SG-2 — Gauge recovery SU(3)xSU(2)xU(1): full dossier — rendered package. Rendered from DOSSIER_SG2_FULL.md; frozen technical content unchanged by rendering.

SG-2 — Gauge recovery SU(3)xSU(2)xU(1): full dossier

What this is. The full, specialist-grade dossier for SG-2 (Gauge recovery — the Standard-Model gauge algebra from internal isometries) on the frozen 13D K₆ branch. It expands the brief closure-attack packet into a believable, checkable, reusable treatment: the rigorous mathematics behind the recovery, the insights that produced it, the certificates and witnesses that back it, an honest map of every open hole, and — most load-bearing — a concrete specialist work plan for closing each hole.

Binding discipline (carried verbatim). STATUS-UPGRADES:0. This dossier reflects the gate's honest current status — DERIVED-GIVEN-E (held) — and never upgrades it. Frozen branch dcc66f1b2685 / manifest meta a5b1e6f9d951 are READ-ONLY. The gauge-group OUTCOME is a rival TIE; the genuine framework-internal contribution is the carrier-forcedness, not the outcome. given-E ≠ derivation of E. AXIOM-CLOSED ≠ proven; selection ≠ derivation; dissolved ≠ solved.

Sources synthesized: the closure-attack dossier …/rendered/TOE/PER_GATE_DOSSIERS/DOSSIER_SG2_GAUGE_RECOVERY_CLOSURE_ATTACK.md; the validator report …/PER_GATE_DOSSIERS/_gate_attack_2026-06-27/reports/SG2_report.json; the re-grade ledger …/rendered/TOE/GATE_REGRADE_BESTCASE_AND_HOLES_2026-06-29.md (SG-2 block); the specialist hole queue …/rendered/TOE/SPECIALIST_HOLE_QUEUE_2026-06-29.md; and the manuscript …/rendered/GUT/GUT.md (§5.1, §3.5, Appendix GS, Appendix D, Appendix C2/C3). Every number, hash, and verdict below is traceable to one of these files.

Published-history note (current board status). The status language in this dossier is a frozen working audit captured on the 2026-06-27 / 2026-06-29 pre-ratification branch, preserved verbatim as the contemporaneous record of the closure work. On the current board (33 requirement-gates: all 33 RESOLVED at +0 · 0 anchored · 0 open, ratified 2026-07-08; the live /gates/ ledger + per-gate dossiers are the source of truth), SG-2 (gauge recovery SU(3)×SU(2)×U(1)) is RESOLVED at +0 (CERTIFIED-IRREDUCIBLE). The frozen per-hole work items and projected endpoints below are the discipline that produced that closure, shown openly — read them as history, not as the current grade.

1. Executive summary + honest status

1.1 Headline

Feed the frozen 13D shape into one rule — forces are the shape's own symmetries — and the Standard Model's exact gauge algebra falls out by equality: 8 + 3 + 1 = 12 generators, rank 4, not one extra and not one missing. The colour force is the symmetry of the flag manifold K₆ = SU(3)/T², the weak force the symmetry of the two-sphere S², hypercharge the symmetry of a folded circle S¹_Y/ℤ₂. No coupling value and no fitted number is consulted in making this read — it is architectural, read off the geometry alone.

1.2 The honest grade (matches the live popup)

Field Value
Gate id SG-2 — Gauge recovery: SU(3)_c × SU(2)_L × U(1)_Y from internal isometries
Status label (binding) DERIVED-GIVEN-E — direction held (re-grade DERIVED-GIVEN-E → DERIVED-GIVEN-E, source: re-grade ledger SG-2 line)
Validator endpoint SG2_GAUGE_RECOVERY_DERIVED_GIVEN_E_WITH_OPEN_CARRIER_WALLS; descent status CONVERGED_WITH_TERMINAL_WALLS; gate status OPEN/wall (two terminal walls remain); STATUS-UPGRADES:0 (source: SG2_report.json)
Frozen hashes it rides Branch dcc66f1b2685 / manifest meta a5b1e6f9d951. Carrier geometry K_gauge = K₆ × S² × S¹_Y with K₆ = SU(3)/T² (R1.2 / R1.4 isometry data). Parity table ac4d2df3e708 (R1.3) is an SG-3/SG-4 object, co-read only because S¹_Y/ℤ₂ is the hyper carrier.
Anchors The observed spectrum E (the SM gauge content as given-E, plus the LEP/SLD light-species count 2.984 ± 0.008 forbidding mirror fermions) and the measured couplings α_i(M_Z) (declared anchors, NOT Gate-2 outputs — handled at SG-7).
Holes 4 open holes (re-grade ledger); 2 terminal walls in the validator (W_R2_NEUTRALITY, W_R3_COMPLETENESS).

