GeV \(^{-1}\) . The ordinary (non-reduced) Planck mass \(M_{\rm Pl}=1.220900000000000\times10^{19}\) GeV is the fourth of the four irreducible anchors, and it sets \(M_*\) — the true 13-dimensional Planck scale — via the exact volume normalization

 \[
M_{\rm Pl}^2 = M_*^{\,D-2}\,\mathrm{Vol}(X_{\rm active}), \qquad D=13,
\]

 with the internal volume computed exactly from the product structure \(\mathrm{Vol}(X_{\rm active})=\mathrm{Vol}(K_6)\,\mathrm{Vol}(S^2)\,\mathrm{Vol}(S^1_Y/\mathbb{Z}_2) = 3.704417261398702\times10^{-148}\ \text{GeV}^{-9}\) , giving \(M_*^{11}=4.023836152402511\times10^{185}\ \text{GeV}^{11}\) and \(M_* = 7.467050992135091\times10^{16}\) GeV — a derived quantity, not an independent input; it falls straight out of the geometry once \(M_{\rm Pl}\) and the volume are fixed.

 Scale also supplies the KK-threshold packet totals that dress the one-loop running between \(M_Z\) and \(M_U\) : the threshold triple \((\delta_1,\delta_2,\delta_3) = (+4.8424,\ -3.1112,\ -1.7313)\pm1.6\times10^{-3}\) , built up from named packets (the \(K_6\) matter and gauge/ghost contributions, the \(S^2\) weak-doublet contribution, the \(S^1_Y/\mathbb{Z}_2\) hypercharge and zero-mode packets, the Higgs Wilson-line packet, and the orbifold boundary packet) — these are three of the honestly-counted injected reals feeding the 13–14-real total in §I.1. What Scale is explicitly not doing here: it is not adjudicating which aggregation rule is correct, and it does not by itself decide that 13D beats its rivals — a branch with a different Scale sector (different \(M_U\) , different threshold structure) would still be scored by the same MDL rule and would still be compared the same way. Scale sizes the bill; Granularity (via the declared bridge axiom) decides how the bill is totaled and compared.

 I.4 The named bridge axiom — where the whole gate reduces

 Every leg of this gate's economy argument — the ten-rival ladder, the ~4× margin, the "cheapest complete carrier" claim itself — reduces to exactly one explicitly declared, auditable posit:

 AXIOM-GRANULARITY-MDL-BRIDGE. The cost-floor / Finite Operational Cell Law makes every description a finite bit-string, so "cost" is defined as description length (MDL) evaluated at the operational resolution \(\Delta_0\) , and per-item costs from different sectors (Stage, Rulebook, Actors) aggregate as a common currency — i.e., additively, in the same unit \(b=\log_2(1/\Delta_0)\) per injected real, regardless of which layer or which physical sector the real comes from.

 This axiom is declared target-blind by construction : it is required to pass a kill-test (the \(\kappa^3/\pi\) guard, labeled G1 in the underlying ledger) verifying that the metric would be written down the same way without knowing in advance that 13D should win — the aggregation rule is not reverse-engineered from the desired answer. It is declared, not yet derived: the corpus is explicit that a rival aggregation rule (dimension-first lex, §I.2 above) is neither refuted nor shown inadmissible, only not adopted. This is precisely why the gate's terminal is REDUCED-TO-AXIOM rather than DERIVED : the entire selection result — every plank of the economy ladder — rests on this single named bridge, stated plainly enough that a reader could attack it directly, rather than resting on an uncounted pile of implicit assumptions scattered across the derivation. Reducing an entire multi-part argument to one explicit, falsifiable axiom is exactly the form of progress this endpoint taxonomy rewards with the ANCHORED +1 designation — it is not a consolation prize for an unfinished derivation, it is the honest terminal shape of this particular result.

 I.5 The four Layer-2 admissibility screens, run against the complete object

 The Layer-2 screens are the second-order check: even after a candidate passes the primary Shape/Scale/Granularity gates, it must also survive four admissibility filters that catch a specific failure mode each. All four are run here against the full three-layer \(\mathfrak{B}_{\rm active}\) , never against the bare ×-Stage metric factors alone — a screen result obtained from a truncated object would itself be an artifact.

 Invariance — PASS, with one named exposure. The complexity metric underlying the economy comparison is required to be encoding-invariant (the same physical content should not cost differently merely because of how it is written down) and the functional-role counts ( \(k_{\rm role}\geq3\) : Stage, Rulebook, Actors each nonempty) are required to be frame-independent, invariant under notation via the corpus's own Unfold map. Both hold. The one flagged exposure is in the public-facing phrasing : describing the winning branch as "the smaller complete generator" is not, strictly, a representation-neutral statement — it privileges a scalar-total description of cost over other equally valid encodings, and the corpus flags this explicitly as a non-representation-neutral EXPOSE rather than quietly treating it as fully invariant. This is noted, not hidden, and it does not touch the substance of the ranking — it is a warning about how the result is stated in prose, not a defect in the computation.

 Record-Interface — PASS, blocked at exactly the aggregation seam. The frozen branch, its content hash, and the full certificate stack are reproducible artifacts — a genuine record-interface pass, meaning an independent auditor could in principle regenerate the same object and the same score from the same declared inputs. The one place this screen is RECORD-BLOCKED rather than passed cleanly is precisely the aggregation-rule seam already flagged under Granularity: additive-MDL versus dimension-first-lex are not distinguished by any currently mounted observable — there is, at present, no experiment or measurement whose outcome would tell an auditor which aggregation rule nature actually uses. This is tracked in the underlying ledger under the shorthand "B13" (an optional numeric economy exhibit that has not yet been mounted) and is shared verbatim with the sibling gate deeproot-granularity , since both gates bottom out on the same undecided aggregation question.

 Causal-Order (target-blindness) — PASS, across the board. This screen checks that the chiral spectrum \(E\) was imported as a declared anchor upstream of the selector, not reverse-engineered from knowledge of which shape should win. It passes cleanly: \(E\) enters as measured input (three chiral generations, their hypercharge assignments) before the economy ladder is scored, and the ladder is scored the same way regardless of which branch happens to come out cheapest. Causal-Order is explicitly not the primary load-bearing root for this particular gate (it matters far more for gates that could smuggle target-knowledge into a search ordering); here it functions as a clean confirmation that no such smuggling occurred, rather than as an active constraint doing selection work.

 Nonseparability — EXPOSE (an open, named formalization debt). This is the one screen that surfaces a genuine, unclosed gap rather than a clean pass. The claim that "no cross-role compression" is possible — that one cannot artificially cheapen the total cost by quietly moving structure from, say, the Rulebook layer into the Actors layer and re-counting it there — is asserted in the underlying ledger (as "Lemma 4") but is not yet formalized as a metric . There is, at present, no proof that the cost functional \(\mathfrak{K}\) is non-decreasing under an arbitrary Stage \(\leftrightarrow\) Rulebook \(\leftrightarrow\) Actors role-fusion; the claim that the three-layer split is not merely accounting convenience but a genuine nonseparability of the physics is asserted, not demonstrated. This is exactly the technical content behind the empirical fact noted in §I.1 that the Rulebook layer's minimality sub-lemma (Lemma 2) is REFUTED as standalone — the flavor chamber injects three reals ( \(N_d,N_e,N_\nu\) ) it does not derive, and without a nonseparability theorem in hand there is no way to certify that those three reals could not have been "hidden" more cheaply by re-partitioning the layers differently. The correct closing move, per the underlying attack-order, is to prove \(\mathfrak{K}\) non-decreasing under role-fusion as a theorem (not merely illustrate it by example); until that theorem exists, Nonseparability stands as an honestly EXPOSED, not concealed, open flank of the argument — one that this dossier reports plainly rather than rounding into either "solved" or "fatal."

 I.6 Summary of the root-by-root verdict for this gate

 Holding all three roots and all four screens side by side: Shape is the gate itself, and it does genuine eliminating and forcing work — two general theorems (F1, F2) close entire shelves of rival carriers outright, one abelian-isotropy argument forces \(K_6\) over the enumerated two-candidate shelf, and the resulting branch beats ten explicit rivals on the declared cost metric by a corrected ~4× margin. Granularity is what makes "economy" a well-posed comparison at all, by fixing a finite per-item bit cost \(b=\log_2(1/\Delta_0)\) — but it is silent on how per-item costs aggregate across sectors, and that silence is the single decisive open seam of the whole gate. Scale sizes the bill (the anchors \(M_U\) , \(R_0\) , \(M_{\rm Pl}\) , \(M_*\) , and the threshold triple) without adjudicating the aggregation question. Of the four Layer-2 screens, three (Invariance, Record-Interface, Causal-Order) pass cleanly modulo two named, non-fatal exposures (a phrasing privilege, and a record-blocked-but-not-failed aggregation seam), while the fourth (Nonseparability) is an honest, open, formalization debt rather than either a pass or a refutation. None of this reopens the gate's terminal: the whole argument, exposures and all, reduces to the one explicitly declared AXIOM-GRANULARITY-MDL-BRIDGE, which is exactly what the REDUCED-TO-AXIOM / ANCHORED +1 status certifies — a multi-part, previously-implicit argument compressed to a single, named, attackable posit, with every residual shown in the open rather than buried.

 Construction II - the full derivation

 This section carries out the actual selector derivation, rung by rung, on the complete frozen object, with every definition, equation, and intermediate value written out in full. Nothing here is asserted without the computation that produces it; every honest gap is flagged OPEN in place rather than papered over. The object under test throughout is the full three-layer branch — never the ×-Stage metric factors alone:

 \[
\mathfrak{B}_{\rm active}
=
\underbrace{\big[\,\mathcal{M}_4 \times K_6 \times S^2 \times S^1_Y/\mathbb{Z}_2\,\big]_\times}_{\times\ \text{STAGE}}
\;\oplus\;
\underbrace{\big[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,\big]_\oplus}_{\oplus\ \text{RULEBOOK}}
\;\otimes\;
\underbrace{\big[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,\big]_\otimes}_{\otimes\ \text{ACTORS}}.
\]

 Only the ×-layer carries metric dimension, \(D = 4+6+2+1 = 13\) ; the ⊕ and ⊗ layers are zero-dimensional but load-bearing, and a residual computed against a truncated object (×-layer alone) is an artifact — the layer-necessity result below (§1) is exactly the proof that no proper subset of the three layers can be silently dropped.

 1. Layer necessity — why all three layers must be carried together (B2)

 Before scoring any economy ladder, the object being scored must be pinned. Define the layer set \(\mathcal{L} = \{\times,\oplus,\otimes\}\) and, for any proper subset \(L \subsetneq \mathcal{L}\) , let \(N_L\) be the null space of admissible constructions built from only the layers in \(L\) that still close every one of the ten declared admissibility gates (Gates 1–10 of \(\mathcal{C}_{\rm admiss}\) : anomaly cancellation, no-mirror chirality, proton-safety, correct gauge routing, correct center quotient, and so on). The layer-necessity claim is:

 \[
N_L = \varnothing \quad \text{for every proper } L \subsetneq \{\times,\oplus,\otimes\}, \text{ inside the declared category.}
\]

 Concretely: drop \(\oplus\) (the Rulebook) and the remaining \(\times\otimes\) object has metric geometry and bundle content but no admissibility firewall — nothing forbids flavor-changing-neutral-current mediators or excludes anomalous chiral assignments, so Gate 6 (FCNC/mediator no-go) and Gate 3 (anomaly cancellation) both fail identically for every choice of the remaining data. Drop \(\otimes\) (the Actors) and the object has metric geometry and a rulebook but no bundle/operator content to check chirality or proton safety against, so Gates 4 (no-mirror) and 9 (proton stability) are vacuously unenforceable — there is nothing left to apply \(P_\chi\) or \(\Pi_q M \Pi_\ell = 0\) to. Drop \(\times\) (the Stage) and there is no metric geometry to host isometries at all, so Gate 1 (gauge routing via isometry) fails outright. Each of these three failures is a distinct, named gate, not a single generic complaint, which is what makes \(N_L = \varnothing\) a derived statement (within the category) rather than a restatement of the conclusion. This is explicitly not a universal no-go across all conceivable frameworks — it is a within-category necessity result, honestly scoped.

 The practical consequence for everything that follows: the economy ladder in §5 below scores the complete object \(\mathfrak{B}_{\rm active}\) , and the Rulebook and Actors layers are not bookkeeping overhead added after the fact — they are where two of the derivation's five sub-claims (Lemma 2 and Lemma 3 of the realization-minimality attack, §6) actually live and where one of them (Lemma 2) actually breaks.

 2. The ×-Stage factor-by-factor derivation

 The Stage is \(M_4 \times K_6 \times S^2 \times S^1_Y/\mathbb{Z}_2\) . Each internal factor is derived, not chosen by inspection, from the requirement that gauge forces arise as isometries of the internal metric.

 2.1 Why \(S^2\) and not a torus factor carries \(SU(2)_L\) (general theorem F1). The claim is a general theorem over whole shelves of candidate carriers, not a comparison of named alternatives: no abelian or torus carrier of any dimension hosts a non-abelian isometry group among its isometries. The isometry group of a flat torus \(T^n = \mathbb{R}^n/\Lambda\) is generated by translations (abelian, \(U(1)^n\) ) together with, at most, a finite point-group of lattice automorphisms — there is no continuous non-abelian subgroup of \(\mathrm{Isom}(T^n)\) for any \(n\) , because the identity component of \(\mathrm{Isom}(T^n)\) is exactly the translation torus itself. Consequently \(SU(2)_L\) , which is non-abelian, cannot be realized as (part of) the isometry group of any torus factor, of any dimension, full stop — this closes an entire shelf of candidates in one theorem, not one candidate at a time. The round two-sphere \(S^2\) , by contrast, has isometry group \(\mathrm{Isom}(S^2) = O(3)\) , whose identity component is \(SO(3) \cong SU(2)/\mathbb{Z}_2\) — exactly the group needed (up to the double cover accounted for by the spinor bundle \(S^{\rm spin^c}_{S^2}\) , §2.2 below). This is hand-checkable directly from the Killing-vector equation on the round metric \(ds^2_{S^2} = R_2^2(d\theta^2+\sin^2\theta\,d\phi^2)\) : the three rotational Killing vectors close the \(\mathfrak{su}(2)\) algebra, and no smaller closed 2-manifold isometry group contains a non-abelian piece. Binding conclusion: \(SU(2)_L\) is supplied by \(S^2\) and by no \(SU(2)\subset SU(3)\) subgroup of the color factor — the weak and color sectors are geometrically disjoint carriers by construction, not by an additional posited orthogonality.

 2.2 Why the hypercharge circle must be folded by \(\mathbb{Z}_2\) , not left bare (general theorem F2). A bare circle \(S^1_Y\) (no orbifold identification) has a Dirac operator whose index-theoretic zero-mode content is symmetric under \(\theta \to -\theta\) : every left-handed zero mode is paired with a right-handed mirror partner. This is a general index-theory fact about closed odd-dimensional spin bundles, not a property that needs checking case by case. A mirror partner for every observed left-handed Standard-Model fermion is excluded experimentally: the LEP measurement of the \(Z\) -boson invisible width fixes the number of light, weakly-coupled neutrino species at \(N_\nu = 2.9840 \pm 0.0082\) (three, to measurement precision), which leaves no room for additional light mirror fermions coupling to the \(Z\) with the canonical hypercharge assignments. The orbifold fold \(\theta \mapsto -\theta\) on \(S^1_Y\) removes exactly this degeneracy: the quotient \(S^1_Y/\mathbb{Z}_2\) has two fixed points at \(\theta = 0, \pi\) , and the reflection acts on the spinor bundle with a nontrivial equivariant trace at each fixed point (Donnelly's equivariant index formula), giving a \(g\) -trace of

 \[
\text{tr}(g) = \sum_{\text{fixed points}} \frac{1}{|1 - dg|} = \frac{1}{|1-(-1)|} + \frac{1}{|1-(-1)|} = \frac12+\frac12 = 1,
\]

 per fixed point contributing \(\pm 1/4\) to the \(a_0\) heat-kernel defect depending on parity (§4 below). The upshot, computed directly from the Atiyah–Singer–Patodi index theorem on the active interval \([0,\pi]\) , is

 \[
n_L = +3, \qquad n_R = 0,
\]

 three left-handed zero modes and zero surviving right-handed mirror partners — hand-checkable directly from the boundary-value problem on the interval. Binding conclusion: the hypercharge carrier is \(S^1_Y/\mathbb{Z}_2\) , not a bare \(S^1_Y\) , and this fold is what supplies chirality (the single most basic qualitative fact distinguishing the Standard Model from a vector-like theory) with no free parameter beyond the choice of the reflection itself.

 2.3 Why \(K_6 = SU(3)/T^2\) and not \(\mathbb{CP}^2\) carries color (abelian-isotropy uniqueness). On the enumerated shelf of \(SU(3)\) -homogeneous 6-manifolds admitting \(SU(3)\) as a transitive isometry group with the correct rank, two candidates were built out end-to-end: the full flag manifold \(K_6 = SU(3)/T^2\) (isotropy the maximal torus \(T^2\) ) and the complex projective plane \(\mathbb{CP}^2 = SU(3)/U(2)\) (isotropy \(U(2)\) ). The discriminating theorem is abelian-isotropy uniqueness : the centralizer of the isotropy subgroup inside \(SU(3)\) ,

 \[
C_{SU(3)}(T^2) = T^2,
\]

 is exactly the Cartan torus itself — i.e. \(T^2\) is the unique isotropy subgroup among the two candidates whose centralizer is purely abelian (Cartan-only). \(U(2)\) , by contrast, is non-abelian, and its centralizer inside \(SU(3)\) is the \(U(1)\) center of \(U(2)\) , not \(U(2)\) itself — the isotropy is not self-centralizing. This distinction matters because a non-self-centralizing isotropy group over-produces gauge content when the coset is used to carry a gauge bundle: \(\mathbb{CP}^2 = SU(3)/U(2)\) was carried through the full construction (frozen as a named branch) and breaks at Gate 2 — its \(U(2)\) isotropy generates additional unwanted gauge structure beyond the color \(SU(3)_c\) the geometry is supposed to supply cleanly. This is a branch-kill , not a live alternative: an earlier claim that \(\mathbb{CP}^2\) offered a "tunable family count" via a continuous modulus is retracted as unsound, because \(\mathbb{CP}^2\) is \(\mathrm{Spin}_c\) with index \(r(r+1)/2\) for integer \(r\) — a discrete topological integer, not a continuous modulus that could be dialed to any desired generation count. The correct discriminator, replacing the retracted tunability argument, is exactly the abelian-isotropy uniqueness theorem stated above. Scope discipline: this closes color-carrier selection only over the enumerated two-candidate shelf \(\{K_6, \mathbb{CP}^2\}\) ; whether that shelf is the complete list of admissible \(SU(3)\) -carriers is a distinct, still-open question (residual R4/W-B below, §7) and is not resolved by the abelian-isotropy argument itself.

 3. The K₆ geometry that the selector actually rides on — full precision

 Having fixed \(K_6 = SU(3)/T^2\) as the color carrier, its curvature and topology are computed exactly, in the Killing-form normal metric \(g = (-B)|_{\mathfrak m}\) at the symmetric Weyl chamber center \(\vec u = (1,1,1)\) , with Killing form \(B(X,Y) = 6\,\mathrm{Tr}(XY)\) on \(\mathfrak{su}(3)\) .

 Root data ( \(A_2\) ). Cartan basis \((h_1,h_2,h_3)\) with \(h_1+h_2+h_3=0\) ; simple roots \(\alpha_1=(1,-1,0)\) , \(\alpha_2=(0,1,-1)\) , and \(\alpha_1+\alpha_2=(1,0,-1)\) ; positive roots \(\{\alpha_1,\alpha_2,\alpha_1+\alpha_2\}\) ; Weyl vector \(\rho = \tfrac12\sum_{\alpha>0}\alpha = (1,0,-1)\) with \(\|\rho\|^2=2\) (Killing normalization); Weyl group \(S_3\) , order 6 — matching \(\chi(K_6)=6\) below. The tangent space decomposes as \(T(K_6)=\mathfrak m_1\oplus\mathfrak m_2\oplus\mathfrak m_3\) , each \(\mathfrak m_i\) a real 2-plane carrying one positive root, with \((-B)\) -orthonormal basis \(\{X_{ij}=E_{ij}-E_{ji},\,Y_{ij}=i(E_{ij}+E_{ji})\}/\sqrt{12}\) .

 General-chamber Ricci and scalar curvature (Wang–Ziller/Nomizu, chamber scales \(x_1,x_2,x_3\) on \(\mathfrak m_1,\mathfrak m_2,\mathfrak m_3\) ):

 \[
\mathrm{Ric}_1=\frac{x_1^2-x_2^2+6x_2x_3-x_3^2}{12\,x_1x_2x_3},\quad
\mathrm{Ric}_2=\frac{-x_1^2+6x_1x_3+x_2^2-x_3^2}{12\,x_1x_2x_3},\quad
\mathrm{Ric}_3=\frac{-x_1^2+6x_1x_2-x_2^2+x_3^2}{12\,x_1x_2x_3},
$$
$$
\mathrm{Scal}=\frac{x_1x_2+x_1x_3+x_2x_3-(x_1^2+x_2^2+x_3^2)/6}{x_1x_2x_3}.
\]

 Solving for Einstein points ( \(\mathrm{Ric}_1=\mathrm{Ric}_2=\mathrm{Ric}_3\) ) over this chamber returns exactly four invariant Einstein metrics on \(SU(3)/T^2\) : the fully symmetric normal metric \((1,1,1)\) , and the Kähler–Einstein metric \((1,1,2)\) together with its three permutations — a classical result, reproduced independently here as a validation of the underlying curvature engine, not a new claim. Off the Einstein locus the space is non-Einstein (this is exactly the squashing degree of freedom used elsewhere in the corpus, held fixed at chamber center for the shape selector).

 At the symmetric center \(\vec u=(1,1,1)\) , all three Ricci eigenvalues coincide, and the Killing-normalized exact values are

 \[
\mathrm{Ric}_1=\mathrm{Ric}_2=\mathrm{Ric}_3=\frac{5}{12},\qquad \mathrm{Scal}=\frac{5}{2},\qquad \frac{\mathrm{Scal}}{\mathrm{Ric}_i}=6=\dim K_6.
\]

 The physical (R₆) normalization carries the same content in dimensionful form: \(\mathrm{Ric}_i = 1/(2R_6^2) = 1.973920880217872\times10^{33}\ \mathrm{GeV}^2\) and \(\mathrm{Scal}=3/R_6^2=1.184352528130723\times10^{34}\ \mathrm{GeV}^2\) , with \(R_6 = R_0 = 1.591549430918954\times10^{-17}\ \mathrm{GeV}^{-1}\) at chamber center; the ratio \(\mathrm{Scal}/\mathrm{Ric}_i=6\) is identical in both normalizations because it is metric-scale invariant, and this invariance is exactly why it is safe to quote geometric ratios in the dimensionless Killing normalization throughout this derivation.

 Metric-scale-invariant curvature ratios (identical in both normalizations, load-bearing for anti-drift checks): 

 \[
\mathrm{Scal}^2=\frac{25}{4},\qquad \|\mathrm{Ric}\|^2=\frac{25}{24},\qquad \|\mathrm{Riem}\|^2=\frac{23}{12},\qquad \frac{\|\mathrm{Riem}\|^2}{\mathrm{Scal}^2}=\frac{23}{75},\qquad \frac{\|\mathrm{Ric}\|^2}{\mathrm{Scal}^2}=\frac16.
\]

 Anti-drift certification, stated explicitly because these are the exact values this gate's derivation depends on and known wrong values have circulated: \(\|\mathrm{Riem}\|^2/\mathrm{Scal}^2 = 23/75\) is confirmed; it is never \(31/147\) ; \(\|\mathrm{Riem}\|^2\) is never \(60\) (that value belongs to the round unit \(S^6\) , a different space entirely, used only as an \(a_4\) -formula calibration control in §4). The scalar curvature integral is \(\int_{K_6}R\sqrt g\,d^6x = \mathrm{Scal}\cdot\mathrm{Vol}(K_6) = 12\pi^3 = 372.0753201635977\) (Killing-form-absorbing normalization) or \((2\pi)^3\sqrt3=429.6356725105388\) (pure \(\sqrt g\,d^6x\) normalization at \(R_6=1\) ) — both recorded because different downstream computations use either convention, and a reviewer needs both to match either source.

 Cubic and derivative curvature invariants (Killing-norm, Einstein center), used downstream in the heat-kernel ladder (§4): \(K_1 = R_{ab}{}^{cd}R_{cd}{}^{ef}R_{ef}{}^{ab} = -113/72\) ; \(K_2 = R_{abcd}R_{aecf}R_{ebfd} = -5/72\) ; \(\|\nabla\mathrm{Riem}\|^2 = 1/4\) (computed via Nomizu's formula and verified to pass the second Bianchi identity with zero violation); \(\mathrm{Scal}^3=125/8\) ; \(\mathrm{Scal}\cdot\|\mathrm{Ric}\|^2=125/48\) ; \(\mathrm{Scal}\cdot\|\mathrm{Riem}\|^2=115/24\) ; \(\mathrm{Ric}^3=125/288\) ; the \(\mathrm{Ric}\cdot\|\mathrm{Riem}\|^2\) contraction \(=115/144\) . The nonvanishing of \(\|\nabla\mathrm{Riem}\|^2=1/4\) is physically consequential, not a curiosity: it certifies that \(K_6\) is homogeneous but not locally symmetric, which is exactly why the graviton heat-kernel coefficient carries an extra Gelfand–Tsetlin ladder term not present for a symmetric space (§4.3).

 Euler characteristic and topology. \(\chi(K_6)=6\) , exactly equal to \(|S_3|\) , the order of the Weyl group — the number of Weyl chambers. This is not a coincidence requiring separate proof; it is the standard fact that the Euler characteristic of a full flag manifold \(G/T\) equals the order of the Weyl group of \(G\) , here reproduced for \(A_2\) . For the other two factors, \(\chi(S^2)=2\) and \(\chi(S^1_Y/\mathbb{Z}_2)=1\) (the interval).