1.3 What this dossier establishes and what it does not

Establishes (given-E, given the selected & frozen geometry): the surviving 4D isometry algebra of the frozen compact factors equals 𝔰𝔲(3)_c ⊕ 𝔰𝔲(2)_L ⊕ 𝔲(1)_Y — equality, not containment. The witness is the simple-summand multiset {8, 3, 1}, rank 4, with no extra unbroken factor and no missing SM factor (the gate's two failure modes). Beyond the bare recovery, three forcedness results — category-internal (their architecture-neutrality is asserted, not proven; see "Does not establish" below) — are the gate's genuine framework-internal content: (C1) K₆ is the unique clean SU(3) carrier by abelian-isotropy uniqueness; (C2) S² is forced for weak by fact F1; (C3) S¹_Y/ℤ₂ is forced for hypercharge by fact F2 — with the cheaper rival carrier CP² = SU(3)/U(2) explicitly killed (built end-to-end and shown to BREAK).

Does not establish, and does not claim: that the SM gauge group is a framework-discriminating result (it is a rival TIE — string, M-, F-theory, noncommutative geometry, and lattice constructions all recover it). That this gauge group is the unique output across all geometries (the route is existence + equality on one frozen branch; given-E ≠ derivation of E). That the carrier-forcedness results are architecture-neutral theorems rather than theorems inside the declared "forces = isometries" category (their neutrality is asserted, not proven — the two terminal walls). It also makes no coupling-unification numerics (that is SG-7).

This is the cardinal honest line, and it is the gate's integrity spine: the OUTCOME is a tie; the discrimination is the carrier-forcedness.


2. The community gap

2.1 The precise open problem

The Standard Model's gauge group, SU(3)_c × SU(2)_L × U(1)_Y, is postulated. The three factors are written in by hand; nothing in the SM explains why there are exactly these three, why their dimensions are 8, 3, and 1, or why the rank is 4. A genuine unified theory is expected to do better — to produce the gauge group from a single deeper structure, rather than assume it.

This is a half-century-old aspiration. Grand unification (Georgi–Glashow SU(5), SO(10), E₆, Pati–Salam) embeds the SM factors in a single larger simple group, but the larger group is itself postulated, and the breaking pattern down to the SM is a separate input. Kaluza–Klein and string/M/F-theory constructions geometrize the gauge group — they read it from the isometries of, or the structure groups of bundles over, a compact internal space — but the compact space (or the bundle, or the brane configuration) is a model choice, tuned to land on the desired group.

2.2 The honest community truth: recovering the group is a tie

Here is the discriminating fact a working physicist must hold onto: recovering SU(3) × SU(2) × U(1) is a constraint that essentially every serious framework satisfies. String constructions arrange it through the structure groups of gauge bundles and intersecting branes; M-theory on G₂ manifolds and F-theory on elliptic fourfolds engineer it through singularity types; Connes–Chamseddine noncommutative geometry produces it from the spectral data of a finite algebra; lattice and deconstruction models build it directly. The outcome therefore discriminates nothing — a theory that reproduces the SM gauge group has passed a filter every credible competitor also passes (source: SG-2 closure-attack dossier §0.1, §1.2 firewall table; re-grade ledger SG-2 "gap" block).

So the community gap is not "produce SU(3) × SU(2) × U(1)" — that is solved many times over. The real, unmet gap is sharper: produce the gauge group from a forced shape — a structure that is not freely tuned to the answer, where the carrier of each force is the unique admissible one rather than one engineering choice among many. None of the rival frameworks derives the group from a forced carrier; they all retain the freedom to arrange it. This is the precise crack SG-2 attacks: not the outcome, but the forcedness of the carriers within a declared grammar.

2.3 State of the art and why prior attempts fall short

Framework How it recovers the group Why it does not close the forced-shape gap
GUT (SU(5)/SO(10)/E₆) embed SM in a postulated simple group the larger group + breaking pattern are inputs; not forced
String / heterotic bundle structure groups on a Calabi–Yau the bundle/CY is a model choice; gauge sourced from bundle, not from a forced isometry
M-theory (G₂) / F-theory singularity types of the compact space singularity structure is engineered to the target
Noncommutative geometry spectral data of a finite algebra the finite algebra is chosen; the "almost-commutative" ansatz is the input
Lattice / deconstruction gauge factors built in directly the construction is the assumption

Every entry recovers the outcome; none turns the carrier into a forced object. SG-2's distinctive move (§3) is to commit to one grammar — gauge forces are the continuous isometries of the compact factors — and then ask, inside that grammar, which shapes are the admissible carriers, and is the survivor unique? That question has a sharp, finite, Lie-theoretic answer, and it is what the gate banks.


3. The construction — rigorous mathematics

Full derivation authority: GUT.md §5.1 (narrative), §3.5 (the menu of internal geometries), Appendix GS (candidate datasheets + facts F1/F2/F3), Appendix D (formal certificate), Appendix C2/C3 (carrier dossiers). This section is the worked, attack-grade construction; it shows the steps, it does not gesture at them.

3.1 The grammar and the pipeline

The category-defining translation of the program (Dictionary row 1; Definition R2.5; GUT.md §2.9): the 4D gauge group is generated by the continuous isometries of the internal compact factors — the Kaluza–Klein mechanism, with gauge bosons appearing as the KK modes of the isometry Killing vectors. Gate 2 is the root gate: Gates 3, 4, 5, 7, 8, 9, 10 all consume its output (source: closure-attack dossier §1.1).