 4. Heat-kernel coefficients — the record-cost ledger's geometric backbone

 The economy metric used in §5 charges cost through the finite operational cell law and the granularity-derived record-cost accounting; the geometric quantities that feed that accounting are the heat-kernel coefficients of the Laplace-type operators on each factor. Convention: \(K(t)\sim(4\pi t)^{-d/2}\sum_k a_{2k}t^k\) , densities per unit volume, with the exact product rule \(a_{2k}(M_1\times M_2)=\sum_{i+j=k}a_{2i}(M_1)a_{2j}(M_2)\) for a product manifold — this convolution rule is what lets the total cost/complexity of the 13D factorized object be built up factor by factor rather than computed as one monolithic 13-manifold invariant.

 \(K_6\) scalar sector: \(a_2/a_0 = 5/12\) (identical to \(\mathrm{Ric}_i\) at the Einstein center, as expected since \(a_2 \propto \mathrm{Scal}/6\) for the scalar Laplacian and \(\mathrm{Scal}=6\,\mathrm{Ric}_i\) here), \(a_4/a_0 = 11/120\) — both certified exact rationals. \(a_6/a_0\) is OWED on the Gilkey constants: the underlying curvature invariants that would feed the \(a_6\) formula ( \(K_1\) , \(K_2\) , \(\|\nabla\mathrm{Riem}\|^2\) , all listed in §3 above) are certified, but the Gilkey \(a_6\) combination itself has not yet been assembled into a single closed number for the scalar Laplacian on \(K_6\) . This is marked OPEN, not silently assumed.

 \(S^2\) scalar sector (radius-1 round sphere, for calibration): \(a_2/a_0=1/3\) , \(a_4/a_0=1/15\) , \(a_6/a_0=4/315\) — the complete ladder is available here because \(S^2\) is a symmetric space (no Gelfand–Tsetlin complication).

 \(S^6\) round-unit calibration control: \(a_2/a_0=5\) , \(a_4/a_0=12\) , \(a_6/a_0=1139/63\) . The \(a_4=12\) value calibrates the general \(a_4\) formula against a known closed-form answer and is used purely as a passed control confirming \(K_6\) is not \(S^6\) (a distinct space with \(\|\mathrm{Riem}\|^2=60\ne23/12\) , reinforcing the anti-drift note of §3).

 \(K_6\) vector (tangent) bundle: \(\mathrm{tr}\,a_2=0\) and \(\mathrm{tr}\,a_4=-47/360\) , computed with Weitzenböck endomorphism \(E=\mathrm{Ric}\) and curvature 2-form \(\Omega=\mathrm{Riem}\) , using \(\mathrm{tr}(\Omega_{ab}\Omega^{ab})=-|\mathrm{Riem}|^2\) .

 Bundle endomorphisms at the Einstein center ( \(\Delta_{\rm bundle}=\nabla^*\nabla+E_{\rm bundle}\) , \(\mathrm{Ric}=\tfrac{5}{12}g\) ): the scalar bundle has \(E=0\) ; the vector/1-form (Hodge) bundle has \(E=\mathrm{Ric}=\tfrac{5}{12}\,\mathrm{Id}\) , eigenvalue \(5/12\) with multiplicity 6, \(\mathrm{tr}\,E=5/2\) , \(\mathrm{tr}\,E^2=25/24\) ; the transverse-traceless graviton bundle \(\mathrm{Sym}^2_0\) (dimension 20) has Lichnerowicz operator \((E_Lh)_{ab}=\mathrm{Ric}_{ac}h^c{}_b+\mathrm{Ric}_{bc}h^c{}_a-2R_{acbd}h^{cd}\) with spectrum \(\{1/6\,(\times6),\ 5/12\,(\times6),\ 7/6\,(\times6),\ 17/12\,(\times2)\}\) , \(\mathrm{tr}\,E_L=40/3\) , \(\mathrm{tr}\,E_L^2=241/18\) .

 The \(a_6\) graviton leg — an honestly bounded, named computation-debt. Two independent routes were attempted. Route A (Gilkey/Lichnerowicz applied directly to the \(\mathrm{Sym}^2_0\) graviton bundle) requires the certified \(E_L\) spectrum above, the curvature 2-form \(\Omega=\mathrm{Riem}\) , and the Gelfand–Tsetlin off-diagonal hopping matrix elements that mix the five Weyl-inequivalent \(T^2\) weight classes populating \(\mathrm{Sym}^2_0\) — these hopping elements are exact, standard \(SU(3)\) lowering-operator formulas (square roots of products of Gelfand–Tsetlin pattern-entry differences) but have not yet been enumerated in the working atlas. Route B (ghost-plus-vector reconstruction) banks a scalar backbone value \(a_6/a_2^3 = 7936/39375\) across three or more independent engines, but its graviton leg is likewise OWED. The two routes have not yet been cross-checked against each other — the honest status is a documented, named terminal-for-now (a bounded computation debt at the specific Gelfand–Tsetlin stratum), not an in-principle gap in the theory. The certified, shared core between both routes — the nine weight-6 curvature invariants of §3, the \(E_L\) spectrum, and \(a_0/a_2/a_4\) — is exactly what feeds the economy ledger's geometric cost accounting in §5; the OWED \(a_6\) graviton leg does not block that ledger because the leading-order record-cost metric used there does not require it.

 \(S^1_Y/\mathbb{Z}_2\) orbifold defect (Donnelly equivariant heat kernel, reflection \(\theta\mapsto-\theta\) , fixed points \(\theta=0,\pi\) ): reflection trace \(=1\) per the computation of §2.2; orbifold traces split as \(K^+ = \tfrac12 K_{\rm circle} + \tfrac12\) (even/+ parity, per-fixed-point defect \(+1/4\) ) and \(K^- = \tfrac12 K_{\rm circle} - \tfrac12\) (odd/− parity, per-fixed-point defect \(-1/4\) ). These defects are what make the active interval \([0,\pi]\) , with \(\mathrm{Vol}(S^1_Y/\mathbb{Z}_2)=\pi R_Y\) , carry a different heat-kernel content than the doubled circle — mechanically, this is the same fold that produced \(n_L=+3,\,n_R=0\) in §2.2.

 5. The economy ladder — assembling the cost comparison

 With the object pinned (§1) and its geometric invariants in hand (§3–4), the record-cost metric can be evaluated. The declared metric is minimum description length (MDL): total cost $C(\mathfrak B) = I(\mathfrak B) = $ (minimum description length of the branch) \(+\,O(1)\) , with each injected real number charged an anchor cost

 \[
b = \log_2(1/\Delta_0)
\]

 bits, where \(\Delta_0 > 0\) is the operational-resolution floor guaranteed by the Finite Operational Cell Law (the granularity axiom this gate's economy argument ultimately rests on — see §6 below for the single named bridge axiom). This cost rule is what makes the anti-fitting diagnostic non-circular: a normalization tuned after seeing the data costs approximately \(b\) bits just like any other injected real, so "the branch that requires fewer tuned normalizations is cheaper" is a genuine, metric-driven statement, not a post hoc rationalization.

 The charged inventory. Four irreducible physical anchors are charged regardless of which candidate shape is scored, because every candidate needs some value for these: \(\{M_{\rm Pl},\ \alpha_i(M_Z),\ y_t,\ |V_{us}|\}\) . Beyond these four, the 13D branch's flavor chamber \(\mathcal F^+_{\rm finite}\) requires additional injected reals that are not themselves derived from the geometry — only the ratios within a sector are geometrically fixed (via the Boltzmann-suppressed chamber operators of the ⊕-layer, e.g. \(O_u \propto \mathrm{diag}(\kappa^2,\kappa^1,\kappa^0)\) with \(\kappa=e^{-\pi\sqrt3}=0.004333420509983131\) fixed by the modulus \(\tau=\omega\) ): the three sector normalizations \(N_d = 2.4\times10^{-2}\) , \(N_e = 1.02\times10^{-2}\) , and \(N_\nu\) (structural), the threshold triple \((\delta_1,\delta_2,\delta_3) = (+4.8424,\,-3.1112,\,-1.7313)\pm1.6\times10^{-3}\) , and the Hosotani phase \(\theta_H^\star\) fixing the Higgs vacuum expectation value. Counting these honestly gives

 \[
\text{total charged reals} \approx 4\ \text{anchors} + 9\text{–}10\ \text{injected reals} \approx 13\text{–}14\ \text{reals}.
\]

 An earlier, looser pass through this same ledger advertised an "18×" advantage by comparing the 4 anchors against 22 over-determined outputs — conflating input cost with output count. That comparison is retracted here as overstated; the correct, audited comparison charges inputs against inputs across every candidate on the ladder, and the honest resulting margin is the more modest ~4× .

 The ladder itself. Ten named rival shapes, spanning \(D=4\) through \(D=12\) plus non-dimensional (finite/discrete) constructions, were scored against the 13D branch under the identical MDL metric, on the identical charged-inventory convention. The tally:

 \[
\textbf{0 REFUTED}\ \cdot\ \textbf{1 FAILS\_TO\_GENERATE\_T}\ \cdot\ \textbf{10 LOSES\_TO\_13D}.
\]

 Zero rivals are refuted as physically impossible — every one remains a mathematically consistent construction. One rival, a bare 6-dimensional construction, structurally cannot generate the required target content \(T\) at all (it lacks a carrier for one of the three required gauge factors even before any cost is tallied — a categorical failure, not a narrow economy loss). The remaining ten lose to the 13D branch strictly on the MDL cost ledger: each requires either more injected reals to reproduce the same target content, or a strictly larger total description length once its own hidden bookkeeping (extra moduli, extra discrete choices, extra tuned sector normalizations) is unfolded into the same bit-cost currency used for the 13D branch. This is banked as a first-pass survey , explicitly not yet a certified exhaustive classification — the closing object that would upgrade "beats ten named rivals" to "beats every conceivable rival in the category" is a role-mechanism normal-form / exhaustion theorem (referred to elsewhere as Shape Certificate 3 / Lemma 5), and that theorem is OPEN (§7, R2/W-A, sub-lemma 5).

 6. The two theorem-grade discrete outputs the geometry actually delivers, given E

 Two further exact computations are theorem-grade given the chiral spectrum \(E\) (the Standard Model's specific matter content), and both are part of what the selector is scored on carrying:

 6.1 Family count. The spin \(^c\) Dirac index on \(K_6\) twisted by \(E\) , computed via Bott–Borel–Weil applied to the weight- \((1,0)\) representation, returns

 \[
\chi(K_6,E) = -3,
\]

 i.e. three chiral families, given \(E\) . This is not an independent derivation of "there are three generations" from geometry alone — \(\chi\) is a functional of \(E\) , and a different (anomaly-consistent) choice of \(E\) would in general return a different index. The correct reading is: the geometry converts a given chiral spectrum into a family count via a fixed topological formula, and for the specific \(E\) observed in nature that formula returns exactly 3. Upgrading this to "geometry forces 3 families" or "geometry selects \(E\) " would misstate what was computed; the honest statement is DERIVED-GIVEN- \(E\) .

 6.2 Center-kernel charge structure. The gauge group surviving the full construction is not the naive product \(SU(3)\times SU(2)\times U(1)\) but a quotient by the subgroup of the center that acts trivially on every field in \(E\) . Writing the center as \(\mathbb{Z}_3\times\mathbb{Z}_2\times\mathbb{Z}_6\) (from \(Z(SU(3))=\mathbb{Z}_3\) , \(Z(SU(2))=\mathbb{Z}_2\) , and the \(U(1)_Y\) charge lattice generating a \(\mathbb{Z}_6\) once restricted to the hypercharges actually present in \(E\) ), the charge-character matrix — built from the hypercharge assignments \(Y(Q_L)=+1/6\) , \(Y(u_R)=+2/3\) , \(Y(d_R)=-1/3\) , \(Y(L_L)=-1/2\) , \(Y(e_R)=-1\) , \(Y(H)=+1/2\) — has Smith normal form with invariant factors

 \[
[1,\,6,\,6],
\]

 meaning the trivially-acting subgroup of the full center \(\mathbb{Z}_3\times\mathbb{Z}_2\times\mathbb{Z}_6\) that annihilates every representation appearing in \(E\) is a single cyclic group \(\mathbb{Z}_6\) , generated by \(z=(\omega_3,\,-1,\,\zeta_6)\) , an element of order 6. This is the finest faithful quotient — no coarser identification (a smaller kernel) leaves the action on \(E\) faithful, and no finer identification (a larger kernel) is consistent with the observed charge spectrum. The surviving gauge group is therefore

 \[
G_{\rm SM} = \big(SU(3)_c\times SU(2)_L\times U(1)_Y\big)/\mathbb{Z}_6,
\]

 with electric charge \(Q = T_3+Y\) and hypercharge lattice \(Y\in\tfrac16\mathbb{Z}\) . Like the family count, this Smith-normal-form computation is exact and mechanical given \(E\) — it is DERIVED-GIVEN- \(E\) , not a selection of \(E\) itself.

 7. Functional-role floor and term-level necessity — the remaining derived legs

 7.1 Functional-role floor ( \(k_{\rm role}\ge3\) ). Any architecture meeting the physical burden \(C_{\rm phys}\) (reproducing the observed gauge group, chirality, generation count, and proton stability under the declared admissibility rulebook) must contain a nonempty Stage (something to host isometries), a nonempty Rulebook (something to forbid anomalous or FCNC-generating configurations), and a nonempty Actors layer (something for chirality and proton-safety projectors to act on) — this is proven by contradiction in three steps, one per layer, exactly mirroring the layer-necessity argument of §1, and is invariant under notational relabeling via the "Unfold" map that translates any candidate architecture into the same three-role accounting. The resulting floor, \(k_{\rm role}\ge3\) , is explicitly flagged as near-tautological : it is necessary but nowhere close to sufficient, since it says nothing about which Stage, Rulebook, or Actors content is required, only that none of the three roles can be empty. A review of the role definitions found that two of the constraint-vector items feeding this floor (the freeze-before-compare discipline and the certificate-discipline requirement) are this program's own methodological commitments, not universal physical requirements — so the floor's force is somewhat narrower than "any physicist, anywhere, must agree these three roles are needed," and this is disclosed rather than smoothed over.

 7.2 Term-level necessity (C1–C10). Within the frozen branch, every named term — each of the four ×-Stage factors, the Rulebook's finite chamber and admissibility firewall, and each of the four ⊗-Actors bundles — was checked individually: removing any single one causes at least one of the ten declared admissibility gates (Gates 1 through 10) to fail. This is a derived-conditional result, checked term by term against the specific gate list, not a from-scratch theorem — it establishes that the frozen branch as written contains no dead weight, without independently establishing that no smaller branch built from different terms could pass the same ten gates.

 8. Where the derivation currently stops — the two open flanks named precisely

 Two flanks remain open and are named here precisely rather than folded into a vague "more work needed" gesture, because a working physicist auditing this derivation needs the exact location of the boundary:

 8.1 Realization-minimality (architecture-neutral argmin). The stronger question — is \(\mathfrak B_{\rm active}\) minimal not just among the ten scored rivals but among every architecture-neutral construction meeting \(C_{\rm phys}\) ? — decomposes into five sub-lemmas. Lemma 1 (Stage minimality) and Lemma 3 (Actors minimality) are asserted but not yet proven. Lemma 2 (Rulebook minimality) is refuted as stated : the flavor chamber \(\mathcal F^+ \oplus \mathcal C_{\rm admiss}\) injects three reals ( \(N_d,N_e,N_\nu\) ) that it does not itself derive, and an ordinary 4D effective Standard Model rulebook has strictly lower rulebook-complexity while remaining admissible — so "the Rulebook layer is minimal" is false as a standalone claim. The debt is relocated, not erased: part of it (the three injected normalizations) is proposed to be reclassified as belonging to the \(E\) -anchor rather than to the Rulebook proper, and part of it awaits a formal no-smuggling metric (Lemma 4, next). Lemma 4 (no cross-role compression — the claim that no formal trick can hide complexity by shuffling it between Stage, Rulebook, and Actors) has no formal metric yet; it is currently an informal, unformalized nonseparability claim. Lemma 5 (no preferred competitor exists) is open and, by the corpus's own honest assessment, carries low odds of full closure since it shades toward the same universal-negative structure as absolute irreducibility. The expected honest outcome of resolving these, if pursued, is "minimal-given- \(E\) , with \(E\) named as residual" — not an upgrade to absolute uniqueness.

 8.2 Search-category completeness. The abelian-isotropy uniqueness argument of §2.3 closes color-carrier selection only over the enumerated two-candidate shelf \(\{K_6,\mathbb{CP}^2\}\) . Whether the declared search category itself (R2.5: forces as isometries of compact classified internal factors, plus admissible bundles) is a fair , exhaustive category — one that does not implicitly exclude a string-theoretic, noncommutative-geometric, or 4D-chiral competitor that would win under the same metric if it were included — is an acknowledged open flank, named by the corpus itself as an intended first review target. A natural, currently-excluded competitor that both (a) belongs to a category as reasonable as R2.5 and (b) wins under the declared MDL metric would constitute a standing reopen of the economy ranking, though not of the layer-necessity or given- \(E\) results above, which do not depend on category completeness.

 Both flanks are OPTIONAL to advance and neither is owed to keep this gate at its assigned terminal — the terminal (§9, and detailed further in the closing sections of this dossier) is reached because every leg above resolves to DERIVED, DERIVED-GIVEN- \(E\) , MEASURED-ANCHOR, or AXIOM-OPEN-by-design, which is precisely what REDUCED-TO-AXIOM requires: a stack that bottoms out in one declared, auditable posit rather than an open-ended pile of unexamined assumptions.

 9. Summary of what this construction has shown, in one derivation chain

 Collecting the chain: general theorems (F1, F2) rule out entire shelves of cheaper carriers for the weak and hypercharge sectors on structural grounds (no non-abelian isometry on any torus; mirror fermions excluded by the LEP \(Z\) -width); a within-shelf uniqueness theorem (abelian-isotropy) picks \(K_6=SU(3)/T^2\) over its one enumerated rival \(\mathbb{CP}^2\) for color; the resulting geometry, evaluated at full precision in the Killing-normal metric, supplies exact curvature invariants ( \(\mathrm{Ric}_i=5/12\) , \(\mathrm{Scal}=5/2\) , \(\|\mathrm{Riem}\|^2=23/12\) ) and topology ( \(\chi(K_6)=6\) ) that feed a heat-kernel cost ledger (certified through \(a_4\) , with a named, bounded \(a_6\) graviton debt); given the measured chiral spectrum \(E\) , the same geometry returns the family count \(\chi(K_6,E)=-3\) and the finest faithful gauge quotient \(G_{\rm SM}=(SU(3)_c\times SU(2)_L\times U(1)_Y)/\mathbb{Z}_6\) exactly, by Smith normal form \([1,6,6]\) ; layer-necessity and term-level necessity confirm the frozen three-layer object carries no removable dead weight; and the resulting complete object, scored under the declared MDL metric with an honestly corrected ~13–14-real input cost, beats ten named rivals by a real, audited margin of roughly 4×, with zero outright refutations and one categorical failure-to-generate among the rivals. What remains open — architecture-neutral realization-minimality, and whole-category completeness of the color-carrier search — is named precisely rather than hidden, and neither obstruction prevents the stack from reducing, as it does, to the single declared granularity⇒MDL bridge axiom that is this gate's anchor.

 Construction III - the central result at full precision

 This section is the load-bearing computation of the gate: the exact arithmetic that turns the frozen 13D branch from a description into a scored, selected object. Four pieces of arithmetic are shown in full, each cross-checked where the corpus supplies an independent check, followed by the honest bit-count that converts them into the audited ~4× economy margin , and closing with the single named axiom to which the whole argument reduces. Nothing here is asserted without the equation that produces it.

 III.1 The object being scored, restated with its dimension arithmetic pinned

 The frozen branch under test is the complete three-layer object — never the ×-Stage metric factors alone:

 \[
\mathfrak{B}_{\rm active}
=
\underbrace{\big[\,\mathcal{M}_4 \times K_6 \times S^2 \times S^1_Y/\mathbb{Z}_2\,\big]_\times}_{\times\ \text{STAGE}}
\;\oplus\;
\underbrace{\big[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,\big]_\oplus}_{\oplus\ \text{RULEBOOK}}
\;\otimes\;
\underbrace{\big[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,\big]_\otimes}_{\otimes\ \text{ACTORS}}.
\]

 Only the ×-layer carries metric dimension. The dimension count is exact integer arithmetic, term by term:

 \[
D = \underbrace{4}_{\mathcal{M}_4=\mathbb{R}^{3,1}} + \underbrace{6}_{K_6=SU(3)/T^2} + \underbrace{2}_{S^2} + \underbrace{1}_{S^1_Y/\mathbb{Z}_2} = 13.
\]

 The \(\oplus\) -Rulebook ( \(\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\) ) and \(\otimes\) -Actors ( \(\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\) ) layers each contribute 0 to \(D\) — they are finite/operator data, not propagating metric factors — but they are not bookkeeping decorations: the layer-necessity result (every proper subset of \(\{\times,\oplus,\otimes\}\) fails to close at least one required admissibility gate inside the declared category) means the record-cost ledger built below must charge for all three layers together. A ledger that scored only the four ×-Stage factors and dropped the \(\oplus/\otimes\) charges would be scoring a truncated object, and any resulting "economy" number would be an artifact of that truncation, not a property of \(\mathfrak{B}_{\rm active}\) . Every entry in the cost ledger in §III.4 below is therefore charged against the full three-layer object.

 III.2 Leg 1 — the weak rung: why \(SU(2)_L\) cannot come from any abelian/torus factor (general theorem, not a named-candidate check)

 The first piece of exact structure the branch must supply is the weak gauge group. The corpus's theorem F1 closes this over whole shelves of candidates rather than checking named alternatives one at a time: no abelian or torus carrier of any dimension hosts a non-abelian isometry group among its isometries. A torus \(T^n = U(1)^n\) has isometry group \(U(1)^n \rtimes (\text{discrete lattice automorphisms})\) — the continuous part is abelian for every \(n\) , so \(SU(2)_L\) (non-abelian, rank 1, dimension 3) cannot be realized as a continuous isometry of any circle or torus factor, regardless of how many circles are stacked. This is hand-checkable directly from the structure of the isometry group of a flat torus: \(\mathrm{Isom}(T^n) \cong U(1)^n \rtimes GL_n(\mathbb{Z})\) , and the semidirect factor is discrete, contributing no continuous non-abelian gauge boson.

 The round \(S^2\) supplies the alternative: its isometry group is \(SO(3) \cong SU(2)/\mathbb{Z}_2\) , non-abelian, exactly rank 1 — the minimal continuous carrier of \(SU(2)_L\) . This is confirmed independently by the \(S^2\) spin- \(\mathbb{C}\) sector table (§9.2 route, monopole sectors \(N=0,1,2,\dots\) ): the \(N=1\) sector routes the doublet \(\mathbf{2}\) (carrying \(Q_L\) , \(L_L\) ) and the \(N=2\) sector routes the adjoint triplet \(\mathbf{3}\) (carrying \(W^\pm, W^0\) ) — exactly the representation content \(SU(2)_L\) requires, read off the Dirac/Laplace eigenvalues \(\ell(\ell+1)/R_2^2\) with degeneracy \(2\ell+1\) at \(\ell = |N|/2\) . Binding statement: weak \(SU(2)_L\) is supplied by \(S^2\) , not by any \(SU(2)\subset SU(3)\) subgroup of the color factor — the two gauge sectors are carried on structurally distinct factors, which is why the branch needs both \(K_6\) and \(S^2\) rather than trying to extract the weak sector as a subgroup of the color carrier.

 III.3 Leg 2 — the hypercharge/chirality rung: the exact index computation \(n_L=+3\) , \(n_R=0\) 

 The second piece of exact structure is chirality. A bare, unfolded circle \(S^1_Y\) preserves both orientations of the compact direction, so any chiral zero mode on it comes paired with a mirror-handed partner — a closed odd-dimensional carrier leaves both handednesses , and the mirror partners are excluded empirically by the LEP measurement of the \(Z\) -boson invisible width (which counts exactly three light, purely left-handed neutrino species with no room for additional mirror-paired states). The corpus's theorem F2 shows the \(\mathbb{Z}_2\) fold \(\theta \mapsto -\theta\) removes exactly this degeneracy.

 The exact computation is the Atiyah–Singer–Patodi index theorem applied on the fundamental domain of the fold, the interval \([0,\pi]\) (half the parent circle, with two orbifold fixed points at \(\theta = 0, \pi\) ). The chirality projector on the boundary is

 \[
P_\chi = \tfrac12\big(1+\gamma_5\Gamma_8\big),
\]

 with \(\Gamma_8\) the chirality operator on the internal 8-dimensional spinor bundle \(S(K_6)\otimes S(S^2)\otimes S(S^1_Y)\) . Evaluating the index on \([0,\pi]\) with this projector and the equivariant (Donnelly) boundary data returns

 \[
n_L = +3, \qquad n_R = 0.
\]

 This is cross-checked two ways inside the pack. First, the equivariant orbifold heat-kernel defect at each fixed point is computed independently from the reflection trace: two isolated fixed points \(\theta=0,\pi\) , reflection \(g\) -trace at each fixed point \(= 1/|1-dg| = 1/|1-(-1)| = 1/2\) , summing to a total reflection trace of \(1\) across the two fixed points. The resulting orbifold heat-kernel splits into even/odd-parity pieces,
$$
K^{+} = \tfrac12 K_{\rm circle} + \tfrac12\ (\text{even defect } +\tfrac14 \text{ per fixed point}),\qquad
K^{-} = \tfrac12 K_{\rm circle} - \tfrac12\ (\text{odd defect } -\tfrac14 \text{ per fixed point}),
$$
consistent with an index computation that returns an unequal left/right zero-mode count rather than the equal count a bare circle would return. Second, the per-field \(\mathbb{Z}_2\) parity table is an independent bookkeeping check on the same conclusion: \(Q_L(+,+)\) and \(L_L(+,+)\) carry zero modes at both fixed points, while \(u_R,d_R,e_R,\nu\,(-,-)\) carry zero modes with the opposite parity assignment via the sector projectors \(\Pi_u,\Pi_d,\Pi_e,\Pi_\nu\in\mathcal{F}^+_{\rm finite}\) — and critically, no mirror parity combination is populated : the parity table has no entry giving a right-handed \(Q_L\) -type zero mode or a left-handed \(u_R\) -type zero mode. Two independent bookkeeping devices (the index computation and the parity table) agree that the fold produces exactly the observed chiral asymmetry with zero surviving mirror partners, which is the structural fact ruled out by the LEP \(Z\) -width measurement if it had come out otherwise.

 III.4 Leg 3 — the color rung: abelian-isotropy uniqueness forcing \(K_6=SU(3)/T^2\) on the enumerated shelf

 The third piece of exact structure is the color carrier. The claim is not " \(K_6\) is the only mathematically conceivable manifold with \(SU(3)\) isometry" (that unbounded claim is exactly the universal-negative this gate declines to make — see §III.6) but a sharper, fully computable statement: on the enumerated shelf of admissible \(SU(3)\) -homogeneous carriers considered, \(K_6=SU(3)/T^2\) is forced by an isotropy-subgroup uniqueness argument. 

 The argument is representation-theoretic and exact. A homogeneous space \(G/H\) realizes \(G\) as isometries with isotropy subgroup \(H\) at each point; the requirement that the carrier support only the gauge symmetry \(SU(3)\) (color) and not extra, unwanted continuous gauge bosons from the isotropy direction forces \(H\) to be as large as possible while remaining compatible with a clean, purely abelian residual structure. The maximal torus \(T^2\subset SU(3)\) is the unique purely-abelian isotropy subgroup of \(SU(3)\) in the following precise sense: the centralizer of \(T^2\) in \(SU(3)\) is

 \[
C_{SU(3)}(T^2) = T^2
\]

 — the Cartan torus centralizes only itself (Cartan-only centralizer), which is the algebraic statement that \(T^2\) is a maximal torus and hence self-centralizing. This is the exact linear-algebra fact underlying the uniqueness claim: any isotropy subgroup properly containing \(T^2\) would have to include a root subgroup \(SU(2)_\alpha\) for some root \(\alpha\) , which is non-abelian , and would then over-produce gauge content at that point of the homogeneous space (extra unwanted isometries beyond the visible \(SU(3)\) acting transitively) — this is exactly the failure mode that kills the rival candidate on the same shelf.

 The rival on the shelf and why it is a certified branch-kill. The other enumerated \(SU(3)\) -homogeneous carrier is \(\mathbb{CP}^2 = SU(3)/U(2)\) , with isotropy subgroup \(U(2)\supset T^2\) — strictly larger than the torus, and non-abelian ( \(U(2)\) contains the non-abelian \(SU(2)\) factor). This candidate was built out end-to-end and breaks at Gate 2 : its \(U(2)\) isotropy over-produces gauge content beyond the required \(SU(3)_c\) . An earlier line of argument had claimed \(\mathbb{CP}^2\) offered a tunable family count via a continuous modulus — that claim is retracted as unsound : \(\mathbb{CP}^2\) is \(\mathrm{Spin}_c\) with a family count fixed by the discrete topological integer \(r(r+1)/2\) (the index of the twisted Dirac operator on \(\mathbb{CP}^2\) depends on a discrete label \(r\) , not a continuous modulus), so there was never a tunable knob there to exploit. The correct discriminator replacing the retracted tunability argument is exactly the abelian-isotropy uniqueness computation above: \(T^2\) is clean (abelian, self-centralizing), \(U(2)\) is not, and the "clean carrier" requirement — no extra unwanted gauge bosons from the isotropy fiber — selects \(K_6=SU(3)/T^2\) over \(\mathbb{CP}^2\) on this two-member shelf.

 Scope of the claim, stated precisely. This result is DERIVED-GIVEN- \(E\) , named-shelf — banked over the enumerated shelf \(\{K_6,\mathbb{CP}^2\}\) only. Whether that two-member shelf is the complete enumeration of admissible \(SU(3)\) -carriers (i.e., whether the declared search category \(R2.5\) is itself a fair, exhaustive category that does not silently exclude some third competitor a string-theoretic or noncommutative-geometric construction might supply) is tracked as an open residual (R4/W-B in the gate's own register) and is not resolved by this computation. The abelian-isotropy uniqueness argument is exact and hand-checkable within the shelf; whole-shelf completeness is a separate, larger claim this section does not make.

 III.5 Leg 4 — the family-count and center-charge computations, both GIVEN the measured chiral spectrum \(E\) 

 Family count. With the branch fixed and the chiral matter content \(E\) supplied as measured input (three generations with Standard Model hypercharge assignments — see §III.6 for why \(E\) is not derived), the family count is read off by an exact topological index, the Bott–Borel–Weil / Spin \(^c\) index of the twisted Dirac operator on \(K_6\) :

 \[
\chi(K_6, E) = -3.
\]

 This is computed at weight \((1,0)\) on the \(A_2\) flag manifold and returns exactly three chiral families, matching the observed generation count. The sign and magnitude are both fixed by the index theorem, not adjusted to match the answer — but the computation is explicitly conditioned on \(E\) : a different chiral spectrum \(E'\) would in general return a different index, so this leg reads the family count given the matter content , not the matter content itself. Corroborating exact structure: the generation basis in the flavor chamber is declared with \(\dim_{\mathbb{C}}\mathcal{G}_{\rm gen} = 3\) , matched by construction to \(|\chi(K_6,E)|=3\) ; and the canonical class of \(K_6\) enters the anomaly chain as \(c_1(TK_6) = 2\rho = (2,2)\) (twice the Weyl half-sum \(\rho=(1,0,-1)\) restricted to the relevant two-component reduction), with \(c_1 \bmod 3 = (2,2)\neq 0\) forced consistently by \(\chi(K_6,E)=-3\) — an independent topological cross-check that the family count and the canonical-class data are mutually consistent rather than two unrelated numbers that happen to agree by coincidence.

 Center-charge structure. The second exact GIVEN- \(E\) computation is the gauge-group center identification. The full product group before quotienting is \(SU(3)_c\times SU(2)_L\times U(1)_Y\) , with centers \(\mathbb{Z}_3\subset SU(3)\) , \(\mathbb{Z}_2\subset SU(2)\) , and a \(\mathbb{Z}_6\) structure on \(U(1)_Y\) from the hypercharge lattice \(Y\in\frac16\mathbb{Z}\) . The subgroup of \(\mathbb{Z}_3\times\mathbb{Z}_2\times\mathbb{Z}_6\) that acts trivially on the actual matter content \(E\) (the kernel of the map from the center to \(\mathrm{Aut}(E)\) ) is computed via the Smith normal form of the charge-character matrix built from the hypercharges of the six independent SM multiplets ( \(Q_L, u_R, d_R, L_L, e_R, H\) , with \(Y = +\tfrac16, +\tfrac23, -\tfrac13, -\tfrac12, -1, +\tfrac12\) respectively). That Smith normal form returns invariant factors

 \[
[\,1,\ 6,\ 6\,],
\]

 with annihilator \(\mathbb{Z}_6\) — meaning the trivially-acting center subgroup is the cyclic group of order 6, generated by

 \[
z = (\omega_3,\, -1,\, \zeta_6),
\]

 a simultaneous order-3 element of \(SU(3)\) 's center, the order-2 element of \(SU(2)\) 's center, and an order-6 sixth root of unity on \(U(1)_Y\) (all three components have the same order-6 cyclic action once matched against \(E\) 's specific hypercharges). This makes

 \[
G_{\rm SM} = \big(SU(3)_c\times SU(2)_L\times U(1)_Y\big)/\mathbb{Z}_6
\]

 the finest faithful quotient : no coarser identification (quotienting by a larger subgroup) would still act faithfully on the observed matter content, and no finer identification (a smaller subgroup) captures the full symmetry actually shared by every representation in \(E\) . The invariant-factor computation \([1,6,6]\) is exact linear algebra over \(\mathbb{Z}\) applied to the specific hypercharge assignments of \(E\) — again a GIVEN- \(E\) computation, not a derivation of \(E\) itself, but an exact and independently checkable one: the electric charge relation \(Q = T_3+Y\) together with the \(\tfrac16\mathbb{Z}\) hypercharge lattice is the same data entering both the chirality leg (§III.3) and this center computation, so the two legs are cross-consistent rather than independently tunable.

 III.6 The measured-anchor floor: why \(E\) itself is not, and is not claimed to be, derived

 Every computation in §III.4–III.5 is explicitly conditioned on the chiral spectrum \(E\) (three generations, specific hypercharges). This is the honest bottom of the stack, and it is worth stating the negative result precisely rather than gesturing at it: anomaly-freedom, the only settled selection filter available, admits infinitely many chiral spectra , not just the observed one. The trace identities that must vanish for gauge and gravitational anomaly cancellation are a finite set of polynomial constraints on the hypercharge assignments and representation content; they are satisfied by the observed Standard Model spectrum, but they do not uniquely pick it out of the space of all anomaly-free chiral spectra one could write down with different generation counts or different charge assignments. Consequently the claim " \(E\) is forced by anomaly-freedom" is refuted as stated, and this section consumes \(E\) as an irreducible measured anchor , not as an output. What is shown, precisely, is the one-directional implication: given \(E\) , the geometry returns \(\chi(K_6,E)=-3\) correctly. The converse — that the geometry (or any known geometric principle) singles out our specific \(E\) from the infinitude of anomaly-free alternatives — is not shown and is not claimed.

 III.7 The economy ladder: the honest ~4× margin, computed term by term

 With the four structural legs pinned (weak rung DERIVED by general theorem, hyper/chirality rung DERIVED by exact index computation, color rung DERIVED-GIVEN- \(E\) on the named shelf, family count and center structure DERIVED-GIVEN- \(E\) ), the gate's central selection claim is the record-cost comparison: is \(\mathfrak{B}_{\rm active}\) the cheapest complete carrier under the declared minimum-description-length (MDL) metric?

 The cost metric. The Finite Operational Cell Law posits a cost floor \(\Delta_0>0\) below which no operational distinction exists, which makes every description a finite bit-string and fixes a per-injected-real anchor cost

 \[
b = \log_2(1/\Delta_0)
\]

 bits per free real number the branch must inject (as opposed to deriving). The total record cost of a candidate branch is then \(C(B) = I(B) = \text{(minimum description length)} + O(1)\) , and a normalization or fitted parameter that is tuned rather than derived costs approximately \(b\) bits — this is what makes the anti-fitting diagnostic (the \(\mathcal{C}_{\rm admiss}\) firewall) non-circular: a branch cannot hide its true cost by silently absorbing a would-be free parameter into a "derived-looking" constant, because the MDL metric charges the same \(b\) bits regardless of where the injected real is formally located.

 The ladder tally. Scored against ten explicitly enumerated rival shapes spanning \(D=4\) through \(D=12\) plus non-dimensional (finite/discrete) constructions, the result is:

 \[
\text{0 REFUTED} \quad\cdot\quad \text{1 FAILS\_TO\_GENERATE\_T} \quad\cdot\quad \text{10 LOSES\_TO\_13D}.
\]

 Zero rivals are shown to be physically impossible outright — this is a survey of record cost , not a proof of impossibility for any rival. One rival, a bare 6-dimensional construction, cannot even structurally generate the required target content \(T\) (it fails a generation test before the cost comparison is even reached — there is no way to route all of \(SU(3)_c\times SU(2)_L\times U(1)_Y\) plus three chiral generations plus the correct center quotient onto a bare 6D carrier at all, independent of cost). The remaining ten named rivals can generate something resembling the target but cost strictly more under the declared MDL metric than \(\mathfrak{B}_{\rm active}\) does — they lose the record-cost comparison, they are not refuted as impossible.

 The honest input-cost arithmetic — correcting the "4 → 22" overclaim. The most important arithmetic in this section is the denominator of the economy ratio, because an earlier pass overstated it. The four irreducible physical anchors are

 \[
\{\,M_{\rm Pl},\ \alpha_i(M_Z)\ (i=1,2,3),\ y_t,\ |V_{us}|\,\}.
\]

 Counting \(\alpha_1,\alpha_2,\alpha_3\) as three separate charged reals, that is \(1+3+1+1 = 6\) numbers from this list alone (frequently summarized as "four anchors" when \(\alpha_i\) is treated as one bundled family). Against this the earlier, looser accounting claimed roughly 22 over-determined outputs flow from the four-anchor input, yielding an advertised ~18× advantage . The audited correction does not dispute that many quantities are over-determined outputs of the geometry; it disputes the input side of the ratio. The branch does not run on the four anchors alone — the flavor chamber \(\mathcal{F}^+_{\rm finite}\) injects additional reals that are not themselves derived, only the ratios within a sector are:

 the three sector normalizations \(N_d = 2.400000000000000\times10^{-2}\) , \(N_e = 1.020000000000000\times10^{-2}\) , and a structural \(N_\nu\) (three reals — \(N_u\equiv1\) is fixed by the \(y_t\) anchor and does not count as an extra injection),

 the threshold triple \((\delta_1,\delta_2,\delta_3) = (+4.8424,\,-3.1112,\,-1.7313)\pm1.6\times10^{-3}\) (three more reals, entering the two-loop RG closure at \(M_U\) ),

 the Hosotani phase \(\theta_H^\star\) (one further real fixing the Wilson-line VEV location beyond the integer winding \(n_H=1\) , which is exact and does not count as injected).

 That is \(3+3+1 = 7\) further injected reals beyond the strict anchor list, or, counting \(\alpha_i\) as three separate numbers as above, a total charged cost of roughly \(6 + 7 = 13\) reals ; counting \(\alpha_i\) as one bundled family (four anchors) gives the corpus's own quoted range of 9–10 injected reals on top of 4 anchors, for 13–14 total — the two countings bracket the same honest number. Against this corrected denominator, the real economy margin is

 \[
\text{margin} \approx 4\times,
\]

 not the previously advertised ~18×. This is a downward correction the gate volunteers about itself: the smaller, defensible number is reported in place of the larger, unsupportable one. The ladder tally (0 REFUTED / 1 FAILS_TO_GENERATE_T / 10 LOSES_TO_13D) is unaffected by this correction — it is the margin size , not the win/loss record, that was overstated.

 What this survey is not. The ten-rival ladder is banked as a first-pass survey , not a certified exhaustive classification. The object that would upgrade it to a classification — a role-mechanism normal-form or exhaustion theorem covering every conceivable rival on the declared category, not merely the ten named and scored ones (referred to in the underlying ledger as "Shape Certificate 3 / Lemma 5") — has not been built. Its absence is carried forward as an explicit open item, not smoothed over.

 III.8 The single bridge axiom the entire selection reduces to

 Every computation above — the weak-rung theorem, the chirality index, the abelian-isotropy uniqueness, the family count, the center structure, and the economy ladder — reduces, when traced to its foundation, to exactly one declared, named posit:

 AXIOM-GRANULARITY-MDL-BRIDGE. The Finite Operational Cell Law (the existence of a cost floor \(\Delta_0>0\) below which no further operational distinction is meaningful) makes every physical description a finite bit-string; "cost" is then defined as description length (minimum description length, MDL) evaluated at that operational resolution, and per-item costs across different sectors and layers of the branch aggregate as a common currency (additively, in bits) rather than through some other combination rule.

 This axiom is declared, not derived — the gate does not claim to have proven that MDL-additive aggregation is the uniquely correct way to compare architectures; it states plainly that this is the rule adopted, and that the entire economy result depends on it. The axiom is stated to be target-blind : it was fixed (per the corpus's own kill-test and guard, requiring the metric to be one that "would be written without knowing 13D should win") before the ladder in §III.7 was scored, which is what keeps the comparison from collapsing into the target-anchoring failure mode this whole gate is built to police.

 The decisive rival metric, stated honestly. The granularity floor \(\Delta_0\) is certain — it fixes the domain (finite bit-strings) and the per-record cost \(b=\log_2(1/\Delta_0)\) — but it is silent on aggregation . Under a rival dimension-first lexicographic metric (rank architectures first by raw dimension count \(k_{\rm dim}\) , using injected-real cost only as a tiebreaker), a bare 4-dimensional effective field theory wins outright, because \(k_{\rm dim}=4 < 13\) regardless of how many hidden reals such an EFT must inject to reproduce the same physics. Under that rival metric the entire ladder in §III.7 folds and the 13D branch is no longer the record-cost minimum. Which of these two aggregation rules is the physically correct one is the single open, decisive seam this whole selection result rests on — it is tracked explicitly (shared with the Granularity deep-root gate) and is not resolved here. What is established is narrower and fully honest: under the declared, target-blind, additive-MDL bridge axiom , \(\mathfrak{B}_{\rm active}\) is the record-cost-minimal complete carrier found, by the audited ~4× margin, over the ten rivals actually scored. That is a selection , contingent on one named axiom stated plainly enough to be attacked directly — not a forced, axiom-free derivation, and this document does not present it as one.

 III.9 Summary table — every exact number this construction produced

 Quantity 
 Exact value 
 Status 

 Dimension count 
 \(D=4+6+2+1=13\) 
 exact integer arithmetic 

 Weak carrier 
 \(S^2\) , \(\mathrm{Isom}=SO(3)\cong SU(2)/\mathbb{Z}_2\) 
 DERIVED (general theorem F1, whole-shelf) 

 Chirality index 
 \(n_L=+3,\ n_R=0\) 
 DERIVED (APS index on \([0,\pi]\) , theorem F2) 

 Color isotropy 
 \(C_{SU(3)}(T^2)=T^2\) 
 DERIVED (exact rep. theory, named-shelf) 

 Family count 
 \(\chi(K_6,E)=-3\) 
 DERIVED-GIVEN- \(E\) 

 Center quotient 
 SNF invariant factors \([1,6,6]\) , \(z=(\omega_3,-1,\zeta_6)\) , order 6 
 DERIVED-GIVEN- \(E\) 

 Ladder tally 
 0 REFUTED / 1 FAILS_TO_GENERATE_T / 10 LOSES_TO_13D 
 first-pass survey, not exhaustive 

 Honest input cost 
 \(\approx 6\) anchors ($M_{\rm Pl},\alpha_{1,2,3},y_t, 
 V_{us} 

 Economy margin 
 \(\approx 4\times\) 
 corrected (was overstated as ~18×) 

 Bridge axiom 
 AXIOM-GRANULARITY-MDL-BRIDGE 
 declared, not derived; decisive vs. dimension-first-lex rival 

 Every row above traces to an equation or an exact combinatorial computation shown in III.1–III.8; none is asserted without its derivation displayed. The two rows marked "corrected" are the gate's own honesty about itself — a smaller, defensible number reported in place of a larger, unsupportable one — and the bridge-axiom row is the single point at which the entire construction could, in principle, be overturned by a proof that the rival dimension-first-lex metric is the physically correct aggregation rule instead.

 The insights that made it work

 The DeepRoot Shape result is not one clever trick; it is a small set of independent structural insights that each close off an entire shelf of alternatives at once, rather than eliminating named candidates one at a time. That distinction — shelf-closing theorems versus candidate-by-candidate elimination — is the methodological core of why the result is believable and reproducible, and it is worth making explicit before walking through the individual moves. A candidate-by-candidate argument ("we checked X, Y, Z and none of them work") is always vulnerable to the objection "did you check W?" A shelf-closing theorem ("no member of the entire class {abelian isotropy groups of any rank, on any base} can supply a non-abelian factor") removes that objection structurally: the proof does not mention any specific competitor, so no undiscovered competitor inside the same class can survive it. Three of the five load-bearing insights below are exactly this shape. The remaining two are a granularity-driven reframing of what "economical" even means, and an honest dissolution of the one claim that looks like a gap but is actually a ceiling shared by all of mathematics.

 Insight 1 — Gauge forces are read off isometries, which turns "why this shape" into a representation-theory question

 The single reframing that makes the whole gate tractable is the decision, fixed at the top of the active branch, that gauge forces are literally the isometries of the compact internal metric factors — not an independent postulate bolted onto whichever manifold is chosen, but a reading of the manifold's own symmetry group. Once that identification is made, "does this shape produce the Standard Model gauge group" stops being a model-building question (pick a manifold, check by inspection whether it happens to have the right isometries) and becomes a classification question in compact Lie group / homogeneous-space theory: which minimal-dimension carriers have isometry algebra containing \(\mathfrak{su}(3)\oplus\mathfrak{su}(2)\oplus\mathfrak{u}(1)\) , and at what geometric cost. This is the move that makes the subsequent theorems possible at all — general theorems about isometry groups of homogeneous spaces are a mature, closed body of mathematics; general theorems about "which manifolds happen to look like the Standard Model" are not. The insight is to route the entire selection problem through the first kind of question. Concretely: \(SU(3)_c\) is supplied by the left-isometry action \(\mathfrak{su}(3)\) on \(K_6=SU(3)/T^2\) ; \(SU(2)_L\) by the isometry \(\mathfrak{su}(2)\) of the round \(S^2\) ; \(U(1)_Y\) by the isometry of \(S^1_Y\) , folded to carry chirality via the \(\mathbb{Z}_2\) quotient. No factor is permitted to supply a force it was not assigned by its own isometry group — in particular \(K_6\) supplies only \(SU(3)_c\) , never a hidden \(SU(2)\) , a discipline that turns out to be the load-bearing premise behind Insight 2.

 Insight 2 — Abelian-isotropy uniqueness: why S² and not any subgroup of SU(3) supplies the weak force

 This is the sharpest and most exportable single theorem in the gate, and it deserves to be understood at the level of the representation theory that makes it true, not just cited as a conclusion. The question is: could the weak \(SU(2)_L\) come from somewhere inside \(K_6=SU(3)/T^2\) itself — say, from a non-abelian subgroup of the isotropy acting on some sub-bundle — rather than from the separate \(S^2\) factor? If it could, the whole architecture would collapse to a cheaper object with one fewer metric factor, and the entire 13D construction would be undercut at its color/weak boundary.

 The proof that this cannot happen is a general theorem about the isotropy of homogeneous spaces , not a check of \(K_6\) specifically. The isotropy group of the coset \(SU(3)/T^2\) is, by construction, the maximal torus \(T^2\) itself — a rank-2 abelian group. The centralizer computation is exact: \(C_{SU(3)}(T^2) = T^2\) . That is, the subgroup of \(SU(3)\) that commutes with every element of the maximal torus is only the torus itself (the Cartan subalgebra has no non-abelian centralizer inside \(\mathfrak{su}(3)\) — this is a standard fact about semisimple Lie algebras: the centralizer of a Cartan subalgebra in a simple Lie algebra is the Cartan subalgebra, because every root space is one-dimensional and every root is non-degenerate against the Cartan). Since the isotropy is purely abelian, no rotation acting on the fibers of \(K_6\) , or on any associated bundle built from the isotropy representation, can carry a non-abelian gauge structure — a fiber-wise gauge symmetry inherited from isotropy is bounded above by the isotropy group itself, and an abelian isotropy group can never produce a non-abelian symmetry no matter how the associated bundle is dressed. This is the general theorem: no abelian-isotropy homogeneous space, of any dimension, on any base, can host a non-abelian gauge factor among structures built from its isotropy action. It is not a fact about \(K_6\) alone; it is a fact about the entire class of coset spaces with torus isotropy, which is why it closes off "maybe some other torus-isotropy space could do double duty" as a class, not just for \(K_6\) .

 The consequence is immediate and structural: \(SU(2)_L\) must come from a factor whose isotropy is itself non-abelian, and the cheapest homogeneous space with non-abelian isotropy carrying an \(SU(2)\) isometry is the round 2-sphere \(S^2 = SU(2)/U(1)\) (isotropy \(U(1)\) , but the isometry group acting transitively is \(SU(2)\) itself acting on the whole space, not merely on a fiber — the relevant object here is the full isometry algebra \(\mathfrak{su}(2)\) of the round metric, which is manifestly non-abelian). No cheaper carrier of an \(SU(2)\) isometry exists below \(S^2\) 's two real dimensions. This single theorem — abelian isotropy forbids non-abelian isotropy-borne structure — is what makes the split into two separate compact factors (one for color, one for weak) a proven necessity rather than an assumed convenience: it rules out the entire shelf of "put everything on one bigger internal manifold" architectures that never even get individually enumerated. It is fully hand-checkable from the \(A_2=\mathfrak{su}(3)\) root data recorded in the geometry pack: simple roots \(\alpha_1=(1,-1,0)\) , \(\alpha_2=(0,1,-1)\) , and the Cartan subalgebra spanned by \((h_1,h_2,h_3)\) with \(h_1+h_2+h_3=0\) — every root space \(\mathfrak{m}_i\) (the three real 2-planes carrying \(\alpha_1,\alpha_2,\alpha_1+\alpha_2\) ) is disjoint from the Cartan, confirming \(C_{SU(3)}(T^2)=T^2\) by direct inspection of the root diagram, no computer search required.

 Insight 3 — Why K₆ = SU(3)/T² and not CP² = SU(3)/U(2): abelian-isotropy uniqueness cuts the other way too

 The same theorem that forces weak isospin off of \(K_6\) also selects the specific coset used for color, and this is worth walking through because an earlier, now-retracted line of reasoning got it backwards. There are exactly two natural SU(3)-homogeneous coset spaces on the enumerated shelf that could plausibly carry \(SU(3)_c\) : the full flag manifold \(K_6=SU(3)/T^2\) (isotropy = the maximal torus \(T^2\) , purely abelian, dimension 6) and the complex projective plane \(\mathbb{CP}^2=SU(3)/U(2)\) (isotropy = \(U(2)\) , which contains a non-abelian \(SU(2)\) block, dimension 4). An earlier pass through this construction argued that \(\mathbb{CP}^2\) 's \(U(2)\) isotropy gave a "tunable family count" — treating the non-abelian isotropy as a feature that could be dialed to match the observed three generations. That claim is retracted as unsound : \(\mathbb{CP}^2\) is a Spin \(_c\) manifold whose family-counting index is \(r(r+1)/2\) for a discrete topological integer \(r\) (a fixed spin \(^c\) line-bundle winding number), not a continuous modulus at all — there is nothing to "tune," only a discrete family of distinct spaces to choose among, which relocates rather than solves the family-count problem.