The recovery is one architectural read:

 K_gauge = K6 × S² × S¹_Y   --isometry algebra-->   g_geom = su(3) ⊕ su(2) ⊕ u(1)
   --quotients/parities/bundle data act-->   g_SM = su(3)_c ⊕ su(2)_L ⊕ u(1)_Y   (EQUALITY)

The load-bearing steps, each named so a reader can attack it:

3.2 Step 1 — carrier identification (the existence leg)

Each compact factor sources exactly one simple summand of the gauge algebra, by the standard isometry algebras of homogeneous spaces:

Gauge bosons are the KK modes of these Killing isometries (Appendix D.1). Attack handle: this is one factor → one summand by construction of K_gauge; the existence leg is given-the-selected-geometry.

3.3 Step 2 — the equality clause (no-extra / no-missing)

The certified claim is equality, not containment. Summing the simple summands:

This is the whole force of the gate: containment ("the algebra contains the SM factors") would be trivially satisfiable by almost any large carrier; equality is the non-trivial statement, and it is exactly what the G02_gauge_recovery certificate's multiset test and no-extra-summand check verify (source: closure-attack dossier §1.1 item 2; GUT.md §5.1 "Surviving / frozen object", line 1735: "simple-summand multiset {𝔰𝔲(3),𝔰𝔲(2),𝔲(1)}, 8+3+1=12 generators, rank 4, no extra summand"). The over-production check is precisely where the cheaper CP² carrier dies (§3.6).

3.4 Step 3 — abelian-isotropy / CSDR centralizer rule (carrier-forcedness leg C1)

Under coset dimensional reduction (CSDR), a homogeneous-space carrier G/H gauges G only if the isotropy subgroup H does not itself act as gauge. The 4D unbroken gauge group is governed by the centralizer C_G(H) (the part of G commuting with the isotropy embedding); a non-trivial centralizer-gauge means the isotropy is "gauge-active" and adds surviving gauge.

The decisive structural fact for colour, G = SU(3):

Abelian-isotropy uniqueness (C1): the maximal torus T² is the unique purely-abelian (maximal-rank) SU(3) isotropy — any larger connected isotropy contains a non-abelian factor. Therefore K₆ is the unique clean SU(3) carrier. This is the gate's genuine framework-internal contribution: not "the SM has SU(3)" (everyone has that), but "this carrier is the clean one, and the cheaper rival is not." Attack handle: "gauge-active = non-trivial centralizer" is a grammar rule; its architecture-neutrality is the open wall R2 (§6.1).

3.5 Step 4 — Fact F1 forces S² for weak (carrier-forcedness leg C2)

GS.2 Fact F1 — abelian geometries carry no non-abelian force. The isometry group of any torus T^n is U(1)^n ⋊ (discrete), which is abelian; an abelian isometry group contains no SU(2). Therefore no torus and no torus orbifold has SU(2) among its isometries (source: GUT.md §3.5 shelf table, lines 2716–2719: tori T^n and T^6/Γ eliminated by "gauge recovery (F1)"). The non-abelian weak force cannot be carried by any abelian factor; it requires a genuinely non-abelian-isometry carrier, and S² = SU(2)/U(1) is the minimal one (dim 2, Isom = SU(2)).

Attack handle: F1 is a clean classification fact inside the category. A string framework routes around it via bundle structure groups — a different category — so F1's architecture-neutrality is conditional (§6.3). For SG-2 within the grammar, the consequence is sharp: S² is forced as the minimal weak carrier.

3.6 The CP² over-production kill (the genuine, non-tie elimination)

This is the part of SG-2 that is not a tie — a framework-internal elimination a rival framework could not write. The cheaper SU(3) carrier candidate is CP² = SU(3)/U(2) (complex projective plane, dim 4). Its isotropy U(2) = (SU(2) × U(1))/ℤ₂ is non-abelian, hence gauge-active. Under the CSDR centralizer rule the carrier then faces a lose-lose fork, stated disjunctively in the manuscript (GUT.md §3.5):

  1. Keep the extra structure → an unwanted SU(2) × U(1) survives from the gauge-active isotropy → the equality clause fails (an extra unbroken factor) → Gate-2 fails (over-production); or
  2. Drop it → the matter is isotropy-locked, weak/hyper forced into colour → axiom A1.4 violated (the matter-routing rule).

Separately (and downstream), CP²'s family count is a tunable bundle choice killed at Gate-4/chirality — but the Gate-2 elimination above does not rely on that and is therefore the stronger, architecture-internal kill.