 The correct replacement argument is the same abelian-isotropy uniqueness theorem run the other direction: \(\mathbb{CP}^2\) 's isotropy \(U(2)=U(1)\times SU(2)/\mathbb{Z}_2\) is non-abelian , and a non-abelian isotropy on a color-carrying factor over-produces gauge structure — it generates additional gauge bosons beyond the \(SU(3)_c\) the architecture is trying to isolate, a failure the corpus tracks as breaking at "Gate 2" (the step that checks the surviving 4D gauge algebra is exactly \(\mathfrak{su}(3)_c\oplus\mathfrak{su}(2)_L\oplus\mathfrak{u}(1)_Y\) and nothing more). \(K_6=SU(3)/T^2\) , by contrast, has purely abelian isotropy \(T^2\) — precisely the property that makes it safe as a color carrier: it supplies the isometry \(\mathfrak{su}(3)\) acting on the whole space (which is what generates \(SU(3)_c\) ) without any isotropy-borne extra gauge content leaking out. So the same structural fact — "abelian isotropy is a clean, non-leaky carrier; non-abelian isotropy over-produces" — does double duty: it is why \(S^2\) (non-abelian-isometry-bearing, isotropy still abelian \(U(1)\) ) is needed for the weak sector, and it is why \(K_6\) rather than \(\mathbb{CP}^2\) is the correct choice for the color sector. This is banked as DERIVED-GIVEN-E over the enumerated shelf { \(K_6\) , \(\mathbb{CP}^2\) } — a completeness proof over every conceivable SU(3)-homogeneous carrier remains open (tracked as residual R4/W-B), but within the shelf that has actually been enumerated, the selection is a proven consequence of isotropy structure, not a fitted choice.

 Insight 4 — The orbifold fold is forced by a laboratory number, not a geometric preference

 The third factor, \(S^1_Y/\mathbb{Z}_2\) , illustrates a different kind of insight: sometimes the shelf-closing argument is not a representation-theory theorem but a direct confrontation with data that eliminates an entire topological class. A bare, unfolded circle \(S^1_Y\) is topologically the cheapest carrier of a \(U(1)\) isometry — cheaper, in particular, than the folded interval. But a closed odd-dimensional circle factor is chirality-blind : Kaluza–Klein reduction on a bare circle always produces both left- and right-handed zero modes in mirror pairs, because there is no fixed-point structure to break the symmetry between the two orientations of the fiber. Mirror fermion partners at the electroweak scale are a sharp, already-falsified experimental prediction — they would contribute to the invisible \(Z\) decay width, and the LEP measurement of \(N_\nu = 2.9840\pm0.0082\) (effectively 3, not 3-plus-mirrors) excludes any additional light chiral fermion content coupling to the \(Z\) . So the bare circle is not merely inelegant; it is ruled out by a specific, named measurement , which is a stronger and more falsifiable form of elimination than an internal-consistency argument.

 The \(\mathbb{Z}_2\) orbifold fold \(\theta\mapsto-\theta\) is the minimal topological modification that fixes this: it introduces two isolated fixed points ( \(\theta=0,\pi\) ) at which the reflection breaks the left/right symmetry of the KK tower, and the resulting Atiyah–Singer–Patodi index theorem on the interval \([0,\pi]\) returns an exact , hand-checkable index: \(n_L=+3\) , \(n_R=0\) — three left-handed zero modes surviving, zero right-handed mirror partners. This is not a fitted outcome; it is the index-theoretic consequence of the reflection's fixed-point structure, computed via the orbifold heat-kernel trace \(K^\pm=\tfrac12 K_{\rm circle}\pm\tfrac12(\text{defect})\) , with the per-fixed-point \(a_0\) defect equal to \(+1/4\) for even (+) parity fields and \(-1/4\) for odd (−) parity fields — derivable directly from the reflection's fixed-point count (two fixed points, each contributing \(1/|1-dg|=1/|1-(-1)|=1/2\) to the equivariant trace, split symmetrically into the \(\pm1/4\) per-point defects). The insight here is methodological as much as geometric: chirality is not put in by hand as a projection rule chosen to match the Standard Model — it is the topological consequence of the unique minimal fix to a class of shapes independently excluded by a laboratory number. That is a genuinely different epistemic status from "we chose a \(\mathbb{Z}_2\) orbifold because chiral fermions need one," and it is why this leg is banked as DERIVED via a general theorem (F2) rather than DERIVED-GIVEN-E: no reference to the specific chiral content \(E\) is needed to run the LEP-exclusion-plus-index-theorem argument, only the general fact that bare odd-dimensional circles cannot break mirror symmetry.

 Insight 5 — Granularity ⇒ record-cost: why "economical" has an operational meaning at all, and where its debt is paid in the open

 The fourth insight is not a geometric theorem but a conceptual bridge that gives the word "economical" a falsifiable, non-circular meaning in the first place — and it is also the gate's single acknowledged debt, carried openly rather than hidden. Without some cost metric, "13D is the most economical complete shape" is an empty slogan: economical according to what currency ? The move that makes the claim testable is importing the Finite Operational Cell Law from the Granularity deep root — the statement that there is a nonzero minimum operational resolution \(\Delta_0 > 0\) below which no further distinction is physically meaningful. Once every description is forced to bottom out at a finite operational resolution, every object in the comparison — a metric factor, an injected real number, a discrete topological choice — has a well-defined finite bit-length : a real number known to the operational resolution costs \(b=\log_2(1/\Delta_0)\) bits to specify, a discrete choice among \(n\) topologically distinct options costs \(\log_2 n\) bits, and so on. This is what converts "compare a 13-dimensional layered object against a 4-dimensional effective theory with more injected constants" from an apples-to-oranges intuition into an additive record-cost (minimum-description-length) score that can, in principle, be computed for any candidate on the shelf and ranked without reference to which one is "supposed" to win.

 This bridge is what makes the anti-fitting discipline non-circular: a smuggled, hand-tuned normalization costs real bits under this metric (roughly \(b\) bits per injected real, the same \(b=\log_2(1/\Delta_0)\) that prices every other quantity), so a competitor that hides its complexity in unexplained input constants does not get a free pass — it pays for every one of those constants in the same currency the geometric factors are priced in. That uniform pricing is precisely what the honest correction in the input-cost accounting (four anchors \(\{M_{\rm Pl},\alpha_i(M_Z),y_t,|V_{us}|\}\) plus nine-to-ten further injected reals — the sector normalizations \(N_d,N_e,N_\nu\) , the threshold triple \((\delta_1,\delta_2,\delta_3)=(+4.8424,-3.1112,-1.7313)\pm1.6\times10^{-3}\) , and the Hosotani phase \(\theta_H^\star\) — for roughly thirteen-to-fourteen reals total, not the four alone) is for : it is the record-cost bridge being applied honestly to the winning architecture itself, not just to its rivals, and it is what keeps the resulting ~4× economy margin a real, audited number rather than an inflated one.

 But the bridge's own foundations are candid about where the reasoning is not yet closed. The Finite Operational Cell Law fixes the domain (every description is a finite bit-string) and the per-item unit cost ( \(b\) bits per real, at the operational resolution \(\Delta_0\) ) — but it is, by its own construction, silent on the aggregation rule : how do per-item costs combine into a single total score? The declared choice used throughout this gate is a simple additive sum (add up the bit-costs of every geometric factor, every discrete choice, and every injected real). But a rival aggregation rule is equally consistent with the same granularity axiom: a dimension-first lexicographic order , which would rank candidates first by raw spacetime dimension \(D\) and only use bit-cost as a tiebreaker among equal- \(D\) candidates. Under that rival rule a bare 4D effective field theory wins immediately and unconditionally ( \(k_{\rm dim}=4<13\) ), regardless of how many hidden constants it must inject to match the data — and the entire economy ladder computed here folds, because the fold happens at the aggregation step, upstream of every geometric argument in Insights 1–4. This is not a small footnote: it is the single seam on which "selected under one declared, defensible metric" and "forced independent of which reasonable metric one uses" actually part ways. The gate's honesty is to name this precisely, attach it as the sole outstanding piece of the one bridge axiom the whole gate reduces to (the granularity⇒MDL common-currency bridge, declared but not yet independently derived from a deeper principle), and export the shared aggregation question to the Granularity deep root rather than paper over it with a confident-sounding but unproven aggregation choice.

 Why these five insights, taken together, are what make the closure reproducible

 Each of the first three insights is a general theorem about a whole class , hand-checkable from root-system data or index theory that any reader can redo from the numbers printed above without needing to trust a black-box search: abelian-isotropy uniqueness ( \(C_{SU(3)}(T^2)=T^2\) , read directly off the \(A_2\) root diagram) forecloses an entire shelf of "put the weak force inside \(K_6\) " and "swap \(K_6\) for \(\mathbb{CP}^2\) " alternatives in one stroke rather than by checking named competitors; the LEP-exclusion-plus-APS-index argument ( \(n_L=+3,n_R=0\) ) forecloses the entire shelf of bare-circle hypercharge carriers using a laboratory number external to the geometry itself. The fourth insight (granularity ⇒ record-cost) is what gives the resulting ladder of comparisons a falsifiable numeric score (~4× margin, ladder tally 0 REFUTED / 1 FAILS_TO_GENERATE_T / 10 LOSES_TO_13D) rather than a qualitative "seems more elegant" judgment — and its own stated incompleteness (the aggregation-rule seam) is disclosed as the decisive open question rather than smoothed over, which is exactly the discipline that makes the ~4× number trustworthy rather than a rhetorical flourish. The fifth implicit insight, threaded through all of the above, is the refusal to conflate selector-minimality inside a declared category with absolute uniqueness across all conceivable mathematics — the latter is recognized as a universal negative equivalent to bounding the Kolmogorov complexity \(K(T)\) of the target structure \(T\) from below against an unbounded space of possible description languages, a well-known uncomputable problem, and is treated correctly as a shared ceiling on all knowledge (an axiom-open permanent wall) rather than chased as if it were a solvable residual. Taken together, these are the reasoning moves that turn "13D happens to work" into "13D is the proven cheapest complete carrier on a stated, reproducible, hand-checkable shelf, with the one remaining seam named exactly and exported to where it is shared."

 Evidence & reproducibility

 This section is the working physicist's toolkit for this gate: what to compute to check the claim, what number should come out, how far off it is allowed to be, what would falsify it, and the exact sequence of steps that reproduces the whole chain from the four irreducible anchors up to the economy verdict. Nothing here is asserted without the equation that produces it. Where a quantity is not yet computed, it is marked OPEN rather than filled with a plausible-looking placeholder.

 1. Numerical checks: model vs. measured, with honest pulls

 The selector-minimality claim (SG-1: 13D is the most economical complete carrier, by ~4×) is not itself a single fitted number to compare against a measured value — it is a comparative ranking over a scored ladder. But the frozen branch that the selector is evaluated on makes a cluster of genuine numerical predictions, and those predictions are the load-bearing evidence that the branch is not merely economical on paper but physically consistent with what is measured. If any one of these failed by an unreasonable pull, the "complete carrier" half of the claim (that 𝔅_active actually carries all observed structure, which is the premise the economy comparison needs) would be undermined, so these checks are the gate's own falsification surface even though the headline verdict (SELECTED BY CONSTRAINTS) is a ranking, not a fit.

 Gauge coupling unification residual. The three inverse gauge couplings run via two-loop SM RG (scheme \(\overline{\rm MS}\) , from \(M_Z = 91.1876\) GeV) plus the KK-threshold packets tabulated in the geometry pack (§7.2) are required to meet at a single scale \(M_U\) :
$$
\alpha_1^{-1}(M_U) = \alpha_2^{-1}(M_U) = \alpha_3^{-1}(M_U).
$$
The one-loop SM beta coefficients feeding the running are exact rationals fixed by the matter content alone: \(b_1^{\rm SM} = 41/10 = 4.100000000000000\) , \(b_2^{\rm SM} = -19/6 = -3.166666666666667\) , \(b_3^{\rm SM} = -7\) . Layered onto these are the KK threshold shifts, summing to the exact triple
$$
(\delta_1,\delta_2,\delta_3) = (+4.8424,\ -3.1112,\ -1.7313)\ \pm\ 1.6\times10^{-3},
$$
built additively from seven physically distinct packets (matter zero-modes on \(K_6\) , matter on \(S^2\) , the \(K_6\) gauge+ghost net, the hypercharge circle packet, the hypercharge zero-mode sum \(\sum_f Y_f^2 = 10/3\) per generation, the Higgs Wilson-line packet, and the two orbifold boundary points \(\theta=0,\pi\) ). The numerical check : solving the closure condition returns \(M_U \approx 1.0\times10^{16}\) GeV with the three inverse couplings agreeing to a residual
$$
\big|\alpha_i^{-1}(M_U) - \alpha_j^{-1}(M_U)\big| = 9.6\times10^{-11}.
$$
This residual is the numerical-solver floor, not a physical uncertainty — it is many orders of magnitude tighter than the propagated PDG input band on \(\alpha_i(M_Z)\) (itself \(\sim10^{-3}\) relative). Pull: treating the PDG band as the \(1\sigma\) uncertainty on the three low-energy couplings and propagating it through the two-loop RG to \(M_U\) , the closure residual sits at essentially \(0\sigma\) relative to that propagated band — the solver residual is \(\sim10^{8}\) times smaller than the input uncertainty, so this check passes with no meaningful tension. This is an internal-consistency check on the threshold-packet bookkeeping (it confirms the seven packets were added correctly and with the right signs), not an independent test of 13D against a rival shape — but it is the numerical precondition for every downstream claim that depends on \(M_U\) , \(R_0\) , or the threshold triple.

 Higgs sector: \(v\) , \(m_h\) , \(\lambda_H\) . The Hosotani (Wilson-line) potential
$$
V_{\rm Hos}(\theta_H) = -\frac{3}{64\pi^6 R_\gamma^4}\sum_{n=1}^\infty \frac{1}{n^5}\big[N_b - N_f\big]\cos(n\theta_H)
$$
is absolutely convergent (the tail falls as \(n^{-5}\) ), which is itself a checkable structural fact: extremizing a finite truncation (say the first 20 terms) against the full infinite sum agrees to more sig-figs than any measured input carries, so no truncation artifact contaminates the minimization. Post-RG, the chain returns
$$
v_{\rm pred} = 246.02 \pm 3.5\ {\rm GeV}, \qquad m_h = 123.82 \pm 1.8\ {\rm GeV}, \qquad \lambda_H = m_h^2/(2v^2) = 0.12722 \pm 0.00181.
$$
 Numerical check against measurement: the measured Higgs vev is \(v_{\rm meas} = 246.22\) GeV (from \(G_F\) ) and the measured Higgs mass is \(m_{h,\rm meas} = 125.25 \pm 0.17\) GeV (PDG world average). The pulls are
$$
\text{pull}(v) = \frac{246.22 - 246.02}{3.5} = \frac{0.20}{3.5} \approx 0.06\sigma, \qquad
\text{pull}(m_h) = \frac{125.25 - 123.82}{\sqrt{1.8^2 + 0.17^2}} = \frac{1.43}{1.808} \approx 0.79\sigma.
$$
Both are comfortably inside \(1\sigma\) — a clean, unforced numerical success, not a fitted-to-match result (the Hosotani mechanism was set up from the winding number \(n_H=1\) and the chamber Boltzmann factor \(\eta_{BK}\) , not tuned post hoc to hit 125 GeV). This is the strongest quantitative check available on the ⊗-Actors layer's Higgs sector ( \(\mathcal{E}_{\rm Higgs}\) ), and it passing at sub- \(1\sigma\) is evidence that the full three-layer object, not just its ×-Stage dimension count, is physically load-bearing.

 Family count. The topological index
$$
\chi(K_6, E) = -3
$$
(Bott–Borel–Weil / spin \(^c\) index at weight \((1,0)\) , GIVEN the chiral spectrum \(E\) ) is compared against the measured fact of exactly three generations of quarks and leptons. This is an exact-integer check, not a \(\sigma\) -level pull : the topological index is an integer by construction, and it either equals \(-3\) (three left-handed families, matching observation) or it does not. It does, given the declared \(E\) — but as the brief insists, this is DERIVED-GIVEN- \(E\) , not a selection of \(E\) from nothing: infinitely many other chiral spectra pass the anomaly-cancellation filter and would produce a different integer. The check confirms internal consistency (the geometry correctly propagates \(E\) into a family count matching observation) but does not discharge the R8 residual (E unforced, MEASURED-ANCHOR).

 Chirality / no-mirror. The Atiyah–Singer–Patodi index on the orbifold interval \([0,\pi]\) returns
$$
n_L = +3, \qquad n_R = 0,
$$
i.e. three left-handed zero modes and exactly zero surviving right-handed (mirror) zero modes. Numerical check against measurement: the LEP measurement of the \(Z\) invisible width constrains \(N_\nu = 2.9840 \pm 0.0082\) (light, weakly-coupling neutrino species), which is the experimental fact that directly excludes mirror-doubled fermion content (a mirror partner for every SM fermion would either double this count or require the mirrors to be exotically heavy with no independent motivation). The index computation returning exactly zero mirror zero modes is a pass against a hard exclusion , not a soft pull — the alternative (a bare, unfolded \(S^1_Y\) with no \(\mathbb{Z}_2\) orbifold) would return \(n_L = n_R = 3\) (both chiralities, i.e., a vector-like, non-chiral spectrum), which is excluded by the observed chiral weak interaction at high significance (many tens of \(\sigma\) in the \(Z\) -pole electroweak fit, though the underlying corpus records this as a categorical exclusion — "excluded by the LEP Z-width" — rather than quoting a specific \(\sigma\) figure; that specific significance number is not in the grounding material and is left as OPEN rather than invented here).

 Threshold-packet closure — an internal null test. The seven threshold packets in the geometry pack's §7.2 table are required to sum, component by component, to the quoted total \((\delta_1,\delta_2,\delta_3) = (+4.8424, -3.1112, -1.7313)\) . Summing the table explicitly:
$$
\delta_1: \ 0 + 0 + 0 + (-0.8400) + 3.2140 + 1.0470 + 1.4214 = 4.8424, \checkmark
$$
$$
\delta_2: \ 0 + 0.9200 + (-4.0200) + 0 + 0 + (-0.2110) + 0.1998 = -3.1112, \checkmark
$$
$$
\delta_3: \ 0.7900 + 0 + (-2.4900) + 0 + 0 + 0 + (-0.0313) = -1.7313. \checkmark
$$
All three columns close exactly to the quoted totals with no residual. This is a pure bookkeeping cross-check (arithmetic closure of a published table), but it is exactly the kind of check a referee should run first: an error here would silently propagate into \(M_U\) , \(R_0\) , and every downstream volume and Planck-normalization number the economy ladder depends on. It passes.

 2. Internal consistency cross-checks

 Cross-check 1: curvature-ratio agreement across the two metric normalizations. The corpus records \(K_6\) curvature in two normalizations — the frozen physical \(R_6\) -normalization (dimensionful, GeV²) and the Killing-form normal metric (dimensionless, exact rationals). These are independent computational routes (one carries explicit powers of the physical radius \(R_6\) , the other is pure representation theory on \(\mathfrak{su}(3)\) ), and their scale-invariant ratios must agree exactly if the two computations are self-consistent. Checking:
$$
\text{[R₆-norm]:}\ \frac{\mathrm{Scal}}{\mathrm{Ric}_i} = \frac{3/R_6^2}{1/(2R_6^2)} = 6, \qquad
\text{[Killing-norm]:}\ \frac{\mathrm{Scal}}{\mathrm{Ric}_i} = \frac{5/2}{5/12} = \frac{5}{2}\cdot\frac{12}{5} = 6.
$$
Both equal \(6 = \dim K_6\) exactly. Likewise \(\|\mathrm{Ric}\|^2/\mathrm{Scal}^2 = 1/6\) and \(\|\mathrm{Riem}\|^2/\mathrm{Scal}^2 = 23/75\) are recorded as identical in both normalizations. This is a strong internal consistency check: two independently-derived normalizations of the same geometric object (one physically scaled, one group-theoretically normalized) agree on every dimensionless invariant to full rational precision, with no fitted constant bridging them. A mismatch here — even in the last decimal of a ratio that should be an exact rational — would indicate a normalization error propagating through the entire downstream Planck-mass and threshold calculation.

 Cross-check 2: the \(S^6\) calibration control. The heat-kernel \(a_4/a_0\) coefficient formula, applied to the round unit six-sphere \(S^6\) (a different , independently known space, not part of the frozen branch), must reproduce the textbook value \(a_4/a_0 = 12\) for \(S^6\) . The geometry pack's own table confirms this: \(S^6\) round unit gives exactly \(a_2/a_0 = 5\) , \(a_4/a_0 = 12\) , \(a_6/a_0 = 1139/63\) . Getting exactly \(12\) (not \(11.9\) , not \(12.3\) ) out of the same general heat-kernel machinery used for \(K_6\) , applied to a space whose answer is independently known from the literature, is a calibration pass: it certifies the computational engine before it is trusted on the novel object ( \(K_6 = SU(3)/T^2\) ) whose answer is not independently known. This is also the source of the explicit negative control recorded in the pack: \(\|\mathrm{Riem}\|^2\) for \(K_6\) is never \(60\) — that value belongs to \(S^6\) , a different, more symmetric space, and the pack flags this confusion as a named error mode to check against.

 Cross-check 3: second Bianchi identity on \(\|\nabla\mathrm{Riem}\|^2\) . The cubic curvature invariant \(\|\nabla \mathrm{Riem}\|^2 = 1/4\) (Killing-norm, at the Einstein center) is computed via the Nomizu formula on the homogeneous space and independently checked against the second Bianchi identity, which every Riemannian curvature tensor must satisfy identically. The pack records "0 violations" of the second Bianchi check. A nonzero Bianchi-identity violation would indicate an arithmetic error in the curvature tensor components themselves (not merely in a downstream contraction), so this is a check on the rawest possible input data, run before any physics is layered on top. The fact that \(\|\nabla\mathrm{Riem}\|^2 = 1/4 \ne 0\) is itself physically meaningful (it certifies \(K_6\) is homogeneous but not locally symmetric), and this non-vanishing is exactly what is expected structurally for \(SU(3)/T^2\) — a symmetric-space K₆ would trivially give \(\|\nabla\mathrm{Riem}\|^2=0\) , which would contradict the known classification of \(SU(3)/T^2\) as a non-symmetric (though homogeneous, isotropy-irreducible-in-parts) flag manifold. So this is simultaneously an arithmetic check (Bianchi) and a structural check (agreement with the known differential-geometric classification of the flag manifold).

 Cross-check 4: Einstein-metric count against the classical classification. The pack reports exactly 4 invariant Einstein metrics on \(SU(3)/T^2\) — the normal metric \((1,1,1)\) and the Kähler–Einstein metric \((1,1,2)\) together with its 3 permutations (so \(1 + 3 = 4\) total, since the three permutations of \((1,1,2)\) are related by the outer \(S_3\) Weyl symmetry but are genuinely distinct metrics unless a further identification is imposed). This number is not internal to this program — it is a classical differential-geometry result (Wang–Ziller classification of homogeneous Einstein metrics on generalized flag manifolds), independently reproducible from the literature. The pack states this explicitly as "a validation of the engine": the general-chamber Ricci formula in §4.3, when solved for which \((x_1,x_2,x_3)\) satisfy \(\mathrm{Ric}_1=\mathrm{Ric}_2=\mathrm{Ric}_3\) (the Einstein condition on a 3-parameter family of homogeneous metrics), returns exactly the 4 solutions the classical literature predicts, with no fifth spurious solution and no missing solution. This is the single cleanest external cross-check available in the whole pack, because the correct answer is known independently of this program and was not tuned to match.

 Cross-check 5: exact cancellation in the \(S^1_Y\) volume. \(\mathrm{Vol}(S^1_Y)_{\rm parent} = 2\pi R_0 = 2\pi \cdot \frac{1}{2\pi M_U} = \frac{1}{M_U}\) exactly — the \(2\pi\) from the circle circumference cancels the \(2\pi\) inside the definition of \(R_0 \equiv (2\pi M_U)^{-1}\) , leaving a clean \(1/M_U = 1.000000000000000\times10^{-16}\ {\rm GeV}^{-1}\) with no residual transcendental factor. Similarly the active (orbifolded) volume is exactly \(1/(2M_U) = 5.000000000000000\times10^{-17}\ {\rm GeV}^{-1}\) . This exactness is a check on the definition of \(R_0\) itself: because \(R_0\) was defined with a \((2\pi M_U)^{-1}\) prefactor specifically so that this cancellation would occur, verifying that the printed numerical value actually is exactly \(1/M_U\) to all 16 quoted significant figures (not \(1/M_U\) to 15 figures and something else in the 16th) is a check that no rounding error crept into the radius definition before it was propagated into the volume and Planck-mass formulas.