The adversarial witness. The program built the CP² swap end-to-end as its own 11D adversarial test (run id wsmnjvt55), and the route BREAKS at Gate 2 by over-producing gauge (source: re-grade ledger SG-2, line 80: "the CP² swap was built end-to-end (run wsmnjvt55) and BREAKS at Gate 2 by over-producing gauge — a structural exclusion, not a tunable-family hand-wave"). This is the corpus's own falsification attempt against itself, and the cheaper carrier failed it. This is the discriminating move: the κ³/π firewall test ("could a rival framework write this exact sentence?") returns No for the CP²-kill, where it returns Yes for the bare outcome (closure-attack dossier §4.1 firewall table).

3.7 Step 5 — Fact F2 forces the fold S¹_Y/ℤ₂ for hypercharge (carrier-forcedness leg C3)

GS.2 Fact F2 — closed odd-dimensional factors produce no net chirality. The chiral index of a Dirac operator on a closed odd-dimensional manifold vanishes identically. So a bare S¹_Y mirrors every fermion: it produces a left-right symmetric (mirror) spectrum (source: GUT.md §3.5 sphere shelf, line 2703: S¹ "None (F2): both handednesses survive → mirror partners ... Eliminated as-is — chirality (mirrors). Rescued by folding → S¹/ℤ₂").

The mirror sector would be visible in the LEP/SLD light-species count, measured at 2.984 ± 0.008 — which shows no mirror partners (source: closure-attack dossier §0; re-grade ledger SG-2, line 129). The repair is the ℤ₂ fold: S¹_Y/ℤ₂ has fixed-point boundaries, and boundaries re-open the chirality channel via the Atiyah–Patodi–Singer (APS) index. The hypercharge U(1)_Y is the translation generator along the parent circle, quantized by its topology.

Attack handle: F2 forces the fold for the chirality reason, which is properly SG-3/SG-4 content; for SG-2 the relevant deliverable is only the carrier identity — the hyper carrier is the folded circle S¹_Y/ℤ₂, not the bare one, and U(1)_Y survives. (Cross-evidence from SG-3: the APS one-sided index on S¹_Y/ℤ₂ returns (n_L, n_R) = (+3, 0); the bare-S¹ control returns (+3, +3) — proving the fold is load-bearing; source: re-grade ledger SG-3, line 175.)

3.8 The full shelf — why the survivors are forced and the rivals killed

The construction is not "K₆, S², S¹ chosen because they work." The manuscript runs the whole homogeneous shelf through F1/F2/F3 and the equality clause, and records the elimination reason for each (GUT.md §3.5, Appendix GS, lines 2691–2719; closure-attack dossier §4.3):

Carrier dim Isom Verdict Reason
K₆ = SU(3)/T² 6 SU(3) Retained clean SU(3) carrier; abelian isotropy (C1)
CP² = SU(3)/U(2) 4 SU(3) Eliminated non-abelian U(2) isotropy → over-produces gauge (Gate-2) or isotropy-locks (A1.4); built & BROKE (run wsmnjvt55)
2 SU(2) Retained minimal non-abelian carrier; required for weak (F1)
S¹_Y/ℤ₂ 1 U(1) Retained hyper carrier; fold re-opens chirality (F2)
bare S¹ 1 U(1) Eliminated as-is F2 mirrors; rescued by folding
S³ ≅ SU(2) 3 SO(4) Eliminated F2 chirality + Occam vs S² (extra dim, extra SU(2), no gate served)
S⁵ ≅ SU(3)/SU(2), S⁷ 5, 7 SO(6), SO(8) Eliminated F2 chirality + wrong surviving group (over-large isometry)
Lens S³/ℤ_n 3 ⊂ SO(4) Eliminated F2 chirality
ℝP^n = S^n/ℤ₂ n quotient O(n+1) Eliminated admissibility (D7, n even) or chirality (n odd)
tori T^n, T^6/Γ n, 6 U(1)^n abelian Eliminated gauge recovery (F1): abelian → no non-abelian force
Calabi–Yau 3-fold, K3 6, 4 none Eliminated F3: no continuous isometries → no forces

The pattern a reader internalizes after F1–F3: even-dimensional spheres can carry handedness, odd-dimensional ones cannot (F2); a sphere's symmetry grows with dimension, usually past what the SM wants; abelian shapes carry no non-abelian force (F1); shapes with no isometry carry no force (F3). The selector is meant to feel mechanical, not curated (GUT.md line 2693).

3.9 Representation / charge recovery (deferred to Gates 3–5)

Appendix D.2–D.3 assign each surviving multiplet its (SU(3), SU(2), Y, Q) from projector data, with Q = T₃ + Y verified componentwise and the global ℤ₆ identification gluing the three centers ([SU(3) × SU(2) × U(1)]/ℤ₆). SG-2 itself is blind to the global quotient — the ℤ₆ is invisible to the Lie algebra and binds only representations. The rep-level recovery and centers are SG-3/SG-4 content and are cited here, not claimed here (source: closure-attack dossier §1.1 item 6).