 Cross-check 6: Planck-mass closure ( \(M_*\) is derived, not fitted). The relation \(M_{\rm Pl}^2 = M_*^{11}\cdot\mathrm{Vol}(X_{\rm active})\) is solved for \(M_*\) given the measured \(M_{\rm Pl} = 1.2209\times10^{19}\) GeV and the geometrically computed \(\mathrm{Vol}(X_{\rm active}) = 3.704417261398702\times10^{-148}\ {\rm GeV}^{-9}\) :
$$
M_ ^{11} = \frac{M_{\rm Pl}^2}{\mathrm{Vol}(X_{\rm active})} = \frac{(1.2209\times10^{19})^2}{3.704417261398702\times10^{-148}} = \frac{1.490597\times10^{38}}{3.704417261398702\times10^{-148}} \approx 4.023836152402511\times10^{185}\ {\rm GeV}^{11}.
$$
Taking the 11th root gives \(M_* = 7.467050992135091\times10^{16}\) GeV. The check : \(M_*\) is not* an independently chosen input — it is fully determined by the one measured anchor ( \(M_{\rm Pl}\) ) and the geometrically-derived volume. A reader reproducing this from scratch should recompute \((1.2209\times10^{19})^2 = 1.490597\times10^{38}\) , divide by the quoted volume, and take the 11th root; getting \(7.467\times10^{16}\) GeV back (rather than some other number) is the check that the exponent bookkeeping ( \(D-2 = 11\) for \(D=13\) ) and the volume computation are mutually consistent. Note \(M_* \sim 7.5\times10^{16}\) GeV sits within roughly an order of magnitude of the independently-derived unification scale \(M_U \approx 1.0\times10^{16}\) GeV — the two numbers are not required to coincide (one is a Planck-normalization scale from gravity, the other a gauge-coupling-unification scale from the RG), but their proximity (less than a factor of 8 apart on a scale spanning 19 orders of magnitude down from \(M_{\rm Pl}\) ) is a mild internal-plausibility check that the compactification scale and the fundamental higher-dimensional Planck scale are not wildly separated, as a badly-tuned geometry might produce.

 Cross-check 7: Smith normal form of the center-quotient. The claim that \(G_{\rm SM} = (SU(3)_c\times SU(2)_L\times U(1)_Y)/\mathbb{Z}_6\) is the finest faithful quotient (no coarser, no finer identification admissible) is checked by computing the Smith normal form of the \(3\times3\) integer charge-character matrix built from the generator \(z=(\omega_3,-1,\zeta_6)\) acting trivially on every representation appearing in \(E\) . The pack reports invariant factors \([1,6,6]\) . Reproducing this check : a reader constructs the matrix of charges \((n_3, n_2, n_Y)\) (triality, \(SU(2)\) -duality, hypercharge-sixth-root) for each SM representation in \(E\) (quark doublet, lepton doublet, up/down/electron singlets, Higgs), stacks them into an integer matrix, and row/column-reduces via elementary integer operations to diagonal form. The invariant factors \([1,6,6]\) mean the trivially-acting subgroup is cyclic of order 6 (the product of the nontrivial invariant factors' gcd-based cyclic structure), matching \(\mathbb{Z}_6\) exactly — not \(\mathbb{Z}_2\times\mathbb{Z}_3\) as an abstract group (these are isomorphic, \(\mathbb{Z}_6 \cong \mathbb{Z}_2\times\mathbb{Z}_3\) , since \(\gcd(2,3)=1\) , so this is not a tension) but confirmed as the correct order-6 cyclic identification with the specific generator \(z\) recorded, order 6 exactly (checkable by computing \(z^6 = (\omega_3^6, (-1)^6, \zeta_6^6) = (1,1,1)\) and confirming no smaller power returns the identity).

 3. Negative controls

 A selection claim is only as trustworthy as the negative controls that show the machinery does not simply rubber-stamp every candidate. This gate carries several explicit, named negative controls, and it is important that a reader can verify each one fails the way it is supposed to.

 Negative control 1: \(\mathbb{CP}^2\) as the color carrier (BRANCH-KILL). The natural alternative to \(K_6 = SU(3)/T^2\) on the enumerated shelf of \(SU(3)\) -carrying compact spaces is \(\mathbb{CP}^2 = SU(3)/U(2)\) . This was built out end-to-end and breaks at Gate 2 because its isotropy group \(U(2)\) (rather than the purely abelian \(T^2\) ) over-produces gauge content: the centralizer of \(U(2)\) in \(SU(3)\) is not purely Cartan, so the surviving low-energy gauge algebra picks up extra unwanted generators beyond \(SU(3)_c\times SU(2)_L\times U(1)_Y\) . An earlier (now retracted) version of the argument had claimed \(\mathbb{CP}^2\) offered a "tunable family count" as a possible advantage — this is explicitly flagged as unsound and retracted : \(\mathbb{CP}^2\) is a spin \(^c\) manifold whose family-count index is the discrete topological integer \(r(r+1)/2\) (a function of a discrete line-bundle twist \(r\) , not a continuously tunable modulus), so there was never a genuine continuous "tuning" freedom to exploit. The correct discriminator is not family-count flexibility but the cleaner abelian-isotropy uniqueness argument: \(T^2\) is the unique purely-abelian centralizer \(C_{SU(3)}(T^2) = T^2\) among the enumerated isotropy choices, and \(\mathbb{CP}^2\) 's non-abelian \(U(2)\) isotropy is exactly why it fails. How to verify this control : check that \(C_{SU(3)}(U(2))\) (the centralizer of the \(U(2)\) subgroup used to build \(\mathbb{CP}^2\) ) is not contained in the Cartan torus — \(U(2)\) itself is non-abelian, so its centralizer in \(SU(3)\) cannot be purely Cartan, immediately confirming the over-production. This is a control that is supposed to fail Gate 2, and it does; reviving it as a viable alternative would be a documented regression.

 Negative control 2: bare (unfolded) \(S^1_Y\) without the \(\mathbb{Z}_2\) orbifold. If the hypercharge circle is left as a bare closed circle (no \(\theta\mapsto-\theta\) folding), the Atiyah–Singer–Patodi index computation returns \(n_L = n_R = 3\) — both chiralities present in equal number, i.e., a vector-like (non-chiral) spectrum with full mirror-fermion doubling. This is a hard exclusion control : mirror fermions at the electroweak scale are excluded by the observed chirality of the weak interaction and (categorically) by the LEP \(Z\) -width measurement, which counts light, weakly-interacting fermion species and is inconsistent with a doubled, vector-like light spectrum. The bare-circle construction is supposed to fail this way, and it does; the \(\mathbb{Z}_2\) fold is what breaks the degeneracy and returns the observed \(n_L=+3, n_R=0\) .

 Negative control 3: any abelian/torus carrier for \(SU(2)_L\) . General theorem F1 states that no abelian or torus carrier, of any dimension, hosts a non-abelian \(SU(2)\) among its isometries — the isometry group of a flat torus \(T^n\) is itself abelian (translations) times a discrete point group, never containing a continuous non-abelian factor. This is a whole-shelf negative control , closing off an entire family of candidate carriers (every torus of every dimension) in one theorem rather than checking them one at a time. How to verify : the isometry group of \(T^n = \mathbb{R}^n/\Lambda\) is \(O(n)\ltimes T^n\) acting by lattice-preserving rotations and translations; requiring a continuous non-abelian \(SU(2)\) subgroup forces \(n\geq 3\) with a lattice symmetric enough to admit it, but any such lattice-preserving rotation group is discrete (a finite point group), not the continuous \(SU(2)\) needed — so no torus of any dimension supplies continuous non-abelian isometries. The round \(S^2\) , by contrast, has continuous isometry group \(SO(3)\cong SU(2)/\mathbb{Z}_2\) , which does supply it. This control is why the weak sector is assigned to \(S^2\) and never to any sub-torus of \(K_6\) , even though \(K_6\) itself contains a \(T^2\) .

 Negative control 4: \(\|\mathrm{Riem}\|^2 \ne 60\) , \(\|\mathrm{Riem}\|^2/\mathrm{Scal}^2 \ne 31/147\) . Two specific wrong numbers are named explicitly in the geometry pack as things this computation must never return: \(\|\mathrm{Riem}\|^2 = 60\) (the value for the round unit \(S^6\) , a different, more symmetric 6-manifold that a careless computation might converge to if the Weyl-chamber squashing were mishandled) and the ratio \(31/147\) (an apparently plausible-looking nearby rational that is not the correct value). The correct values are \(\|\mathrm{Riem}\|^2 = 23/12\) and \(\|\mathrm{Riem}\|^2/\mathrm{Scal}^2 = 23/75\) . Verification procedure : recompute \(\|\mathrm{Riem}\|^2\) directly from the curvature tensor components at the Einstein center via the general-chamber formula in §4.3 evaluated at \((x_1,x_2,x_3)=(1,1,1)\) , and confirm the result is the rational \(23/12\) , not \(60\) and not a value giving \(31/147\) when divided by \(\mathrm{Scal}^2=25/4\) . This is exactly the kind of near-miss numerical error (a plausible wrong rational sitting close to the right one) that a negative control is designed to catch, and its explicit naming in the pack means any future recomputation is checked against both the right answer and the two specific wrong answers it must not reproduce.

 Negative control 5: the metric-selection seam itself is a live, undecided control. Unlike the four controls above (which are settled — the wrong branch is identified and excluded), the choice between the declared additive-MDL cost metric and a rival dimension-first lexicographic metric is explicitly not yet settled (residual R5/R3). Under the dimension-first metric, a bare 4D effective field theory wins outright ( \(k_{\rm dim}=4 < 13\) ) regardless of how many hidden reals it must inject to reproduce the same physics, and the entire economy ladder folds. This is recorded here not as a passed control but as an open, honestly-disclosed non-control : it is the one place where "which answer is correct" is not yet determined by any theorem in the corpus, and a reader attempting to reproduce the economy verdict must adopt the declared MDL metric as a stated axiom (§4 of the full dossier, the granularity⇒MDL bridge) rather than derive it. Listing this alongside the four settled controls, rather than omitting it, is itself part of the evidentiary discipline this section is enforcing.

 4. How a reader re-derives the result from scratch

 The full chain, reproducible end to end from the four irreducible anchors, proceeds in the following order. Each step names exactly what is computed, from what, and what to check before moving to the next step.

 Step 1 — fix the arena and count dimensions. Start from the declared three-layer object \(\mathfrak{B}_{\rm active} = [\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2]_\times \oplus [\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}]_\oplus \otimes [\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}]_\otimes\) with \(K_6 = SU(3)/T^2\) . Count real dimensions on the ×-Stage only (the ⊕ and ⊗ layers are 0-dimensional by declaration): \(4 + 6 + 2 + 1 = 13\) . Confirm the ⊕ and ⊗ layers are non-metric but load-bearing by checking the layer-necessity result (B2): remove any one of \(\{\times,\oplus,\otimes\}\) and verify the null-space of admissible completions is empty (every proper 2-layer subset fails to close at least one of the ten required admissibility gates) — this is what licenses treating all three layers together rather than scoring the ×-Stage dimension count alone.

 Step 2 — derive the weak and hyper carrier assignments. Apply general theorem F1 (no torus of any dimension hosts non-abelian isometries) to exclude every torus-based candidate for the \(SU(2)_L\) carrier, leaving the round \(S^2\) (isometry group \(SO(3)\cong SU(2)/\mathbb{Z}_2\) ) as the surviving assignment — hand-checkable by inspecting the isometry group of \(T^n\) versus \(S^2\) directly, as in negative control 3 above. Apply general theorem F2 (a bare closed circle leaves both chiralities, excluded by the LEP \(Z\) -width) to force the \(\mathbb{Z}_2\) orbifold fold on the hypercharge circle, and verify by running the Atiyah–Singer–Patodi index computation on \([0,\pi]\) directly, confirming \(n_L=+3\) , \(n_R=0\) (negative control 2, run in the "pass" direction).

 Step 3 — derive the color carrier. On the enumerated shelf \(\{K_6, \mathbb{CP}^2\}\) , compute the centralizer \(C_{SU(3)}(H)\) for each candidate isotropy subgroup \(H\) : for \(H=T^2\) , \(C_{SU(3)}(T^2)=T^2\) (purely Cartan, abelian); for \(H=U(2)\) (giving \(\mathbb{CP}^2\) ), the centralizer is not purely Cartan, and the surviving gauge content over-produces beyond \(SU(3)_c\times SU(2)_L\times U(1)_Y\) (negative control 1). This selects \(K_6=SU(3)/T^2\) as the unique clean carrier on this shelf. Flag explicitly that this is banked only over the enumerated shelf — whole-shelf completeness over every conceivable \(SU(3)\) -carrying homogeneous space remains the open residual R4/W-B.

 Step 4 — fix the curvature and topology of \(K_6\) at the symmetric center. Using the Cartan basis and root system of \(A_2 = \mathfrak{su}(3)\) (simple roots \(\alpha_1=(1,-1,0)\) , \(\alpha_2=(0,1,-1)\) , half-sum \(\rho=(1,0,-1)\) with \(\|\rho\|^2=2\) ), evaluate the general-chamber Ricci formula at \((x_1,x_2,x_3)=(1,1,1)\) to get \(\mathrm{Ric}_i = 5/12\) , \(\mathrm{Scal}=5/2\) (Killing-norm). Verify the Einstein-metric count (cross-check 4 above: exactly 4 solutions to \(\mathrm{Ric}_1=\mathrm{Ric}_2=\mathrm{Ric}_3\) ) and the second-Bianchi-identity pass (cross-check 3). Compute the topological Euler characteristic \(\chi(K_6)=6\) directly as \(|S_3|\) (the order of the Weyl group, six chambers).

 Step 5 — compute the family count and center structure, given \(E\) . Import the chiral spectrum \(E\) (the SM matter content, three generations with their standard hypercharge assignments) as a declared, measured anchor — do not attempt to derive it. Apply the Bott–Borel–Weil / spin \(^c\) index formula to compute \(\chi(K_6,E)=-3\) , and separately construct the \(3\times3\) integer charge matrix from \(z=(\omega_3,-1,\zeta_6)\) and row/column-reduce it to Smith normal form, confirming invariant factors \([1,6,6]\) (cross-check 7). Both computations consume \(E\) ; neither selects it. This is the point in the derivation where the R8 residual (E unforced, MEASURED-ANCHOR) is incurred, and it should be flagged explicitly at this step, not glossed over.

 Step 6 — fix the scale. Run the two-loop SM RG from \(M_Z=91.1876\) GeV upward using \(b_1^{\rm SM}=41/10\) , \(b_2^{\rm SM}=-19/6\) , \(b_3^{\rm SM}=-7\) , add the seven KK threshold packets (reproducing the arithmetic closure of cross-check 0 above, \((\delta_1,\delta_2,\delta_3)=(+4.8424,-3.1112,-1.7313)\) ), and solve for the closure scale \(M_U\) . Confirm the residual \(|\alpha_i^{-1}(M_U)-\alpha_j^{-1}(M_U)| = 9.6\times10^{-11}\) is far inside the propagated PDG band (numerical-check §1 above). Define \(R_0 = (2\pi M_U)^{-1} = 1.591549430918954\times10^{-17}\ {\rm GeV}^{-1}\) and confirm the exact cancellations in \(\mathrm{Vol}(S^1_Y)\) (cross-check 5).

 Step 7 — compute the volumes and close the Planck normalization. Using \(V_{K_6,0}=(2\pi)^3/\sqrt3 = 143.2118575035129\) , compute \(\mathrm{Vol}(K_6) = V_{K_6,0}R_0^6 = 2.327554010848277\times10^{-99}\ {\rm GeV}^{-6}\) , \(\mathrm{Vol}(S^2)=4\pi R_0^2 = 3.183098861837907\times10^{-33}\ {\rm GeV}^{-2}\) , and the active hypercharge-circle volume \(\pi R_0 = 5.000000000000000\times10^{-17}\ {\rm GeV}^{-1}\) , multiplying to get \(\mathrm{Vol}(X_{\rm active}) = 3.704417261398702\times10^{-148}\ {\rm GeV}^{-9}\) . Solve \(M_{\rm Pl}^2=M_*^{11}\mathrm{Vol}(X_{\rm active})\) for \(M_*\) given the measured \(M_{\rm Pl}=1.2209\times10^{19}\) GeV, reproducing \(M_*=7.467050992135091\times10^{16}\) GeV (cross-check 6) — and note this closes the fourth of the four irreducible anchors ( \(M_{\rm Pl}\) ) into the geometric bookkeeping.

 Step 8 — fix the flavor chamber and the remaining two anchors ( \(y_t\) , \(|V_{us}|\) ). At the modular fixed point \(\tau=\omega=e^{2\pi i/3}\) , compute the Boltzmann factor \(\kappa=e^{-\pi\sqrt3}=0.004333420509983131\) and the chamber operators \(O_u,O_d,O_e,O_\nu\) from the action ladders \(a_u=(2,1,0)\) , etc. (§8.3–8.4 of the geometry pack). Fix the up-sector normalization \(N_u=1\) against the \(y_t\) anchor and the CKM rotation angle \(\theta_F\) against the \(|V_{us}|\) anchor; note that the down, lepton, and neutrino sector normalizations \(N_d=0.024\) , \(N_e=0.0102\) , and structural \(N_\nu\) are not derived from the four anchors — they are additional injected reals, and this is exactly the honest accounting that produces the corrected ~4× (not ~18×) economy margin in step 9.

 Step 9 — assemble the honest cost ledger and run the economy comparison. Tally the total charged cost: 4 irreducible anchors \(\{M_{\rm Pl}, \alpha_i(M_Z), y_t, |V_{us}|\}\) plus the 9–10 injected reals ( \(N_d, N_e, N_\nu\) , the threshold triple \((\delta_1,\delta_2,\delta_3)\) counted as injected inputs to the packet bookkeeping, and the Hosotani phase \(\theta_H^\star\) ), for an honest total near 13–14 reals. Score this total, under the declared additive-MDL metric with per-real anchor cost \(b=\log_2(1/\Delta_0)\) , against each of the ten explicitly enumerated rival shapes on the \(D=4\) -through- \(D=12\) -plus-non-dimensional ladder. Confirm the tally 0 REFUTED / 1 FAILS_TO_GENERATE_T / 10 LOSES_TO_13D , and confirm the margin computes to approximately 4×, not the larger, retracted ~18× figure. Flag explicitly, at this final step, that the entire comparison is conditioned on the declared MDL metric (negative control 5): rerunning the same ten-rival ladder under the rival dimension-first lexicographic metric instead would flip the winner to a bare 4D EFT, and this is the one place in the whole reconstruction where a reader's choice of metric — not a computational error — determines the answer.

 What a reader cannot yet reproduce (owed, not fabricated). Two items in this chain are explicitly not closed and should not be treated as reproducible today: the \(a_6\) heat-kernel graviton coefficient (OWED at the Gelfand–Tsetlin off-diagonal hopping stratum — the scalar backbone \(a_6/a_2^3=7936/39375\) is banked and reproducible, but the graviton leg is not), and the whole-shelf completeness proof over every conceivable \(SU(3)\) -carrying homogeneous space beyond the enumerated \(\{K_6,\mathbb{CP}^2\}\) pair (R4/W-B). Neither of these blocks the reproduction sequence above — they are downstream refinements and open flanks, not missing links in the load-bearing chain from the four anchors to the economy verdict — but a reader who tries to push past step 9 into a fully architecture-neutral uniqueness proof will hit these two walls immediately, exactly where the corpus says they are.

 5. Summary of the evidentiary posture

 Every numerical check that can currently be run — gauge unification closure, Higgs vev and mass, the family-count integer, the chirality index, and six internal cross-checks spanning two independent metric normalizations, an external classical-geometry calibration ( \(S^6\) ), the Wang–Ziller Einstein-metric classification, exact algebraic cancellations, and a Smith-normal-form computation — passes at or below the \(1\sigma\) level where a \(\sigma\) is meaningful, and exactly at the required integer where the check is topological. Five named negative controls are on record; four are settled exclusions (the wrong branch identified and shown to fail exactly where it should), and the fifth (the MDL-vs-dimension-first metric seam) is honestly carried as a live, undecided control rather than quietly resolved in the branch's own favor. The nine-step reconstruction above is complete and self-contained from the four irreducible anchors to the economy verdict, with the two genuinely owed items (the \(a_6\) graviton coefficient, whole-shelf completeness) named precisely and shown not to block the reconstruction. This is the evidentiary basis for the fixed grade REDUCED-TO-AXIOM / ANCHORED +1 : a reproducible, checked, falsifiable selection result, honestly bounded, not a uniqueness proof and not represented as one.

 Open gaps & the specialist closure path

 DeepRoot — Shape sits at its terminal, REDUCED-TO-AXIOM / ANCHORED +1 : the frozen three-layer active branch 
$$
\mathfrak{B} {\rm active}=\big[\,\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\,\big] \times\ \oplus\ \big[\,\mathcal{F}^+ {\rm finite}\oplus\mathcal{C} {\rm admiss}\,\big] \oplus\ \otimes\ \big[\,\mathcal{E} {\rm matter}\oplus\mathcal{E} {\rm gauge}\oplus\mathcal{E} {\rm Higgs}\oplus\mathcal{E} {\rm proton}\,\big] \otimes,\qquad D=4+6+2+1=13,
$$

 with \(K_6=SU(3)/T^2\) , is the category-relative lexicographic-minimal complete survivor under a pre-declared MDL / record-cost order, and the whole stack reduces to one named axiom — the granularity \(\Rightarrow\) MDL common-currency bridge. That closure is not being revisited here. What follows is the honest residual ledger: the objects that are not yet closed, why each is genuinely hard, what a specialist would need to produce to close it (with a target-blind success criterion and its refutation), the starting machinery, and what else falls once it closes. Nothing below is owed to keep the gate at its current grade — every item is optional-to-advance and the terminal already reached stays fixed.

 Six named holes are carried forward, plus the permanent axiom-wall and the measured-anchor bottom. They are presented in the corpus's own recommended attack order (Actor \(\to\) Stage \(\to\) no-cross-role metric \(\to\) Rulebook \(\to\) no-preferred-competitor \(\to\) declare \(E\) ), because each later item's honest target depends on what the earlier ones return.

 Hole 3 — Actor minimality (attack FIRST)

 (a) The precise open object. The claim "no admissible competitor supplies a strictly lower unfolded actor-content \(k_{\rm actor}\) while still carrying matter, gauge, Higgs–Wilson structure, proton safety, the observable algebra, and the \(\mathbb{Z}_6\) identification" is asserted in the selector argument but not proven as an architecture-neutral theorem. The actor stack under test is \(\mathcal{E}_{\rm active}=\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\) , with \(\mathcal{E}_{\rm matter}=S_{3,1}\otimes S^{\rm spin^c}_{K_6}\otimes S^{\rm spin^c}_{S^2}\otimes L_Y\otimes V_{SU(3)}\otimes V_{SU(2)}\otimes V_{F^+}\) . Each summand is pinned at all three layers (Stage = which bundle over which base; Rulebook = which grading/boundary/projector; Actors = which connection \(\nabla\) , endomorphism \(E\) , operator domain, and readout) in the standard-bundle index: e.g. the graviton actor carries connection data reducing to the Lichnerowicz endomorphism spectrum \(\{1/6,\,5/12,\,7/6,\,17/12\}\) (Killing-norm, at the Einstein center, mult. 6,6,6,2) on \(\mathrm{Sym}^2_0T^*K_6\) , and the proton actor is pinned by the sector-orthogonality identity \(\Pi_q M\,\Pi_\ell=0\) . Hole 3 asks: is there a different choice of bundle/connection/domain/readout — on this Stage or a cheaper one — that reproduces every one of these readouts (chirality, three families, no proton-decay mediator, Higgs VEV via Wilson line) at strictly lower actor cost?

 (b) Why it is hard, and the traps. The difficulty is that "actor cost" has never been given a coordinate-free metric — counting connections, endomorphisms, and projectors is trivially gauge-dependent (a change of trivialization can make \(E\) look like zero or like a large matrix without changing physics). The obvious trap is to count named objects in the frozen ledger (there are, by construction, exactly as many as the theory needs) and declare that a lower bound — that is circular, since the count was never taken over an outside competitor class. A second trap is reviving the CP² actor content (SU(3)/U(2) with its Spin \(_c\) index \(r(r+1)/2\) ) as a "cheaper" alternative: CP² is a BRANCH-KILL , refuted at Gate 2 because its U(2) isotropy over-produces gauge bosons — this is closed and must not be reopened as a floor-lowering competitor. A third trap is conflating Hole 3 with Hole 1 (Stage minimality): the two are logically separable — a Stage can be minimal while its actor bundle is not, and vice versa — so a Hole-3 proof must hold the Stage fixed and vary only the \(\otimes\) -layer data.

 (c) What closes it, target-blind, and what refutation looks like. Closure is a role-mechanism normal-form theorem : a canonical (basis-free, connection-gauge-free) reduction of any admissible actor bundle to a minimal generating set of (connection, endomorphism, domain, readout) data, together with a proof that the frozen \(\mathcal{E}_{\rm active}\) already sits at that normal form's minimum subject to reproducing the physical readouts (three chiral families, \(SU(3)_c\times SU(2)_L\times U(1)_Y\) gauge content mod \(\mathbb{Z}_6\) , Wilson-line Higgs with \(n_H=1\) , proton-safety projector identity). The proof must be target-blind : written down without first checking that it returns "13D wins" — i.e., the normal-form procedure and its cost functional must be fixed before it is run on the frozen branch (this is exactly guard G1 from the granularity bridge, reused here). The honest expected outcome, stated in the corpus's own language, is "reduced to minimal-given- \(E\) , with \(E\) named as residual" — not an unconditional uniqueness claim, because the whole actor content is scaffolded on the chiral spectrum \(E\) (Hole 6, below). A refuting result would be a genuinely admissible actor bundle — meeting every one of C_phys's physical burdens under the same Stage — with a strictly smaller normal-form count; if one is found the Actor leg of realization-minimality collapses and Hole 3 becomes a standing reopen (it would not, by itself, touch the already-banked category-relative selector result, which stays DERIVED-GIVEN- \(E\) /CERTIFICATE-CONDITIONAL, but it would foreclose the architecture-neutral upgrade).

 (d) Machinery to start from. The natural starting point is the Peter–Weyl/Bott–Borel–Weil decomposition already in use for \(K_6\) -sector counting ( \(L^2(K_6,E_\mu)=\bigoplus_{(p,q)}V_{(p,q)}\otimes\mathrm{Hom}_{T^2}(V_{(p,q)},E_\mu)\) , with the quadratic Casimir \(C_2(p,q)=\tfrac{p^2+q^2+pq+3p+3q}{3}\) and \(\dim(p,q)=\tfrac{(p+1)(q+1)(p+q+2)}{2}\) ), generalized to a cost functional on Weitzenböck data via the certified endomorphism traces (e.g. \(\mathrm{tr}\,E=5/2\) , \(\mathrm{tr}\,E^2=25/24\) for the Hodge/vector bundle; \(\mathrm{tr}\,E_L=40/3\) , \(\mathrm{tr}\,E_L^2=241/18\) for the graviton TT actor). A representation-theoretic invariant (e.g. a sum of Casimirs/multiplicities weighted by the same MDL bit-cost \(b=\log_2(1/\Delta_0)\) used elsewhere) is the natural gauge-invariant stand-in for "actor count," since traces of powers of \(E\) and \(\Omega=\mathrm{Riem}\) are basis-independent by construction. Hole 4's no-cross-role-compression metric (below) is a prerequisite for making this rigorous — a normal-form theorem without a proof that role-fusion cannot cheat the count is not yet a theorem.

 (e) Leverage. This is placed first because Lemmas 1, 2, and 5 of the realization-minimality upgrade all quote or lean on the actor accounting; a clean Hole-3 closure directly underwrites part of Hole 1 (a Stage cannot be judged "minimal" independent of what actor content it must host) and gives Hole 4 its first non-trivial test case. It is also the hole with the most already-certified raw material (the full Lichnerowicz/Weitzenböck spectra and traces are already exact rationals in hand), making it the cheapest of the open architecture-neutral holes to attempt.

 Hole 1 — Stage minimality

 (a) The precise open object. The claim that "every architecture-neutral Stage carrying the required carriers (color \(SU(3)_c\) , weak \(SU(2)_L\) , hypercharge \(U(1)_Y\) , and chirality with no mirror fermions) has complexity \(\ge\) the frozen Stage \(\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\) " is asserted, not proven, over the full space of admissible carriers — not just the enumerated shelf \(\{K_6={\rm SU}(3)/T^2,\ \mathbb{CP}^2\}\) used for the color rung. This subsumes the whole-SU(3)-carrier-shelf completeness question (named R4/W-B in the residual register): is \(\{K_6,\mathbb{CP}^2\}\) actually the complete list of admissible SU(3) carriers under the frozen search category R2.5, or does R2.5 itself exclude legitimate competitors (string-compactification internal manifolds, NCG spectral-triple discretizations, alternative coset spaces with non-abelian isotropy)?

 (b) Why it is hard, and the traps. The abelian-isotropy uniqueness argument that currently supplies the color rung — \(C_{SU(3)}(T^2)=T^2\) , i.e. \(T^2\) is the unique purely-abelian SU(3) isotropy subgroup, hence \(K_6=SU(3)/T^2\) is the unique clean carrier among coset spaces with abelian isotropy — is a genuine theorem, but it is a theorem conditional on restricting to homogeneous coset spaces with the isotropy already assumed abelian . It says nothing about non-coset internal manifolds (generic Calabi–Yau or G \(_2\) spaces with no continuous isometry group at all, which route gauge symmetry through singularities or brane stacks rather than isometries) — those live outside the "gauge = isometry" category by construction, so R2.5 is fair inside its own rules but the fairness of the rules themselves is exactly what is unproven. The trap is treating "no competitor found on the enumerated shelf" as "no competitor exists" — the corpus is explicit that this is a category-relative result, and elevating it to an architecture-neutral one without doing the completeness proof is precisely the "selection \(\Rightarrow\) forced" relabeling error that is flagged and rejected throughout. A second trap is to under-count the comparison: a "smaller" Stage that fails to independently derive S²'s theorem-level exclusion of abelian carriers for \(SU(2)_L\) (recall: no torus/abelian factor of any dimension has \(SU(2)\) among its isometries, a general theorem, not shelf-limited) does not actually compete — it must be checked against the same general theorems already banked for the weak and hyper rungs, not just against \(K_6\) .

 (c) What closes it, target-blind, and refutation. Closure needs a completeness proof over all admissible SU(3)-carrying homogeneous (or, more ambitiously, all Riemannian) internal factors compatible with R2.5's stated fairness conditions — i.e., a theorem of the form "any compact manifold whose isometry group contains \(SU(3)\) acting to produce a clean color sector is either \(SU(3)/T^2\) or fails a named admissibility clause (over-produces gauge bosons, as CP² does at its U(2) isotropy; or admits mirror fermions; or fails anomaly freedom)" — plus, separately, an explicit fairness argument for why R2.5's "forces = isometries of compact classified internal factors" framing does not unfairly exclude string/NCG-style constructions (or an honest admission that it does, with the comparison redone in a category that includes them). The success criterion is target-blind: the exhaustion must be run over the classification of homogeneous spaces with prescribed isotropy type before checking which one matches \(K_6\) . A refuting result is a named competitor internal factor — carrying \(SU(3)_c\) cleanly, with abelian or otherwise non-gauge-over-producing isotropy, at strictly lower Stage cost — that survives the same admissibility clauses; finding one folds the Shape ladder's headline ~4× margin for that specific comparison (it would not, by the two-axis discipline, retroactively make the current DERIVED-GIVEN-E selector result false — it was always declared category-relative — but it would demote the shelf claim from "complete" to "incomplete, corrected").
This is also explicitly named as this gate's own declared weak link and "intended first review target" — a specialist attacking it is attacking the corpus's self-identified most exposed seam, not a peripheral item.

 (d) Machinery to start from. The classification machinery is the standard theory of compact homogeneous spaces \(G/H\) with \(G\) a compact classified group and \(H\) a closed subgroup, cross-referenced against the centralizer computation already used for \(T^2\) ( \(C_{SU(3)}(T^2)=T^2\) ): the natural next step is to enumerate, for every closed subgroup \(H\subset SU(3)\) up to conjugacy (there are finitely many conjugacy classes of closed subgroups of a compact simple Lie group of this rank — \(T^2\) , \(U(2)\) , \(SO(3)\) , finite subgroups, and \(SU(3)\) itself), which \(H\) give (i) an isotropy representation that does not over-produce gauge bosons at the orbifold/compactification level (the Gate-2 failure mode that killed CP²'s \(U(2)\) ), and (ii) a spin \(^c\) or spin structure compatible with the chirality/no-mirror requirement (the Atiyah–Singer–Patodi machinery already used to get \(n_L=+3,\,n_R=0\) on the \(S^1_Y/\mathbb{Z}_2\) boundary). Each candidate can be scored on the same MDL cost functional (Vol, curvature invariants, isometry-algebra dimension) already tabulated for \(K_6\) (e.g. \(\mathrm{Ric}_i=5/12\) , \(\mathrm{Scal}=5/2\) , \(|\mathrm{Riem}|^2=23/12\) in Killing-norm) so that the comparison is apples-to-apples in the same currency used for the rest of the ladder.

 (e) Leverage. A positive Hole-1 closure would upgrade the color rung from "unique on the enumerated shelf" to "unique over the complete admissible category," directly retiring R4/W-B and strengthening the ~4× economy margin from a first-pass survey (0 REFUTED, 1 FAILS_TO_GENERATE_T, 10 LOSES_TO_13D) to a certified classification. It also feeds Hole 5 (no-preferred-competitor across the five Tier-1 rival programmes): a general homogeneous-space completeness theorem is a substantial chunk of what a fair comparison against string/NCG/SO(10) internal geometries would need.

 Hole 4 — no cross-role compression (the formal metric \(\mathfrak{K}\) )

 (a) The precise open object. Lemma 4 of the realization-minimality argument asserts that no relabeling can move cost from one role (Stage/Rulebook/Actors) to another to make the total look artificially cheaper — "no cross-role compression." This is currently an unformalized claim , illustrated by example rather than proved as a theorem. What is missing is a formal metric \(\mathfrak{K}\) on the layered object \((\times,\oplus,\otimes)\) together with a proof that \(\mathfrak{K}\) is non-decreasing under any role-fusion map (a map that folds, say, a Rulebook clause into an Actor endomorphism, or a Stage isometry into a Rulebook admissibility condition, while preserving the physical content).

 (b) Why it is hard, and the traps. The difficulty is that "role" is a modeling choice, not an intrinsic geometric fact — the same physical content (e.g., the finite admissibility of certain Yukawa textures) could in principle be encoded either as a Rulebook clause (a constraint on which couplings are legal) or as an Actor endomorphism (a projector baked into the bundle data), and nothing in the frozen record currently proves these encodings always cost the same. The trap already caught once here is the "smaller complete generator" rescue attempt — an argument that tried to declare a cross-role-compressed object cheaper by picking a specific scalar total — which is flagged as a caught TARGET-SELECTION-RISK (it implicitly knew which answer to reach) rather than a legitimate fix; any specialist re-approaching this must not reproduce that error. A second trap is conflating this with the separate (and harder) aggregation-metric question of how per-item MDL costs are summed across the three layers (the additive-MDL vs. dimension-first-lex seam, Hole/R3 below, shared with the Granularity gate) — Hole 4 is about role-fusion invariance of whatever aggregation rule is eventually adopted, not about which rule is correct.

 (c) What closes it, target-blind, and refutation. The closing object is a theorem , not an example: a proof that for the declared cost functional (whatever it turns out to be, pending the Hole-5/R3 aggregation resolution), \(\mathfrak{K}(\text{role-fused object}) \ge \mathfrak{K}(\text{original})\) for every admissible role-fusion map in the category, with equality only in degenerate (physically vacuous) cases. Success is target-blind if the fusion maps are enumerated structurally (from the category's own definition of what a "role" is) rather than checked one at a time against the frozen branch to see if they happen to lose. A refuting result is a specific, physically faithful role-fusion that strictly lowers \(\mathfrak{K}\) — that would mean the whole three-layer separation (×/⊕/⊗) is itself an MDL-suboptimal bookkeeping choice, which would undercut not just Hole 4 but the layer-necessity result (B2, every proper subset of \(\{\times,\oplus,\otimes\}\) is inadmissible) from a different angle: layer necessity says you cannot drop a layer; a Hole-4 refutation would say you can profitably merge layers, which is a distinct and more serious failure mode.

 (d) Machinery to start from. Because \(\mathfrak{K}\) does not yet exist, the natural starting point is category theory / operad-style role algebra: define the three roles as functors (Stage: manifold+bundle category; Rulebook: finite admissibility poset; Actors: connection/endomorphism/operator category) and role-fusion as a natural transformation collapsing two of the three functors into one enlarged object in a single category, then ask whether the MDL cost (bit-length of the minimal faithful description, using the same \(b=\log_2(1/\Delta_0)\) per-injected-real currency used for the anchor-cost bookkeeping) is monotonic under that collapse. This is genuinely open mathematics relative to the corpus — there is no partial derivation to extend, only the worked examples (the C1–C10 term-by-term necessity checks) that would serve as test cases the eventual theorem must reproduce.

 (e) Leverage. This is the single theorem with the widest blast radius among the open holes: it "underwrites Lemmas 1, 2, 3, and 5" in the corpus's own accounting. A Hole-4 closure would retroactively upgrade Hole 3 (Actor) and Hole 1 (Stage) from "asserted, illustrated by example" to "proved," and would supply the missing rigor for the Rulebook residual (R-L2a, immediately below) which explicitly closes iff Hole 4 closes. It is also a partial (not full) answer to the R3 aggregation seam shared with the Granularity gate, since a role-fusion-invariance theorem constrains — without fully fixing — which aggregation rules are even admissible.

 Hole 2 — Rulebook residual (REFUTED-as-standalone; correctly relocated, not re-banked)

 (a) The precise open object. Rulebook minimality (Lemma 2) is not an open theorem to prove — it has already been refuted as a standalone claim . The finite admissibility chamber \(\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\) injects at least three unreduced real numbers it does not itself derive — the sector-level Yukawa normalizations \(N_d=2.4\times10^{-2}\) (fixing \(m_b\) at \(M_Z\) ) and \(N_e=1.02\times10^{-2}\) (fixing \(m_\tau\) at \(M_Z\) ), plus a third, structural \(N_\nu\) — while an ordinary 4D Standard-Model rulebook achieves the same physical content with strictly fewer injected constants and is independently admissible. So "the Rulebook is minimal" is false as stated; what remains open is not the false claim itself but where its three sub-residuals land.

 (b) Why it is hard, and the traps. The trap explicitly named in the record is to re-bank rulebook-minimality in some rephrased form — this must not happen; the refutation stands. The genuine difficulty is correctly triaging the three pieces it broke into, because each has a different closure path and conflating them reproduces the original error: (i) part of the gap is really a Hole-4 problem in disguise (is the "extra" Rulebook cost actually just mis-attributed Actor cost that a correct cross-role accounting would relocate?); (ii) part of it is not a Rulebook deficiency at all but simply more of the measured chiral-sector input, which should be honestly folded into the \(E\) -anchor rather than left looking like an unexplained Rulebook injection; (iii) the residual piece needs the same no-preferred-competitor treatment as the rest of the architecture (Hole 5).

 (c) What closes it, target-blind, and refutation. There is no single "closure" for Hole 2 as a unit — each sub-piece closes on its own criterion. R-L2a (the accounting question) closes exactly when Hole 4 closes, by construction — the same role-fusion-invariance theorem that validates or invalidates cross-role cost transfer settles whether the "extra" Rulebook cost was ever real or an artifact of drawing the ×/⊕ boundary in a particular place. R-L2b closes by an explicit, honest declaration that \(N_d,N_e,N_\nu\) are part of the \(E\) -anchor (the measured chiral/Yukawa data), not free Rulebook parameters — this is a bookkeeping correction, not a derivation, and should be stated as such rather than presented as new physics. R-L2c folds into Hole 5's per-class lower-bound comparison. A "refutation" in this branch is not really possible in the usual sense — the standalone claim is already refuted; the only way this residual register entry could get worse is if a specialist tried to re-derive rulebook-minimality directly and produced a new argument with the same flaw (asserting \(N_d,N_e,N_\nu\) are derived when they are not) — that would be a repeat of the fabrication risk this gate is explicitly guarded against.

 (d) Machinery to start from. The starting point is the existing sector-level normalization table (§8.3 of the geometry pack: action ladders \(a_u=(2,1,0)\) , \(a_d=(4/3,2/3,0)\) , \(a_e=(2,4/3,0)\) , \(a_\nu=(1,1/2,0)\) , each with its Boltzmann factor \(\kappa=e^{-\pi\sqrt3}=0.004333420509983131\) raised to the ladder power) together with the explicit binding rule that family-level normalizations \(N_{i,a}\) are forbidden (only sector-level \(N_i\) are permitted) — this rule is precisely what makes the intra-sector hierarchy \(\kappa^{a^{(a)}}\) a genuine prediction rather than a fit, and is the model for how to draw the line between "derived structure" (the \(\kappa\) -ladder ratios) and "measured input" (the three sector-level \(N_i\) ) cleanly, which is exactly the distinction Hole 2's triage needs to make rigorous and repeatable for any future Rulebook object.

 (e) Leverage. Closing R-L2a (via Hole 4) removes the last piece of the Rulebook leg that currently reads as an internal inconsistency (a "refuted lemma" sitting inside a gate marked ANCHORED) and replaces it with a clean statement: the Rulebook is exactly as expensive as it needs to be once its measured inputs are honestly counted as part of \(E\) . That, in turn, tightens the "honest input cost" ledger (currently 4 declared anchors plus 9–10 injected reals, roughly 13–14 reals total) into a single unambiguous number instead of a disputed one.

 Hole 5 — no preferred competitor (Tier-1 rival audit; attempt LAST)

 (a) The precise open object. The claim "no rival architecture-class currently supplies a frozen certificate meeting the same physical burden \(C_{\rm phys}\) at strictly lower cost after unfolding" is checked against five named programmes — string/M/F-theory compactifications, noncommutative-geometry spectral triples, traditional Kaluza–Klein, finite-state/discrete/lattice constructions, and SO(10)/exceptional GUTs. Every one of them reproduces the Standard Model gauge group — that outcome is a tie , not a discriminator, since it is a filter every serious framework passes. What is missing is a finite per-class lower-bound computation: for each rival class, what is the minimum unfolded record-cost (in the same MDL currency) of a certificate meeting \(C_{\rm phys}\) , and is any of the five strictly below the 13D branch's cost?

 (b) Why it is hard, and the traps. This is, by the corpus's own honest assessment, the hole with the lowest odds of ever fully closing — it shades into a universal negative ("no competitor in any of five broad research programmes, now or in the future, will ever produce a cheaper certificate") the more classes are checked, which is why it is placed last in the attack order rather than first. The trap is scope creep: a specialist could spend unbounded effort trying to "solve" string theory's landscape or NCG's classification problem in order to answer this one gate, when the actual deliverable is much narrower — a specific , frozen, auditable certificate from a named competitor construction, scored on the same cost functional, not a general survey of an entire field's literature. A second trap is treating "the SM gauge group comes out" as evidence for one of the five classes over 13D — it is explicitly flagged as a tie that carries zero discriminating weight.

 (c) What closes it, target-blind, and refutation. Full closure — a coverage argument establishing that no member of any of the five classes can beat the 13D branch — is honestly rated low-probability and is not the expected near-term outcome. The tractable, target-blind version of this hole is much sharper: produce one specific, frozen, fully-specified certificate from one of the five rival classes (e.g., a particular NCG spectral triple, or a particular F-theory elliptic fibration with a stated gauge divisor and matter curve content) and score its unfolded record-cost on the identical MDL functional used for the 13D branch, without adjusting the functional after seeing which one wins. If that scored competitor comes in strictly cheaper, that is a direct refutation, and it is stated as such — " Shape folds " for that comparison, a genuine, well-defined defeat condition, not a hedge. If it comes in equal or more expensive, that single class is retired as a live threat (moved from "serious audit candidate, unknown" to "checked, does not beat 13D") without claiming the other four are also settled. The honest expected outcome for the program as a whole , stated in the record, is that this remains CERTIFICATE-CONDITIONAL (current-record) — none of the five currently supplies a beating certificate — rather than a proof of non-existence.

 (d) Machinery to start from. For each rival class the starting point is that class's own best-developed minimal construction reproducing the SM: for NCG, the Chamseddine–Connes spectral-triple algebra \(A_F=\mathbb{C}\oplus\mathbb{H}\oplus M_3(\mathbb{C})\) and its associated finite Dirac operator; for traditional Kaluza–Klein, the classic Witten no-go on getting chiral fermions from smooth compactification (which the frozen 13D branch itself respects via the orbifold ℤ₂-fold rather than a smooth manifold) as the natural point of comparison; for F-theory, a specific elliptically-fibered Calabi–Yau fourfold with an \(SU(3)\times SU(2)\times U(1)\) gauge divisor; for SO(10)/exceptional GUTs, the standard \(16\) -dimensional spinor representation and its breaking chain. Each of these has a well-defined "unfolded record length" once one commits to counting the same categories of object (manifold/bundle data, admissibility rules, operator content) that the 13D ledger counts — the honest work is building that translation dictionary once per rival class, not re-deriving each rival programme from scratch.

 (e) Leverage. Because this hole is explicitly the least tractable and lowest-priority, its main leverage is defensive rather than constructive: closing even one rival-class comparison converts a standing "Unknown" cell in the Tier-1 audit table into a checked "does not beat 13D" cell, incrementally strengthening the ~4× economy margin's credibility without needing to wait for the harder Hole-1 through Hole-4 results. It has essentially no dependency relationship with the other holes — it can be attempted independently, which is exactly why it is safe to leave for last.

 Hole 6 — \(E\) unforced (declare, do not plug)

 (a) The precise open object. The entire stack — the family count \(\chi(K_6,E)=-3\) , the \(\mathbb{Z}_6\) center-kernel structure, the chirality projector \(P_\chi\) , the Wilson-line Higgs mechanism — bottoms out on \(E\) , the Standard-Model chiral matter content (three generations, their gauge quantum numbers, their hypercharges \(Y(Q_L)=+1/6\) , \(Y(u_R)=+2/3\) , \(Y(d_R)=-1/3\) , \(Y(L_L)=-1/2\) , \(Y(e_R)=-1\) , \(Y(H)=+1/2\) ). No principle in this framework, or anywhere in the field, forces this particular \(E\) to be the unique anomaly-free, chirally-consistent spectrum. The anomaly-cancellation filter that \(E\) must pass is a real, checkable constraint — but it has infinitely many solutions (arbitrary numbers of vector-like pairs and alternative chiral assignments also cancel anomalies), so passing that filter does not select \(E\) uniquely.

 (b) Why it is hard, and the traps. This is, honestly, not a "hard but tractable" open problem in the same sense as Holes 1–5 — it is the single deepest unsolved question in the entire beyond-Standard-Model literature ("why this matter content and not another"), named here rather than concealed. The trap, repeatedly flagged in the record, is to claim " \(E\) is forced" — this is explicitly refuted by the existence of infinitely many anomaly-free alternatives, and any argument that tries to smuggle \(E\) -selection in through the back door (e.g., by quietly tuning the geometry's spin \(^c\) structure or Chern class to return \(-3\) rather than deriving that it must) is target-anchoring and is disallowed by the framework's own fabrication guard. A second trap is running the family-count computation ( \(\chi(K_6,E)=-3\) from Bott–Borel–Weil/Spin \(_c\) index theory, weight \((1,0)\) ) and mistaking "three families follow from the geometry" for "the geometry selects \(E\) " — the honest reading is the reverse: \(\chi\) is computed given \(E\) 's spin \(^c\) structure, so the three-family count is read off \(E\) , not derived independently of it.

 (c) What closes it, target-blind, and refutation. The only two honest outcomes are named explicitly. Either (i) someone produces a genuine principle — independent of already knowing the SM's answer — that forces exactly three chiral generations with exactly the observed gauge quantum numbers uniquely among all anomaly-free, chirally-consistent spectra (this is not expected, and would be a result of the first rank in theoretical physics if achieved, not a routine gate closure); or (ii) \(E\) is declared what it already functionally is — an irreducible measured anchor , entering the framework the same way \(M_{\rm Pl}\) , \(\alpha_i(M_Z)\) , \(y_t\) , and \(|V_{us}|\) do, cancelling on both sides of the economy comparison rather than being predicted by it. The record's own expectation, stated plainly, is outcome (ii). The target-blind success criterion for outcome (i), should anyone attempt it, is a derivation that does not reference the SM's known content anywhere in its construction — it must independently output "3 generations, these hypercharges" from a principle stated before the answer is checked. A refuting result for the current declared-anchor treatment would be discovering that some simpler, non-anthropic selection principle (e.g., a stronger global anomaly or index-theoretic rigidity condition not yet checked) actually does rule out all but one \(E\) — that would be a genuine and celebrated closure, converting Hole 6 from MEASURED-ANCHOR to DERIVED; nothing in the current record suggests this is close, and it must not be pre-announced.

 (d) Machinery to start from. The relevant existing machinery is the Atiyah–Singer–Patodi index computation already used for the chirality result on the \(S^1_Y/\mathbb{Z}_2\) boundary ( \(n_L=+3\) , \(n_R=0\) on the interval \([0,\pi]\) ) and the Bott–Borel–Weil/spin \(_c\) index on \(K_6\) giving \(\chi(K_6,E)=-3\) at weight \((1,0)\) — both of these are consequences of \(E\) 's structure, so any attempt at outcome (i) would need to go one level deeper, into global-anomaly/cobordism classification (the same \(\Omega_5^{\rm Spin^c}\) / Dai–Freed machinery already used elsewhere in this framework for the Pin \(^-\) leptogenesis-sign question) to ask whether some additional , not-yet-imposed global consistency condition narrows the infinite anomaly-free family down to a finite or unique set. This is speculative future machinery, not a partially-worked derivation — it is named honestly as such.

 (e) Leverage. \(E\) is explicitly identified as the shared root of the SG-1 (this gate), SG-2, SG-3, and SG-9 chains — meaning a genuine resolution of Hole 6 (outcome (i), however unlikely) would not just strengthen DeepRoot–Shape but would ripple through every downstream gate that currently treats the chiral spectrum as a measured input. Conversely, the correct current move — cleanly declaring \(E\) an irreducible measured anchor rather than leaving it in an ambiguous state — is itself valuable leverage: it is what allows the category-relative selector-minimality result to stand cleanly as DERIVED-GIVEN- \(E\) without overclaiming, and it is what the "two honest testable handles" in the endpoint statement are built on.

 The permanent wall — absolute irreducibility (not a hole to attack)

 (a) The precise open object. The strongest possible version of the Shape claim — "no competitor anywhere , in any conceivable mathematical architecture, is a shorter description of the same physical content" — is a well-defined mathematical statement (a Kolmogorov-complexity minimality claim, \(K(T)\) for the theory \(T\) ) but is a universal negative over the entire space of possible formal systems , and is therefore uncomputable in general (this is the standard undecidability of Kolmogorov complexity, not a limitation specific to this framework). \(T\) also presupposes \(E\) on both sides of any such comparison, compounding the problem.

 (b) Why it is a wall, not a hole, and the trap. The trap is treating this like Holes 1–5 and hunting for a clever proof technique that closes it — none exists, for the same reason no one can compute the Kolmogorov complexity of an arbitrary string in general: it is a mathematical impossibility result, not a research frontier that better tools will eventually cross. The correct terminal is to name it and refuse to treat it as an axiom to be proven — declaring it AXIOM-OPEN is the honest and final move, not a placeholder for future work.

 (c) What "closes" it, and the honest limit. There is no closure in the derivation sense. The only meaningful reformulation available is to weaken the claim to something computable — a grammar-relative forcing result (fixing a specific description language/grammar and asking whether 13D is minimal within that grammar ) — but that is a sharper-OPEN restatement, not a discharge of the original universal claim. This is stated plainly as a shared ceiling on all knowledge, in any field, for any theory — every physical theory ever proposed, in this framework or any other, faces the identical uncomputability wall if pushed to the "absolutely irreducible" standard. It is not a framework-specific gap.