3.10 What "closed" means here, and its conditionality

"Closed" for SG-2 is a rigid algebra-equality recovery + three carrier-forcedness theorems, all category-internal — strictly not a cross-geometry uniqueness proof, not "the SM gauge group is forced by nature," not a coupling-unification result. It is conditional on three things:

The genuine, defensible content (survives the honest accounting):

Genuine content (framework-internal, category-relative) Mechanism
Surviving algebra equals 𝔰𝔲(3) ⊕ 𝔰𝔲(2) ⊕ 𝔲(1) (no extra/missing) isometry algebra of K₆ × S² × S¹_Y; multiset + no-extra-summand check
K₆ = SU(3)/T² is the unique clean SU(3) carrier abelian-isotropy uniqueness; C_SU(3)(T²) = T² adds no gauge
forced for non-abelian weak fact F1
folded S¹_Y/ℤ₂ forced for hypercharge, no mirrors fact F2 + LEP/SLD 2.984 ± 0.008
CP² over-produces gauge → fails Gate-2 OR isotropy-locks (A1.4) U(2) non-abelian isotropy; built end-to-end and BROKE

The conditional / non-genuine content (the rival-TIE and cross-geometry leg):

Quantity Honest reality
"the SM gauge group SU(3) × SU(2) × U(1)" rival TIE — string/M/F/NCG/lattice all recover it; not framework-discriminating
"this gauge group is unique" NOT claimed — given-E; existence + equality on one branch
coupling values α_i(M_Z) declared anchors, not Gate-2 outputs; unification is SG-7
centers / ℤ₆ / charge tables deferred to Gates 3–5; SG-2 is Lie-algebra-blind to the global quotient

4. The insights we used

These are the moves that made the progress believable and reproducible — now shared at working-physicist depth so a reader can both check them and build on them.

4.1 Forces = isometries: commit to one grammar, then ask "is the carrier forced?"

The pivotal reframe is not trying to out-engineer the rivals at producing the outcome (a losing game — everyone produces it). It is to commit to a single grammar (gauge = continuous isometries, KK) and then ask the sharp, finite question the grammar makes possible: among the shapes the grammar admits, which can carry each force, and is the survivor unique? The grammar converts a vague "produce the group" ambition into a bounded Lie-theoretic classification problem with a yes/no answer. This is why SG-2 can bank something the rivals cannot: a forcedness claim, not an outcome claim.

4.2 Abelian-isotropy uniqueness — the one structural fact that does the work

The whole carrier-forcedness for colour rests on one clean piece of Lie theory: the maximal torus is the unique connected abelian subgroup of maximal rank, and its centralizer in G is itself. For SU(3), C_SU(3)(T²) = T² (Cartan only), so the CSDR reduction off K₆ = SU(3)/T² survives exactly 𝔰𝔲(3) and no more. Any larger isotropy contains a non-abelian factor, is gauge-active, and over-produces. This single fact — deformation-proof, number-free, hand-checkable — is the gate's spine. It is why "K₆ is the clean carrier" is a structural statement, not a curated preference.

4.3 The equality discipline — containment is trivial, equality is the gate

A subtle insight that separates SG-2 from a weak "the algebra contains the SM" claim: the gate's content is equality. The over-production check (no extra unbroken factor) is what kills the cheaper rivals; the under-production check (no missing factor) is the existence leg. Stating the gate as equality is what makes it falsifiable — find a surviving extra U(1) or a missing SU(2) and the gate falls — and it is precisely the predicate the G02 certificate machine-checks.

4.4 Build your own adversary — the CP² end-to-end break

The most credibility-bearing insight is methodological: rather than assert CP² fails, the program built the CP² swap end-to-end (run wsmnjvt55) as an adversarial witness and let it break at Gate 2 by over-producing gauge. A theory that builds its own cheapest rival and shows it failing is doing the opposite of curve-fitting — it is the structural-exclusion analogue of a controlled negative experiment. This is why the CP²-kill is bankable as framework-internal where the bare outcome is not.

4.5 The firewall test (κ³/π discipline at the framing level)

The honesty insight that keeps the gate from over-claiming: for every Gate-2 claim, ask "could a rival framework write this exact sentence?" If yes, it is a TIE leg (label it shared); if no, it is genuine framework-internal. Run on SG-2: "the low-energy algebra is 𝔰𝔲(3) ⊕ 𝔰𝔲(2) ⊕ 𝔲(1)" → Yes (shared); "K₆ is the unique clean SU(3) carrier by abelian isotropy" → No (framework-internal). This test is what forbids banking the OUTCOME as a framework win — the cardinal overclaim — and it is the gate's integrity spine (source: closure-attack dossier §4.1).

4.6 Architectural read with no number consulted — the un-tunable property

A property worth its own line: the Gate-2 read consults no coupling value and no UV number. The α_i(M_Z) are declared anchors handled at SG-7, not Gate-2 outputs; the validator's negative controls confirm the gate does not smuggle them in (NC_GROUP_TARGET_LOADED_SCOPE, NC_M_R_AS_ANCHOR, NC_TARGET_ANCHOR all pass; source: SG2_report.json negative_control_results). Because the gate consults no fitted number, it cannot be quietly tuned — there is nothing to dial. This is what makes the falsifiable bet (§5.4) honest rather than a hedge.