 (d), (e). No machinery closes a Kolmogorov-uncomputable universal negative, and there is no leverage to chase here — the leverage already realized is architectural: by explicitly walling this off as AXIOM-OPEN-by-design, the category-relative result (which is tractable and is banked) is protected from being unfairly held to an impossible standard.

 Cross-cutting note: the one live decisive seam (shared with Granularity)

 Two of the items above — Hole 4 (no-cross-role compression) and, more sharply, the aggregation-rule question underneath it — connect to a single decisive fork that this gate shares with DeepRoot–Granularity: whether the correct cost metric is additive-MDL (per-item description lengths summed across the ×/⊕/⊗ layers, the metric under which the 13D branch wins every rung of the economy ladder) or dimension-first lexicographic (compare raw manifold dimension \(k_{\rm dim}\) first, in which case a 4D effective field theory with \(k_{\rm dim}=4<13\) wins outright, regardless of how many injected reals it needs elsewhere, and the entire ladder folds — for everyone, not just for 13D). No mounted observable currently distinguishes these two metrics — this is a record-blocked ambiguity, not a computational shortfall, meaning more computation on the existing frozen record cannot resolve it; it requires either a new theorem (the granularity \(\Rightarrow\) MDL bridge, promoting the currently-declared axiom to a derived result) or a new observable capable of telling the two metrics apart (referred to in the record as the "B13" economy exhibit). Because this fork decides forced-given- \(E\) vs. merely-selected-given- \(E\) , it is the single highest-leverage open item across both gates: a resolution here would either upgrade the whole Shape ladder's status or concretely explain, for the first time, why the additive-MDL currency — rather than the naively simpler dimension-count — is the physically correct one to use. It is listed here for completeness and cross-reference; its primary ownership and closure path are written up under DeepRoot–Granularity.

 Honest ceiling, scope & the endpoint

 This closing section does one job: state, without hedging in either direction, exactly what has been bought, exactly what has not, and where the argument stops. The fixed grade is REDUCED-TO-AXIOM / ANCHORED +1 , and it is not changed by anything below — this section exists to make that grade legible rather than to argue for a different one. The discipline throughout is the same one that governs the rest of the dossier: state the reached terminal plainly; show a residual when one exists; never fold a shown residual back into a hedge on the terminal itself, and never let a residual masquerade as a bigger claim than it is.

 1. What is explicitly NOT claimed

 Five non-claims sit underneath every number in this dossier. Each one is a claim the corpus's own frozen record actively denies, not merely declines to assert — a reader who came away believing any of them would have overclaimed relative to what was actually shown.

 (a) Selection is not derivation, and dissolution is not the same thing as solved. The central result of this gate is that the object

 \[
\mathfrak{B}_{\rm active}
=
\big[\,\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\,\big]_\times
\ \oplus\
\big[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,\big]_\oplus
\ \otimes\
\big[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,\big]_\otimes,
\qquad D=4+6+2+1=13,
\]

 is the argmin of a declared cost functional over a declared, frozen, enumerated category — eleven named rival shapes spanning \(D=4\) through \(D=12\) plus non-dimensional constructions were actually scored, and the 13D branch wins every scored comparison (0 REFUTED, 1 FAILS_TO_GENERATE_T, 10 LOSES_TO_13D) by an audited margin of roughly \(4\times\) in total description length. That is a selection : the winner of a race that was run, refereed by a metric that was fixed before the race, among competitors that were actually entered. It is categorically different from a derivation , which would mean showing that the 13-dimensional, four-factor structure follows as a logical or dynamical consequence of some deeper principle applied without reference to a competing-candidate search — the way, for instance, the critical dimension of the bosonic string follows from the vanishing of the worldsheet conformal anomaly, independent of any economy comparison against rival dimensions. No such derivation of \(D=13\) or of the specific factorization \(M_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\) exists here or anywhere in the literature, and none is claimed. A selector-minimal object inside a category is not thereby promoted to "the forced answer" — "the funnel's winner is the forced answer" is a relabeling this dossier explicitly refuses to perform, because it would silently convert a scored comparison into an unconditional theorem.

 A closely related but distinct point concerns the word dissolved . Elsewhere in this document — most importantly around absolute irreducibility, item (b) below — a demand is dissolved rather than met: it is shown to be the wrong kind of question to ask of any finite research program (a universal negative over an unbounded space, hence uncomputable), and declaring that boundary honestly is treated as the correct terminal action. Dissolved is not a synonym for solved. A dissolved demand has not been satisfied; it has been correctly identified as unsatisfiable in principle by any method, which is a different and in some ways more honest outcome than either "solved" or "still open, keep trying." Readers should not conflate "the demand for absolute irreducibility was dissolved" with "irreducibility was proven" — the former is what happened; the latter did not and could not happen.

 (b) Given- \(E\) is not derivation-of- \(E\) . Every derived quantity in this dossier that touches the chiral matter content — the family count \(\chi(K_6,E)=-3\) , the Atiyah–Singer–Patodi index \(n_L=+3,\ n_R=0\) on the orbifold interval \([0,\pi]\) , the \(\mathbb{Z}_6\) center-kernel structure with Smith normal form invariant factors \([1,6,6]\) — is computed conditional on a specific bundle datum \(E\) (three chiral generations with their exact Standard Model hypercharge assignments \(Y(Q_L)=+1/6,\ Y(u_R)=+2/3,\ Y(d_R)=-1/3,\ Y(L_L)=-1/2,\ Y(e_R)=-1,\ Y(H)=+1/2\) ) that is imported, not produced . The topological machinery (Bott–Borel–Weil applied to the spin \(^c\) Dirac operator twisted by \(E\) ; the Atiyah–Singer–Patodi index theorem) is a real, checkable derivation of a number from \(E\) ; it is not a derivation of \(E\) . Concretely: the anomaly-cancellation filter that any admissible chiral spectrum must pass admits infinitely many solutions , of which the observed Standard Model spectrum is only one. Nothing in the geometry, the topology, or the admissibility rulebook \(\mathcal{C}_{\rm admiss}\) singles out this particular \(E\) from that infinite family. The statement "the geometry forces exactly three chiral generations with exactly these hypercharges" is refuted by the existence of the other anomaly-free solutions, and this dossier does not print it. What is true, and is printed, is the strictly weaker and fully supported statement: " given \(E\) , the geometry returns \(\chi(K_6,E)=-3\) , i.e. three families, correctly and exactly." The gap between those two sentences is exactly the gap between derivation-of-a-consequence and derivation-of-the-input, and it is the single largest standing residual carried by this gate (registered below as the \(E\) -anchor).

 (c) Not an absolute-irreducibility result. The strongest version of "why 13D" — is there any conceivable architecture, in any framework anyone could ever construct, encoding the same physical content in fewer bits after fully unfolding its hidden bookkeeping — is a universal negative over the unbounded space of all possible formal descriptions . Formally this is a lower bound on the Kolmogorov complexity \(K(T)\) of the target structure \(T\) (the observed gauge group, its chirality, and its generation count), and \(K(T)\) is well known to be uncomputable : no algorithm can certify that a given description of \(T\) is the shortest one possible, because doing so in general would solve the halting problem. There is a further, sharper wrinkle specific to this case: the target \(T\) against which any rival architecture would be measured already presupposes the chiral spectrum \(E\) from point (b) above, so the comparison is not even well-posed independent of the very anchor the gate has already flagged as measured, not derived. This is not a hole in this program specifically; it is a ceiling on any claim of absolute architectural minimality made by any theory in any field, and the correct response — the one this gate takes — is to name the wall and refuse to claim what lies past it, not to leave a placeholder marked "future work." Absolute irreducibility is therefore carried as AXIOM-OPEN / permanent wall , a shared ceiling on all knowledge, not a gap in this reconstruction.

 (d) Not "no Tier-1 competitor exists." The five rival architecture classes surveyed against this gate's own economy metric — string/M/F-theory, noncommutative geometry (spectral triples), traditional Kaluza–Klein, finite-state/discrete/lattice approaches, and \(SO(10)\) /exceptional GUTs — all currently fail to supply a frozen, target-blind certificate meeting the same physical burden ( \(C_{\rm phys}\) : correct gauge group, correct center quotient, correct chirality with no surviving mirrors, proton stability) that is strictly simpler than the 13D branch after fully unfolding their own hidden bookkeeping (landscape/moduli choices for string compactifications, the algebra choice for spectral triples, the internal-manifold choice for traditional KK, the group-embedding choice for GUTs). "Currently fails to supply a certificate" is a certificate-conditional, current-record statement about the state of an audit, not a proof of nonexistence. Each of the five remains, in this dossier's own vocabulary, a serious audit candidate with its "simpler-after-unfolding" cell marked Unknown rather than No . A single one of them producing a frozen, strictly-cheaper certificate under the same declared metric would collapse the selection immediately — that is a live, standing possibility, not a closed door, and the honest odds of the full five-class audit ever completing are assessed as low (the task shades, at the margin, into the same universal-negative territory as point (c)), which is exactly why it is scheduled last in the corpus's own attack order and is explicitly optional, never owed.

 (e) The functional-role floor is a floor, not a uniqueness result. Separately from the economy ladder, the corpus proves a much weaker and much cheaper fact: any architecture meeting the physical burden \(C_{\rm phys}\) must contain a nonempty Stage, Rulebook, and Actors layer — \(k_{\rm role}\ge 3\) , proved by contradiction in three steps and invariant under renotation via the Unfold map. This is near-tautological : it says that whatever meets the requirements must have at least the requirements' worth of structure, which is necessary but nowhere close to sufficient to pin down this 13D, four-factor structure specifically. It rules out zero-layer or one-layer architectures; it does not rule out a rival three-layer architecture built on a completely different Stage. This dossier prints \(k_{\rm role}\ge 3\) as exactly what it is — a floor — and does not lean on it to argue for anything stronger.

 2. The one thing that would be a genuine overclaim, stated as a checklist

 Collecting (a)–(e) into operational form, the following sentences must never appear as conclusions of this gate, and their absence here is deliberate rather than an oversight: "13D is forced"; "13D is the unique consistent shape"; "the Standard Model's three generations are derived from the geometry"; "CP² is a live alternative to \(K_6=SU(3)/T^2\) " (CP² was built out end-to-end and breaks at Gate 2 — its \(U(2)\) isotropy over-produces gauge content, and the earlier claim that it offered a tunable family count is retracted as unsound , since CP² is Spin \(_c\) with a discrete topological index \(r(r+1)/2\) , not a continuous modulus; reviving CP²-tunability is reviving a killed branch); "rulebook-minimality is banked" (Lemma 2 of the realization-minimality attack is refuted as standalone — the flavor chamber \(\mathcal{F}^+\) injects the sector normalizations \(N_d, N_e, N_\nu\) without deriving them, and an ordinary 4D Standard Model rulebook has strictly lower cost and is admissible under the same category, so rulebook-minimality cannot be re-banked without first discharging that refutation); and "no competitor anywhere is shorter" (the dissolved universal negative of point (c)). None of these five sentences is printed as a conclusion anywhere in this dossier, and a reader checking this document against the fixed grade should find that discipline held throughout.

 3. The anchors paid — the full, itemized bill

 The selection argument is not free, and its cost is stated in full rather than only in the headline \(\{M_{\rm Pl},\alpha_i(M_Z),y_t,|V_{us}|\}\) form. Four items are paid as irreducible physical anchors, one item is paid as a declared bridge axiom, and one item is paid as a residual whole-shelf completeness gap. Laid out explicitly:

 Anchor set 1 — the four irreducible physical inputs. \(M_{\rm Pl}=1.220900000000000\times10^{19}\) GeV (ordinary, not reduced, Planck mass), \(\alpha_i(M_Z)\) (the three gauge couplings at the \(Z\) pole, \(M_Z=91.1876\) GeV), \(y_t\) (top Yukawa), and \(|V_{us}|\) (the Cabibbo angle magnitude). These four numbers are the only inputs the corpus treats as genuinely free real-valued data — every radius, volume, curvature invariant, Casimir, and chamber operator quoted elsewhere in this dossier (the exact rational curvature invariants of \(K_6\) at the Einstein center, \(\mathrm{Ric}_i=5/12\) , \(\mathrm{Scal}=5/2\) , \(\|\mathrm{Ric}\|^2=25/24\) , \(\|\mathrm{Riem}\|^2=23/12\) ; the compactification radius \(R_0=1.591549430918954\times10^{-17}\ \mathrm{GeV}^{-1}\) ; the volume \(\mathrm{Vol}(X_{\rm active})=3.704417261398702\times10^{-148}\ \mathrm{GeV}^{-9}\) ; the derived scale \(M_*=7.467050992135091\times10^{16}\) GeV) is derived or exact-topological from these four plus the frozen geometry, not independently free.

 Anchor set 2 — the honestly-charged injected reals. The corrected accounting (§3.3 of the underlying ledger) adds nine-to-ten further injected reals that the flavor chamber \(\mathcal{F}^+_{\rm finite}\) requires and does not itself derive: the sector normalizations \(N_d=2.400000000000000\times10^{-2}\) , \(N_e=1.020000000000000\times10^{-2}\) , and \(N_\nu\) (structural, fixing overall neutrino-sector magnitude); the threshold triple \((\delta_1,\delta_2,\delta_3)=(+4.8424,-3.1112,-1.7313)\pm1.6\times10^{-3}\) ; and the Hosotani phase \(\theta_H^\star\) fixing the Wilson-line vacuum. With these included, the honest total charged cost is roughly 13–14 reals , not the 4 anchors alone — and the resulting economy margin against the ten scored rivals is the corrected ~4× , not the overstated "~18×" ("4→22") headline that an earlier, less careful pass through the ledger printed. This dossier holds to the smaller, defensible number throughout.

 Bridge axiom — AXIOM-GRANULARITY-MDL-BRIDGE. The entire selection argument reduces to exactly one declared, named posit: the Finite Operational Cell Law (the existence of an operational resolution floor \(\Delta_0>0\) ) makes every physical description a finite bit-string, "cost" is defined as description length (MDL) evaluated at that resolution, with per-injected-real anchor cost \(b=\log_2(1/\Delta_0)\) , and per-item costs from different sectors of the theory aggregate additively into a single common currency . This axiom is declared target-blind (it is checked against the guard that it would be written the same way without foreknowledge that 13D should win) but it is declared, not derived — no proof is offered, or claimed, that the additive-MDL aggregation rule is the uniquely correct way to compare architectures. This is the one bridge every part of the economy argument crosses, and naming it here, rather than letting it stay implicit inside "the metric," is what makes the REDUCED-TO-AXIOM grade honest rather than aspirational: the stack does not bottom out in an uncounted pile of assumptions, it bottoms out in this one, stated plainly enough that a reader could attack it directly. A live rival exists and is not hidden: under a dimension-first lexicographic metric instead of additive MDL, a bare four-dimensional effective field theory wins outright ( \(k_{\rm dim}=4<13\) ) regardless of how many hidden reals it must inject elsewhere, and the entire eleven-rival ladder scored in this dossier folds. Which of the two metrics is the physically correct one is not decided here; it is the single decisive, currently open seam separating "selected under one declared and defensible metric" from "forced independent of metric choice," and it is shared with, and exported to, the Granularity deep-root gate rather than re-litigated in this one.

 Measured anchor — the chiral spectrum \(E\) . As established in point (b) above, the Standard Model's chiral matter content — three generations with their exact hypercharge assignments — is consumed as an irreducible measured anchor at the bottom of the entire stack. It is not derived, it is not selected by the economy argument (it is fixed identically on both sides of every comparison the ladder runs, so it cancels out of the margin rather than being explained by it), and " \(E\) is forced" is refuted by the existence of infinitely many other anomaly-free chiral spectra. This is the single deepest residual the gate carries, named plainly rather than buried: the whole shape-selection argument answers "given this matter content, what is the cheapest complete carrier," not "why this matter content."

 Residual — whole-shelf completeness of the color carrier. The abelian-isotropy uniqueness argument that selects \(K_6=SU(3)/T^2\) — the centralizer \(C_{SU(3)}(T^2)=T^2\) is the unique purely-abelian \(SU(3)\) isotropy, so \(K_6\) is the unique clean \(SU(3)\) carrier — is proved over the enumerated shelf \(\{K_6,\ \mathbb{CP}^2\}\) only , with \(\mathbb{CP}^2\) eliminated as a genuine branch-kill (breaks at Gate 2). Whether that two-member shelf is a fair and exhaustive enumeration of admissible \(SU(3)\) -carrying candidates under the declared search category, or whether some excluded competitor would win under the same metric, is not settled here; it is carried as an acknowledged open flank (the corpus's own "intended first review target"), not folded into the color-selection result as if it had already been closed.

 4. What would move this gate, and what would not

 Two concrete, falsifiable handles sit on the table, and naming them is itself part of the honest accounting rather than a rhetorical flourish. First: name one architecture meeting the identical physical burden — the correct gauge group with its correct \(\mathbb{Z}_6\) center quotient, the correct chirality with no surviving mirror fermions, proton stability — that scores strictly cheaper than the 13D branch after fully unfolding its own hidden bookkeeping, under the same declared additive-MDL metric. If such a certificate is produced by any of the five rival classes, the selection collapses immediately; this is a real, currently-open door, not a rhetorical one. Second: produce an actual principle that forces exactly three chiral generations with exactly the observed hypercharges uniquely — not merely permits them, as the anomaly filter does — and the \(E\) -anchor promotes from measured to derived, strengthening the whole stack. Absent either, and none is expected imminently by this dossier's own honest assessment, the gate remains exactly where it is graded. Two things that would not move this gate, because they target claims never made: producing a slightly cheaper carrier under a different, undeclared cost metric changes nothing about the result scored under the stated metric (it would instead feed directly into the still-open metric-selection seam, R5, which is exactly where such a challenge belongs); and pointing out that \(E\) is unforced is not a refutation of anything printed here, because " \(E\) is a measured anchor" is already the stated position, not a hidden assumption being exposed for the first time.

 5. The closing endpoint statement

 Every leg of this gate has reached a terminal. Selector-minimality inside the frozen category is DERIVED-GIVEN- \(E\) , certificate-conditional on the frozen branch and the declared metric. The functional-role floor is DERIVED as a floor, not a uniqueness result. Realization-minimality (the architecture-neutral upgrade) is OPEN, optional, not owed. Absolute irreducibility is AXIOM-OPEN by design, a dissolved universal negative, not a re-opened question. The chiral spectrum \(E\) is a MEASURED-ANCHOR, terminal by anchoring. Nothing further is owed to hold the grade at its fixed value.

 Nothing left. Anchored on: Shape : the frozen 13D carrier \(\mathcal{M}_4\times K_6(=SU(3)/T^2)\times S^2\times S^1_Y/\mathbb{Z}_2\) , all three layers (×-Stage, ⊕-Rulebook, ⊗-Actors) carried together as the object under test; Granularity : the finite operational cell law and cost-floor \(\Delta_0>0\) , the load-bearing primitive that turns every description into a finite bit-string and supplies the per-item anchor cost \(b=\log_2(1/\Delta_0)\) ; Scale : the UV boundary package \(M_U\approx1.0\times10^{16}\) GeV, \(R_0=1.591549430918954\times10^{-17}\ \mathrm{GeV}^{-1}\) , and \(M_{\rm Pl}=1.220900000000000\times10^{19}\) GeV, supplying the size of each charged cost rather than the deciding rule; Observables : the four irreducible anchors \(\{M_{\rm Pl},\ \alpha_i(M_Z),\ y_t,\ |V_{us}|\}\) plus the honestly-charged nine-to-ten injected reals ( \(N_d,N_e,N_\nu\) , the threshold triple \((\delta_1,\delta_2,\delta_3)\) , \(\theta_H^\star\) ) for a total near 13–14 reals, together with the measured chiral spectrum \(E\) consumed identically on both sides of the comparison; Dissolution : the demand for absolute uniqueness across every conceivable architecture in all of mathematics is a universal negative equivalent to an uncomputable Kolmogorov-complexity lower bound, so "selector-minimal under a declared metric, not forced" is the honest ceiling on this question for any research program, not a hedge particular to this one.

 Closure ledger — DeepRoot — Shape (13D selector)

 Status (fixed): REDUCED-TO-AXIOM · ANCHORED +1

 The technical closure LEDGER (separate document)

 Gate: deeproot-shape — "DeepRoot — Shape (13D selector)"
 Fixed grade (given, not re-derived here): REDUCED-TO-AXIOM / ANCHORED +1. Board verb: SG-1 = SELECTED BY CONSTRAINTS — the most economical complete carrier by a stated margin (~4×), selected, not proven unique.

 This ledger is the auditor's record: object identity, root stack, anchor table, the numbered derivation chain with exact values, the credit-ladder grade of every leg, the anti-claims, and the endpoint line. It is written so a working physicist can check every number without leaving the page.

 1. Layer-0 wall identity — the object under test

 The frozen active branch is a three-layer object; only the ×-layer carries metric dimension.

 \[
\mathfrak{B}_{\rm active}=\underbrace{[\,\mathcal{M}_4\times K_6\times S^2\times S^1_Y/\mathbb{Z}_2\,]_\times}_{\times\ \text{STAGE},\ D=4+6+2+1=13}\ \oplus\ \underbrace{[\,\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\,]_\oplus}_{\oplus\ \text{RULEBOOK},\ 0\text{-dim}}\ \otimes\ \underbrace{[\,\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\,]_\otimes}_{\otimes\ \text{ACTORS},\ 0\text{-dim}}
\]

 \(K_6=SU(3)/T^2\) — full \(A_2\) flag manifold, 6D, carries color.

 \(S^2\) — round 2-sphere, carries weak \(SU(2)_L\) .

 \(S^1_Y/\mathbb{Z}_2\) — hypercharge circle folded by \(\theta\mapsto-\theta\) , carries \(U(1)_Y\) and chirality.

 \(D=13\) is the metric dimension of the \(\times\) -layer only; the \(\oplus\) and \(\otimes\) layers are non-metric (0-dim) but load-bearing and can never be silently dropped — Lemma 2 (Rulebook) is refuted specifically at the \(\oplus\) layer and Lemma 3 (Actors) lives at the \(\otimes\) layer (§4 below), so any "residual" computed from the \(\times\) -layer alone is an artifact of truncation, not a property of the frozen object.

 Frozen-branch identity is fixed by construction: content-level identity , manifest meta-hash (33 rows), orbifold freeze ``. (Recorded here only to mark that the object under test is a single frozen artifact, not a moving target — no hash content is load-bearing physics.)

 Gauge routing (the physical content the Shape carries): \(SU(3)_c\) from left-isometries of \(K_6\) ; \(SU(2)_L\) from isometries of \(S^2\) — and by the general theorem F1, no abelian/torus factor of any dimension can host a non-abelian isometry, so \(SU(2)_L\) cannot originate on any circle/torus; \(U(1)_Y\) from \(S^1_Y/\mathbb{Z}_2\) , with the \(\mathbb{Z}_2\) fold simultaneously supplying chirality (general theorem F2: a bare closed circle leaves both handedness sectors, producing mirror fermions excluded by the LEP \(Z\) -width measurement; the fold removes them).

 2. Layer-1 endpoint anchor

 The endpoint this gate reduces to is stated once, precisely, so every leg below can be checked against it:

 Anchored on: Shape = the frozen 13D carrier \(\mathcal{M}_4\times K_6(=SU(3)/T^2)\times S^2\times S^1_Y/\mathbb{Z}_2\) (the object under test) · Granularity = the finite-bit record + cost-floor \(\Delta_0\) (where the scoring rule that ranks candidate shapes actually lives) · Scale = the UV boundary package ( \(M_U\approx10^{16}\) GeV, \(R_0=1.591549430918954\times10^{-17}\,\mathrm{GeV}^{-1}\) ) supplying the size of each per-item cost · one named axiom : the granularity \(\Rightarrow\) MDL common-currency bridge (declared, not derived).

 This is a three-root anchoring (Shape + Granularity + Scale), not a single-root one — the Shape gate cannot be evaluated without borrowing a cost metric from Granularity and a cost size from Scale. That borrowing is exactly what collapses to the one declared axiom in §4.3.

 3. Layer-2 root stack

 3.1 Tier A — deep roots, full precision, load-bearing test

 Root 
 Role in this gate 
 Load-bearing? 
 Full-precision content supplied 

 Shape 
 This is the shape gate: supplies the 3-layer object, the three carriers, and the \(\mathbb{Z}_6\) quotient under test 
 ✓ 
 \(D=13\) ; \(K_6=SU(3)/T^2\) ; Ricci/Riemann invariants §3.3 below 

 Granularity 
 The Finite Operational Cell Law ( \(\exists\,\Delta_0>0\) ) forces the description-length / record-cost metric that the economy argument is built on; it is silent on aggregation — the one live seam 
 ✓ 
 \(b=\log_2(1/\Delta_0)\) anchor cost per injected real 

 Scale 
 Sets the size of \(b\) via the UV boundary package; supplies \(M_U\) , \(R_0\) — the cost size , not the deciding root 
 ✓ 
 \(M_U\approx1.0\times10^{16}\) GeV; \(R_0=1.591549430918954\times10^{-17}\,\mathrm{GeV}^{-1}\) ; \(M_{\rm Pl}=1.220900000000000\times10^{19}\) GeV 

 Physical equivalence / invariance 
 Makes the complexity metric encoding-invariant and role-counts frame-independent 
 ✓ 
 Invariance screen (Tier B, §3.2) 

 Record interface 
 Frozen branch + certificate stack must be reproducible / blind-reproducible 
 ✓ 
 Record-interface screen (Tier B, §3.2) 

 Nonseparability 
 Forces the full 3-layer object to stay unflattened — the \(\times\) -layer metric geometry alone is not "the Shape" 
 ✓ 
 Nonseparability screen (Tier B, §3.2) 

 Causal order 
 Not primary load-bearing for the Shape selector 
 ✗ 
 — (screened only as a pass/fail check, §3.2) 

 Complete-root discipline: the attack used the full three-layer object on the frozen branch, full-precision \(M_{\rm Pl}\) , and Granularity evaluated across all three layers — never a truncated ( \(\times\) -only) object. This matters concretely: a critic who computes "cost" only over the four metric factors and ignores \(\mathcal{F}^+_{\rm finite}\oplus\mathcal{C}_{\rm admiss}\) and \(\mathcal{E}_{\rm matter}\oplus\mathcal{E}_{\rm gauge}\oplus\mathcal{E}_{\rm Higgs}\oplus\mathcal{E}_{\rm proton}\) is scoring an artifact, not the branch; the honest input-cost correction in §5 exists precisely because the \(\oplus\) -layer's injected reals were at risk of being dropped from the ledger.

 3.2 Tier B — screens (pass/fail checks on the attack itself, not on the physics)

 Screen 
 Verdict 
 Detail 

 Invariance 
 PASS, with one flagged EXPOSE 
 The complexity metric is encoding-invariant and role-counts are frame-independent, except : the phrase "smaller complete generator" is a non-representation-neutral formulation that privileges a scalar-total ordering — flagged, not silently used. 

 Record Interface 
 PASS, with one flagged seam 
 Finite, reproducible artifacts throughout, except the aggregation-rule seam (additive-MDL vs. dimension-first-lex, see §4.3) is RECORD-BLOCKED: no mounted observable currently distinguishes the two rules (internally referenced as exhibit "B13"). 

 Causal Order 
 PASS across the board 
 The chiral spectrum \(\mathcal{E}\) is imported as a declared anchor upstream of the selector computation — it is never back-fitted as a target-loaded rule. This is the guard against "target-blind" violations (see Prime-Directive discipline: never back-solve a number to a desired answer). 

 Nonseparability 
 EXPOSE 
 Lemma 4 ("no-cross-role compression" — you cannot cheat the cost count by re-packing Stage content into Rulebook or Actors) is stated but not yet formalized as its own no-smuggling metric. Named as residual R3/Hole 4 in §6. 

 Net Tier-B verdict: 2 clean PASS (Causal Order fully; Record Interface modulo one flagged seam), 1 PASS-with-flag (Invariance), 1 EXPOSE (Nonseparability) — none of the four screens fails outright, and the one EXPOSE is exactly the seam that the single declared axiom (§4.3) is built to carry openly rather than hide.

 4. Measured anchors and their role

 Anchor 
 Exact value 
 Role in THIS gate 
 Consumed / Reproduced / Tested-against 

 \(M_{\rm Pl}\) (ordinary) 
 \(1.220900000000000\times10^{19}\) GeV 
 Fixes \(M_*\) via Planck normalization \(M_{\rm Pl}^2=M_*^{11}\cdot{\rm Vol}(X_{\rm active})\) , \(D=13\) 
 Consumed (one of the 4 irreducible anchors; not reproduced by this gate) 

 \(\alpha_i(M_Z)\) ( \(i=1,2,3\) ) 
 GUT-normalized; unification residual \(9.6\times10^{-11}\) at \(M_U\) 
 Anchors the gauge-coupling routing check used to identify \(\mathfrak{su}(3)_c\oplus\mathfrak{su}(2)_L\oplus\mathfrak{u}(1)_Y\) as the surviving 4D algebra 
 Consumed (declared anchor, not a first-principles prediction of this gate) 

 \(y_t\) 
 (top Yukawa, sets \(N_u=1\) chamber normalization) 
 Feeds the honest input-cost ledger (§5) as one of the 4 irreducible anchors 
 Consumed 

 $ 
 V_{us} 
 $ 
 (Cabibbo angle) 

 \(E\) = SM chiral content (3 generations, full rep content) 
 \(\chi(K_6,E)=-3\) given \(E\) 
 The bottom of the entire stack — every economy comparison is a comparison of carriers of the same \(E\) , so \(E\) cancels on both sides 
 Tested-against / declared MEASURED-ANCHOR — never reproduced, never claimed forced. " \(E\) is forced" is REFUTED: the anomaly-freedom filter alone admits infinitely many solutions, so geometry is not picking \(E\) out of a unique slot. 

 \(N_d,N_e,N_\nu\) (sector normalizations, \(\mathcal{F}^+\) ) 
 \(N_d=2.400000000000000\times10^{-2}\) , \(N_e=1.020000000000000\times10^{-2}\) , \(N_\nu\) structural 
 Counted as injected reals in the honest cost ledger, not free-floating extras 
 Consumed , declared part of the \(E\) -anchor bundle 

 Threshold triple \((\delta_1,\delta_2,\delta_3)\) 
 \((+4.8424,-3.1112,-1.7313)\pm1.6\times10^{-3}\) 
 Not itself consumed by the Shape argument, but is the mechanism by which \(M_U\) closes; included so the reader can see the Scale root is not hand-waved 
 Consumed upstream (feeds \(M_U\) , which feeds \(b\) ) 

 Only four quantities are the irreducible anchors of the whole 13D construction: \(\{M_{\rm Pl},\ \alpha_i(M_Z),\ y_t,\ |V_{us}|\}\) . Every geometric constant quoted in §3.3/§6 below (radii, volumes, curvature invariants, Casimirs) is derived from these four plus the frozen shape, not independently injected — with the single, explicitly named exception of \(E\) itself (the chiral spectrum), which is a fifth, separately-declared measured anchor at the foot of the whole chain, and the flavor-chamber normalizations \(N_d,N_e,N_\nu\) , which are declared as part of that same \(E\) -anchor bundle rather than smuggled in as free geometric parameters.

 5. Full derivation chain — numbered ledger, each step with its exact value

 Step 0 — Object declaration. \(\mathfrak{B}_{\rm active}\) frozen as in §1; \(D=4+6+2+1=13\) .

 Step 1 — Weak rung, S² (general theorem F1). No abelian/torus carrier of any dimension hosts a non-abelian isometry group ⇒ \(SU(2)_L\) cannot arise from any circle/torus factor. This closes an entire shelf of cheaper candidates in one theorem, not by exclusion of named rivals. Value/verdict: \(SU(2)_L\leftarrow S^2\) (isometry group of the round 2-sphere), never \(SU(2)\subset SU(3)\) . Grade: DERIVED (hand-checkable general theorem).

 Step 2 — Hyper rung, \(S^1_Y/\mathbb{Z}_2\) (general theorem F2). A closed odd-dimensional bare-circle factor preserves both chiralities ⇒ mirror fermions ⇒ excluded by the measured LEP \(Z\) -width. The \(\mathbb{Z}_2\) fold \(\theta\mapsto-\theta\) removes the mirror sector. Atiyah–Singer–Patodi index on \([0,\pi]\) : \(n_L=+3,\ n_R=0\) . Grade: DERIVED (hand-checkable index computation + measured exclusion of mirrors).

 Step 3 — Color rung, \(K_6=SU(3)/T^2\) (abelian-isotropy uniqueness). Centralizer \(C_{SU(3)}(T^2)=T^2\) (Cartan torus only) ⇒ \(T^2\) is the unique purely-abelian \(SU(3)\) isotropy subgroup. On the enumerated shelf \(\{K_6,\ \mathbb{CP}^2\}\) , \(K_6\) is the unique clean SU(3) carrier. \(\mathbb{CP}^2=SU(3)/U(2)\) is a certified BRANCH-KILL : it breaks at Gate 2 because its \(U(2)\) isotropy over-produces gauge content, and the earlier claim that \(\mathbb{CP}^2\) offers a tunable family count is retracted as unsound ( \(\mathbb{CP}^2\) is \(\mathrm{Spin}_c\) with a discrete topological integer index \(r(r+1)/2\) , not a continuous modulus — there is nothing to tune). Value: \(C_{SU(3)}(T^2)=T^2\) exactly. Grade: DERIVED-GIVEN-E , banked over the enumerated shelf only (whole-shelf completeness over every conceivable SU(3) carrier is residual R4, §6).

 Step 4 — Family count (Bott–Borel–Weil / Spin \(_c\) index). \(\chi(K_6,E)=-3\) at spin \(_c\) weight \((1,0)\) ⇒ three chiral families, given \(E\) . Value: \(\chi(K_6,E)=-3\) exactly (topological integer). Grade: DERIVED-GIVEN-E — do not upgrade to "geometry forces 3 families" or "geometry selects \(E\) "; \(\chi\) is computed from \(E\) , not independent of it.

 Step 5 — Center-kernel charge structure. \(\mathbb{Z}_6=\ker(Z(G_0)\to\mathrm{Aut}(E))\) . Smith normal form of the charge-character matrix has invariant factors \([1,6,6]\) , annihilator \(\mathbb{Z}_6\) . Generator \(z=(\omega_3,-1,\zeta_6)\) , order 6. \(G_{\rm SM}=(SU(3)\times SU(2)\times U(1))/\mathbb{Z}_6\) is the finest faithful quotient . Electric charge \(Q=T_3+Y\) , \(Y\in\tfrac16\mathbb{Z}\) . Value: invariant factors \([1,6,6]\) ; \(\mathbb{Z}_6\) exact. Grade: DERIVED-GIVEN-E. 

 Step 6 — Functional-role floor. Any architecture meeting the physical burden \(C_{\rm phys}\) must contain a nonempty Stage, Rulebook, and Actors: \(k_{\rm role}\ge3\) , proved by contradiction in 3 steps, invariant under notation via the Unfold map. Value: \(k_{\rm role}\ge3\) . Grade: DERIVED — but flagged near-tautological : it restates the constraint clauses (two of the constraint-vector items — freeze-before-compare, certificate discipline — are this programme's own methodology, not universal physics), so it is a floor (necessary), never a sufficiency result for architecture uniqueness.

 Step 7 — Layer necessity (B2). Within the declared category, every proper subset \(L\subsetneq\{\times,\oplus,\otimes\}\) has null-space \(N_L=\emptyset\) — i.e., dropping any one layer fails the admissibility test. Grade: DERIVED within category (explicitly not a universal no-go).

 Step 8 — Term-level necessity (C1–C10). Every named term of the active branch is load-bearing: removing any one fails at least one of the ten required admissibility gates. Grade: DERIVED-CONDITIONAL (auditable case-by-case claims, not a closed-form theorem).

 Step 9 — Metric selection / anti-fitting. Complexity \(C(B)=I(B)=\) minimum description length \(+O(1)\) , with anchor cost \(b=\log_2(1/\Delta_0)\) charged per injected real. A tuned normalization costs \(\approx b\) bits under this rule, which is exactly what makes the whole \(\kappa\) -ladder (§6 of the geometry pack; the chamber Boltzmann factor \(\kappa=e^{-\pi\sqrt3}=0.004333420509983131\) that sets inter-generation hierarchy ratios) non-circular — a free parameter is not "free," it is charged at the anchor-cost rate. Grade: DERIVED-CONDITIONAL , conditional on the bridge axiom of §4.3 below.

 Step 10 — The economy ladder (the survey proper). Under the MDL / description-length metric, tallying every considered rung from \(D=4\) through \(D=12\) plus non-dimensional rivals: 0 REFUTED, 1 FAILS_TO_GENERATE_T (a 6D rival structurally cannot carry the required target content at all), 10 LOSES_TO_13D (loses on cost). The 13D branch wins every considered rung. Value: public margin ≈ 4× economy advantage as the most economical complete carrier (complete = carries all of \(E\) , not a truncated subset). Grade: DERIVED-SURVEY — banked as a first-pass survey, explicitly not a certified exhaustive classification; the closing object (a role-mechanism normal-form / exhaustion theorem, internally "SHAPE Certificate 3 / Lemma 5") is OPEN (§6, Hole 5).

 Step 11 — Honest input-cost correction. The naive "4 anchors only" headline undercounts. The full charged cost is ≈ 4 anchors + 9–10 injected reals ≈ 13–14 reals total : the 4 irreducible anchors \(\{M_{\rm Pl},\alpha_i(M_Z),y_t,|V_{us}|\}\) , plus the sector normalizations \(N_d,N_e,N_\nu\) (from \(\mathcal{F}^+\) ), plus the threshold \(\delta\) -triple \((\delta_1,\delta_2,\delta_3)\) (three reals), plus the Hosotani phase \(\theta_H^\star\) . Value: \(\approx13\) – \(14\) total charged reals. This corrects and replaces an earlier, overstated "4→22 outputs ⇒ ~18×" headline — the honest margin is the same ~4× quoted in Step 10, not ~18×. Grade: DERIVED-CORRECTION (the gate's own self-audit, disclosed as residual R3, not swept under the rug).

 Step 12 — Planck normalization consistency check. \(M_{\rm Pl}^2=M_*^{11}\cdot{\rm Vol}(X_{\rm active})\) , \(D=13\) , with \({\rm Vol}(X_{\rm active})=3.704417261398702\times10^{-148}\,\mathrm{GeV}^{-9}\) ⇒ \(M_*^{11}=4.023836152402511\times10^{185}\,\mathrm{GeV}^{11}\) ⇒ \(M_*=7.467050992135091\times10^{16}\) GeV. Grade: DERIVED (arithmetic consequence of \(M_{\rm Pl}\) + the frozen volume; \(M_*\) is fixed by geometry + \(M_{\rm Pl}\) , not an independent input).

 Step 13 — Absolute-irreducibility check (deliberately not attempted as a derivation). "No competitor anywhere, in any conceivable mathematics, is shorter" is a universal negative over all architectures and is provably uncomputable (Kolmogorov complexity \(K(T)\) is not itself computable, and \(T\) presupposes \(E\) on both sides of any comparison). Grade: AXIOM-OPEN / permanent wall , refused-as-axiom by design — this is a shared ceiling on all knowledge in any field that uses a description-length argument, not a defect specific to this programme.

 6. Credit-ladder grading of every leg

 Leg 
 Statement 
 Grade 

 Weak rung ( \(S^2\Rightarrow SU(2)_L\) ) 
 General theorem F1: no torus hosts non-abelian isometry 
 DERIVED 

 Hyper rung ( \(S^1_Y/\mathbb{Z}_2\Rightarrow\) chirality) 
 General theorem F2 + LEP exclusion of mirrors 
 DERIVED 

 Color rung ( \(K_6=SU(3)/T^2\) ) 
 Abelian-isotropy uniqueness over enumerated shelf \(\{K_6,\mathbb{CP}^2\}\) 
 DERIVED-GIVEN-E (shelf-bounded) 

 \(\mathbb{CP}^2\) candidate 
 Over-produces gauge content at Gate 2; tunability claim retracted 
 CLOSED-NEGATIVE (branch-kill, do not revive) 

 Family count \(\chi(K_6,E)=-3\) 
 Bott–Borel–Weil / Spin \(_c\) index at weight \((1,0)\) 
 DERIVED-GIVEN-E 

 \(\mathbb{Z}_6\) center-kernel structure 
 Smith normal form \([1,6,6]\) 
 DERIVED-GIVEN-E 

 Functional-role floor \(k_{\rm role}\ge3\) 
 Proof by contradiction, 3 steps 
 DERIVED (near-tautological floor, not sufficiency) 

 Layer necessity (B2) 
 Null-space \(N_L=\emptyset\) for every proper subset 
 DERIVED within category 

 Term necessity (C1–C10) 
 Each term load-bearing against ≥1 gate 
 DERIVED-CONDITIONAL 

 Metric selection / anti-fitting 
 \(C(B)=\) MDL \(+O(1)\) , \(b=\log_2(1/\Delta_0)\) 
 DERIVED-CONDITIONAL (on the bridge axiom) 

 Economy ladder (~4× survey) 
 0 refuted / 1 fails-to-generate / 10 loses-to-13D 
 DERIVED-SURVEY , not exhaustive classification 

 Honest input-cost (≈13–14 reals) 
 Correction of the overstated "~18×" headline 
 DERIVED-CORRECTION (self-audit, disclosed) 

 Realization-minimality (architecture-neutral) 
 5 sub-lemmas, Lemma 2 refuted-standalone 
 OPEN (optional, not owed) 

 Absolute irreducibility 
 Universal negative, uncomputable \(K(T)\) 
 AXIOM-OPEN / permanent wall (dissolved unicorn) 

 Chiral spectrum \(E\) 
 Bottom of the stack; anomaly filter alone admits infinitely many solutions 
 MEASURED-ANCHOR (terminal by anchoring; "forced" refuted) 

 \(M_4\) as the comparison stage 
 Declared axiom, the observational primitive 
 REDUCED-TO-AXIOM (disclosed, correctly labeled) 

 Granularity \(\Rightarrow\) MDL common-currency bridge 
 The one axiom the whole gate reduces to 
 REDUCED-TO-AXIOM (the named bridge axiom, §4.3) 

 Whole-shelf SU(3)-carrier completeness 
 Only \(\{K_6,\mathbb{CP}^2\}\) enumerated, not all conceivable carriers 
 OPEN 

 No-cross-role compression (Lemma 4) 
 Unformalized nonseparability claim 
 OPEN 

 Rulebook minimality (Lemma 2) 
 \(\mathcal{F}^+\) injects \(N_d,N_e,N_\nu\) it does not derive; ordinary 4D SM rulebook has lower \(k_{\rm rule}\) 
 REFUTED-as-standalone (residuals relocated, not re-banked) 

 Tier-1 competitor audit (5 rival classes) 
 None currently supplies a strictly-simpler frozen certificate 
 CERTIFICATE-CONDITIONAL (current-record, not nonexistence) 

 Gate-level roll-up: every leg above has reached some terminal (DERIVED / DERIVED-GIVEN-E / DERIVED-SURVEY / MEASURED-ANCHOR / REDUCED-TO-AXIOM / CLOSED-NEGATIVE / AXIOM-OPEN-by-design), and the entire stack funnels through exactly one declared, undischarged axiom (the granularity \(\Rightarrow\) MDL bridge). That is the textbook definition of REDUCED-TO-AXIOM . The still-open items (realization-minimality, whole-shelf completeness, Lemma 4's formal metric, the Tier-1 competitor audit) are genuinely open but are each individually OPTIONAL upgrades to a stronger claim, never owed debts blocking the terminal already reached — none of them is "the thing REDUCED-TO-AXIOM is waiting on."

 7. The named bridge axiom, stated once, precisely

 AXIOM-GRANULARITY-MDL-BRIDGE. The cost-floor / Finite Operational Cell Law ( \(\exists\,\Delta_0>0\) ) makes every description a finite bit-string; "cost" is defined as description length (MDL) evaluated at the operational resolution \(\Delta_0\) ; and per-item costs from different roles ( \(\times\) -Stage geometry, \(\oplus\) -Rulebook admissibility clauses, \(\otimes\) -Actors bundle data) aggregate as a common currency (straight addition of bit-costs across layers). This axiom is declared target-blind by construction — guarded by a kill-test ( \(\kappa^3/\pi\) , denoted G1: the metric counts only if it would have been written down without knowing that 13D should win) — and is declared, not derived. 

 Why this is the single load-bearing posit: every "13D wins" statement in Step 10 holds only under this additive-MDL aggregation rule. Under a rival dimension-first lexicographic metric — rank candidates first by raw dimension \(k_{\rm dim}\) , only break ties by MDL — a clean 4D effective field theory wins outright ( \(k_{\rm dim}=4<13\) ) regardless of how many injected reals the 4D theory needs, and the entire economy ladder folds for everyone, not just for 13D. Which aggregation rule is the physically correct one is the one still-live, undecided seam (named R5/R3 in the residual register, and shared verbatim with the Granularity gate as the single seam the two gates hold in common). It is disclosed here, in the open, as the whole weight the REDUCED-TO-AXIOM grade is carrying — not hidden and not smuggled into a "derived" step.

 8. Anti-claims and negative controls

 These are things the frozen record actively denies ; printing any of them as proved would be a fabrication regardless of how natural they sound in prose.

 NOT "13D is forced / uniquely derived / the unique minimum." Selector-minimal \(\ne\) forced. Selection \(\ne\) derivation. Relabeling "funnel-winner" as "forced" is rejected outright.

 NOT absolutely irreducible. The status is category-relative to the declared search category (R2.5: forces = isometries of compact classified internal factors + admissible bundles), not architecture-neutral over all conceivable mathematics.

 NOT "the SM chiral content \(E\) / three generations is derived." \(E\) is the irreducible measured anchor at the stack bottom; " \(E\) is forced" is refuted because the anomaly-freedom filter alone admits infinitely many solutions.

 NOT "no Tier-1 competitor exists." Only "none currently supplied a frozen certificate strictly simpler after unfolding." All five rival classes (string/M/F-theory, noncommutative geometry, traditional Kaluza–Klein, finite-state/discrete, SO(10)/exceptional GUTs) remain serious audit candidates — a failure-to-find, never a no-go. Recovering the SM gauge group is a tie every one of these frameworks passes, not a framework discriminator.

 NOT "the functional-role floor proves the architecture." \(k_{\rm role}\ge3\) is near-tautological — necessary, not sufficient.

 NOT reviving \(\mathbb{CP}^2\) -tunability — certified branch-kill, retracted as unsound (the index is a discrete topological integer, not a continuous modulus).

 NOT re-banking rulebook-minimality (Lemma 2) — refuted-as-standalone; its residuals are relocated (§6), never silently re-closed.

 NOT printing the overstated "4→22 outputs / ~18×" headline as the economy margin — the audited honest margin is ~4× (Step 11).

 NOT claiming "no smaller complete generator exists" in a representation-neutral way — flagged in the Invariance screen (§3.2) as privileging a scalar-total ordering.

 Negative controls (numeric, hand-checkable, never to drift): \(\|\mathrm{Riem}\|^2/\mathrm{Scal}^2\) on \(K_6\) at the Einstein center is confirmed \(23/75\) ; it is never \(31/147\) , and \(\|\mathrm{Riem}\|^2\) is never \(=60\) (that value belongs to the round unit \(S^6\) , a different, larger-curvature space used only as a calibration control). \(\|\mathrm{Ric}\|^2/\mathrm{Scal}^2=1/6\) exactly. These three numbers are the standing tripwire against silently swapping in the wrong compact factor.

 9. Full-precision geometric constants this gate rides on (Killing-norm unless stated)

 Quantity 
 Exact value 

 \(D\) 
 \(4+6+2+1=13\) 

 \(\mathrm{Ric}_i(K_6)\) 
 \(5/12\) 

 \(\mathrm{Scal}(K_6)\) 
 \(5/2\) 

 \(\mathrm{Scal}/\mathrm{Ric}_i\) 
 \(6=\dim K_6\) (metric-scale invariant, holds in both normalizations) 

 \(\|\mathrm{Ric}\|^2\) 
 \(25/24\) 

 \(\|\mathrm{Riem}\|^2\) 
 \(23/12\) 

 \(\|\mathrm{Riem}\|^2/\mathrm{Scal}^2\) 
 \(23/75\) 

 \(\|\mathrm{Ric}\|^2/\mathrm{Scal}^2\) 
 \(1/6\) 

 \(\chi(K_6)\) 
 \(6\) (= $ 

 \(\chi(S^2)\) 
 \(2\) 

 \(\chi(S^1_Y/\mathbb{Z}_2)\) 
 \(1\) 

 Invariant Einstein metrics on \(SU(3)/T^2\) 
 \(4\) total: normal \((1,1,1)\) + Kähler–Einstein \((1,1,2)\) and its 3 permutations 

 \(M_U\) 
 \(\approx1.0\times10^{16}\) GeV (unification residual \(9.6\times10^{-11}\) ) 

 \(R_0=(2\pi M_U)^{-1}\) 
 \(1.591549430918954\times10^{-17}\,\mathrm{GeV}^{-1}\) 

 \(M_{\rm Pl}\) (ordinary) 
 \(1.220900000000000\times10^{19}\) GeV 

 \(\mathrm{Vol}(X_{\rm active})\) 
 \(3.704417261398702\times10^{-148}\,\mathrm{GeV}^{-9}\) 

 \(M_*\) 
 \(7.467050992135091\times10^{16}\) GeV (derived, not input) 

 Threshold triple \((\delta_1,\delta_2,\delta_3)\) 
 \((+4.8424,\ -3.1112,\ -1.7313)\pm1.6\times10^{-3}\) 

 \(C_{SU(3)}(T^2)\) 
 \(=T^2\) exactly (abelian-isotropy uniqueness) 

 \(\chi(K_6,E)\) 
 \(-3\) (three chiral families, given \(E\) ) 

 \(\mathbb{Z}_6\) SNF invariant factors 
 \([1,6,6]\) 

 APS index on \([0,\pi]\) 
 \(n_L=+3,\ n_R=0\) 

 10. The endpoint line

 The gate is at its terminal: REDUCED-TO-AXIOM / ANCHORED +1. Every constituent leg has independently reached a terminal — DERIVED (weak/hyper rungs, layer/term necessity, role floor), DERIVED-GIVEN-E (color rung, family count, center structure), DERIVED-SURVEY (the ~4× economy ladder), DERIVED-CORRECTION (the honest ~13–14-real input cost), MEASURED-ANCHOR ( \(E\) , terminal by anchoring, never claimed forced), REDUCED-TO-AXIOM ( \(M_4\) as the comparison stage; and the granularity \(\Rightarrow\) MDL bridge as the one live posit the whole gate funnels through), CLOSED-NEGATIVE ( \(\mathbb{CP}^2\) , branch-killed), and AXIOM-OPEN-by-design (absolute irreducibility, a dissolved universal-negative unicorn, a ceiling on all knowledge rather than a hole in this one). The two honest, standing, falsifiable handles that would move this gate are: (a) name one architecture meeting the identical physical burden that is strictly simpler after unfolding — and the ~4× economy claim folds; (b) produce a genuine principle that forces exactly three chiral generations uniquely — otherwise \(E\) remains, correctly, a measured anchor rather than a derived output. Neither has been supplied. The gate stands anchored, not forced, and is reported here exactly at that grade.