5. Evidence & reproducibility

5.1 The witness ledger (each leg graded + reproduces flag)

# Witness Asserts Grade Reproduces?
W1 Isometry-algebra identification (K₆→𝔰𝔲(3), S²→𝔰𝔲(2), S¹_Y→𝔲(1)) each carrier sources one simple summand hand-checkable Yes — standard homogeneous-space isometry algebras
W2 Generator / rank count 8+3+1=12, rank 4 multiset exactly {8,3,1} hand-checkable Yes — dim SU(3)=8, SU(2)=3, U(1)=1; rank 2+1+1=4
W3 Equality clause (no extra, no missing) G02 multiset + no-extra-summand passes machine-lane AUDIT / BLOCKED — certificate folder referenced; input absent on disk
W4 Abelian-isotropy uniqueness (C1) T² unique purely-abelian SU(3) isotropy; K₆ unique clean carrier symbolic Yes — max torus is unique connected abelian subgroup of max rank
W5 Fact F1 S² forced for weak hand-checkable Yes — Isom(T^n)=U(1)^n abelian; contains no SU(2)
W6 Fact F2 bare S¹_Y mirrors; fold forced hand-checkable Yes — odd-dim closed Dirac index vanishes; APS boundary re-opens chirality
W7 CP² over-production verdict U(2) gauge-active → CP² fails Gate-2 OR isotropy-locks symbolic / build Yes — centralizer of U(2)⊂SU(3) non-trivial; 11D CP² build (wsmnjvt55) BROKE
W8 Architectural-read discipline recovery uses no α_i, no UV number symbolic/audit Yes — validator NC checks pass
W9 Freeze hashes every carrier object content-addressed machine-lane Yes in principle — re-hash R1.2/R1.4 carrier data + recompute meta-hash

5.2 What reproduces, plainly

5.3 What does NOT yet reproduce (honest gaps)

5.4 The falsifiable bet (confident, testable — not a hedge)

The honest edge is a confident testable bet: the architectural read consults no coupling and no fitted number, so it cannot be quietly tuned. Once the G02 certificate is mounted and re-run it must return exactly {8,3,1}, rank 4, with no extra unbroken factor and no missing one. The falsification map (GUT.md §6.12) is explicit: find a surviving extra U(1) or a missing SU(2), and Gate 2 falls (status → Open / not claimed). The one thing we do not claim is the one nobody can: that this gauge group is the unique output of all possible geometries — that cross-geometry uniqueness is unprovable in principle for every framework (§7).

5.5 How to re-run / re-derive (reproducibility recipe)

  1. Algebra recovery (hand): confirm Isom(SU(3)/T²)=SU(3), Isom(S²)=SU(2), Isom(S¹)=U(1) from textbook homogeneous-space isometry theory; sum dims (8+3+1=12) and ranks (2+1+1=4); verify the multiset is exactly {8,3,1}.
  2. C1 (group theory): verify T² is the unique connected abelian subgroup of maximal rank in SU(3) and that C_SU(3)(T²)=T² (Cartan only); confirm any larger connected H contains a non-abelian factor.
  3. F1/F2 (hand): Isom(T^n)=U(1)^n (abelian, no SU(2)); the Dirac index on a closed odd-dim manifold vanishes; APS boundary index on the fold is one-sided.
  4. CP² kill: compute C_SU(3)(U(2)) and confirm U(2) is gauge-active; the disjunctive fork follows.
  5. Certificate (machine, BLOCKED until mounted): mount certificates/G02_gauge_recovery/, re-run the multiset + no-extra-summand check on R1.2/R1.4 isometry data, re-hash carriers (dcc66f1b2685 / a5b1e6f9d951), confirm the D.4 exotics ledger marks every candidate Absent/Massive.

5.6 Validator state (frozen)

The unified validator returns SCHEMA_CONVERGED with STATUS-UPGRADES:0 and an explicit firewall (AUDIT_DONE → THEORY_CONFIRMED and SCHEMA_CONVERGED → TRUE are forbidden inferences). The descent certificate is CONVERGED_WITH_TERMINAL_WALLS (9 nodes, 8 edges, zero breaks, at least one terminal wall remains → gate status OPEN/wall). Leaf results: L_RECOVERY = DERIVED-GIVEN-E; L_ANCHOR_ALPHA3 = measured-ANCHOR; four NAMED_AXIOM leaves (L_AX_ISOMETRIES, L_AX_CLEAN_CARRIER, L_AX_NONABELIAN, L_AX_HYPERFOLD); two WALL leaves (W_R2_NEUTRALITY, W_R3_COMPLETENESS). Structural posit IDs: AXIOM-CLEAN-CARRIER, AXIOM-HYPER-CARRIER-FOLD, AXIOM-NONABELIAN-NEEDS-NONABELIAN-CARRIER, AX_FORCES_ARE_ISOMETRIES, P2_SHAPE_FROZEN_13D. Floor count = 1 (only the α₃ anchor is counted; E and the frozen shape are necessary glue, not counted). (Source: SG2_report.json.)


6. Open gaps + closure path — the specialist work plan

The most load-bearing section. For each open hole: (a) precise statement, (b) why it's hard / traps to avoid, (c) exactly what closes it (success criterion + what a refuting result looks like), (d) machinery & inputs with exact paths, (e) leverage. The four holes are from the re-grade ledger and hole queue; the two terminal walls (W_R2_NEUTRALITY, W_R3_COMPLETENESS) correspond to Holes 1 and 2. Physics only; no device engineering.

The decisive structural fact for this work plan: SG-2 carries no tunable real, no scale, no coupling — α_i(M_Z) are SG-7 anchors. So the residuals do not terminate on a numeric invariant; they terminate on structural / spectral truths (the observed gauge group E, the LEP/SLD count) or are grammar / claim-boundary items, plus one computation-debt certificate. There is therefore almost nothing to target-fit except the framing, where the κ³/π blade forbids banking the OUTCOME tie. Every genuine-physics path below is a bounded, finite Lie-theory computation on a small object.

6.1 Hole 1 (HIGHEST LEVERAGE) — architecture-neutrality of the abelian-isotropy / CSDR-centralizer uniqueness (C1) — validator wall W_R2_NEUTRALITY

(a) Precise statement. "K₆ is the unique clean SU(3) carrier" is currently a theorem asserted inside the "forces = isometries" + CSDR grammar, not proven as a standalone theorem. The missing object is a CSDR-centralizer uniqueness theorem over the SU(3) subgroup lattice.

(b) Why it's hard / traps to avoid. The hardness is not the computation (it is finite and small) — it is the category-relativity: "clean = abelian isotropy = trivial centralizer-gauge" is a grammar rule, and a bundle/brane reviewer is outside its scope. Full category-neutrality (proving "forces = isometries" is the right category) is not available — that is a modeling choice, not a theorem (R8/wall). Trap to avoid #1: do not over-claim the result as architecture-neutral full stop — the honest ceiling is "DERIVED-CLOSED inside CSDR." The frozen docs explicitly deny "K₆ is the unique clean SU(3) carrier as an architecture-neutral / proven theorem" (re-grade ledger bright-line). Trap #2: prove the CP² exclusion without invoking the anti-fitting rule — the manuscript is explicit that the centralizer argument is the standalone one, and leaning on anti-fitting would be a weaker, target-flavored kill. Trap #3: do not reverse-engineer; the test is whether the enumeration returns T²-uniqueness, not whether you can arrange it.

(c) Exactly what closes it. Prove: Among all homogeneous spaces SU(3)/H with H a closed connected subgroup, the carriers whose CSDR reduction yields exactly 𝔰𝔲(3) as the surviving 4D gauge algebra (no extra factor, no isotropy-locking) are precisely those with H a maximal torus; up to conjugacy H = T², so K₆ is unique. Success criterion: the finite enumeration returns H = T² as the unique pass. Refuting result (also a valid close): if a second H passes the equality clause, C1 weakens to "unique among ..." — a sharper-OPEN; if T² itself fails the equality clause under careful CSDR, the whole carrier claim falls (very unlikely; standard coset reduction). Either way the wall is resolved to a definite verdict.

(d) Machinery & inputs. (1) Enumerate the closed connected subgroups H ⊂ SU(3) up to conjugacy: {1, U(1), T², SU(2), U(2), SU(3)} (~6 conjugacy classes; the lattice is small). (2) For each, compute the CSDR-surviving gauge algebra = centralizer data C_SU(3)(H) and check the equality clause (𝔰𝔲(3) exactly, no extra, no locking). (3) Confirm H = T² is the unique pass. Known anchors already in hand: C_SU(3)(T²) = T² (Cartan only); U(2) is non-abelian/gauge-active. Tools: CSDR / coset dimensional-reduction machinery; the SU(3) subgroup lattice. Start from: closure-attack dossier §4.2 (AXIOM-CLEAN-CARRIER, target-blind); GUT.md §3.5 + Appendix GS.5; SG2_report.json leaf L_AX_CLEAN_CARRIER. Named axiom floor (κ³/π-clean): AXIOM-CLEAN-CARRIER"a homogeneous carrier G/H is 'clean' iff its isotropy H is a maximal torus of G (so its CSDR centralizer adds no non-abelian gauge); for SU(3) the unique clean carrier is SU(3)/T² = K₆."

(e) Leverage. HIGHEST closeable target. This is the gate's principal framework-internal claim; converting it from assertion to DERIVED-CLOSED-inside-CSDR theorem is the single biggest real gain. Cross-gate: K₆ is the same carrier SG-3 uses for the family index (χ = −3), so hardening C1 strengthens the carrier identity SG-3 depends on. Specialist: hand to a CSDR / coset-reduction specialist.

6.2 Hole 2 — SU(3)-carrier completeness over the homogeneous shelf — validator wall W_R3_COMPLETENESS

(a) Precise statement. C1 covers the purely-abelian-isotropy class and kills CP² explicitly, but there is no single delivered enumeration theorem over the whole homogeneous SU(3) coset shelf showing every non-T² carrier over-produces gauge or fails chirality. The clean-class result is in hand; the completeness ledger is not.

(b) Why it's hard / traps. The homogeneous shelf is short and classifiable, so the bounded part is genuinely closeable. The trap is scope creep into the unbounded universal negative: "no manifold of ANY kind (incl. non-homogeneous, unclassified) with SU(3) isometries survives" is a universal negative, unprovable in principle (it is a dissolved unicorn — frame as a shared ceiling, never as an open weakness). Trap to avoid: do not market the homogeneous-shelf result as "unique carrier full stop"; the honest endpoint is "unique carrier among homogeneous SU(3) carriers." A useful dimensional anchor from the classification: dim H ≤ 4 ⟹ dim M ≥ 4; S⁵/Wu manifolds are killed by odd-dimensionality (re-grade ledger SG-1, line 89).

(c) Exactly what closes it. Assemble the GS.5/GS.6 coset shelf + relevant products (notably Witten's CP² × S² × S¹, which passes Gate 2 but dies at chirality, GS.11) into one completeness ledger recording, per carrier: surviving gauge algebra (centralizer), equality verdict, chirality verdict. Certify the homogeneous shelf. Success criterion: every non-T² homogeneous SU(3) carrier has a recorded elimination row (over-production / wrong-group / chirality). Refuting result (valid close): if a non-T² homogeneous carrier survives both the equality clause and chirality, C1's "unique" downgrades — a genuine, publishable finding.

(d) Machinery & inputs. Take the GS.5 coset shelf + GS.6 funnel rows for SU(3) carriers (already in GUT.md §3.5 / Appendix GS — the shelf table is reproduced in §3.8 above); for each record (algebra, equality, chirality); add the product carriers (Witten's CP² × S² × S¹). Tools: homogeneous-spaces classification + CSDR centralizer check. Start from: closure-attack dossier §4.3 (AXIOM-SU3-CARRIER-SHELF, target-blind); hole queue SG-2 item 2; GUT.md §3.5 shelf + GS.11. Named axiom floor: AXIOM-SU3-CARRIER-SHELF"the complete shelf of internal-isometry SU(3) carriers in the declared category is {SU(3)/T², SU(3)/U(2)=CP², and products thereof with abelian/S² factors}; every member except SU(3)/T² over-produces gauge or delivers the family count only as a tunable bundle choice."

(e) Leverage. HIGH. Turns "unique clean carrier" into "unique carrier among homogeneous SU(3) carriers" — a bounded, genuine strengthening. Reachable to DERIVED for the homogeneous shelf; the non-homogeneous case stays the explicitly-flagged residual scope. Specialist: homogeneous-spaces specialist.

6.3 Hole 3 — F1 generalization (no abelian carrier supplies non-abelian SU(2))

(a) Precise statement. Fact F1 is stated as a class fact for tori/torus-orbifolds; it is not packaged as a closed theorem with S²-forcedness as a corollary at full generality.

(b) Why it's hard / traps. It is not hard — it is essentially a one-line no-go. The only trap is the category boundary: F1 holds inside "forces = isometries"; a string framework routes around it via bundle structure groups (the other category). Trap: do not claim F1 as architecture-neutral against bundle constructions — state it as a theorem inside CSDR.

(c) Exactly what closes it. Package the no-go: Isom(M) abelian ⟹ CSDR-surviving gauge algebra abelian ⟹ SU(2)_L requires a non-abelian-isometry carrier ⟹ S² = SU(2)/U(1) is the minimal one. Success criterion: the three implications each verified within CSDR; S²-forcedness drops out as a corollary. Refuting result: only if some abelian carrier sneaks a non-abelian factor via bundle data — which is the other category, outside scope (so within-grammar it cannot refute).

(d) Machinery & inputs. (1) Isom(T^n) = U(1)^n abelian (textbook). (2) CSDR from an abelian-isometry carrier yields only abelian gauge (centralizer of an abelian group adds no non-abelian survivor). (3) S² is the minimal non-abelian-isometry carrier (dim 2, Isom = SU(2)). Start from: closure-attack dossier §4.4; hole queue SG-2 item 3; GUT.md §3.5 shelf (tori rows). Named axiom: AXIOM-NONABELIAN-NEEDS-NONABELIAN-CARRIER"a non-abelian gauge factor can survive only from a carrier with non-abelian isometry; therefore SU(2)_L requires a non-abelian-isometry carrier, minimal S² = SU(2)/U(1)." Hand to the same CSDR specialist as Hole 1.

(e) Leverage. MEDIUM. Reaches AXIOM-CLOSED, promotable to DERIVED with the bounded no-go proof; strengthens C2 (S² forced) from stated-fact to theorem.

6.4 Hole 4 (CHEAPEST / BLOCKED) — G02 multiset / no-extra-summand certificate is AUDIT (input absent on disk)

(a) Precise statement. "No extra unbroken factor survives, no SM factor is missing" — the gate's load-bearing equality claim — is machine-witnessed by the certificates/G02_gauge_recovery/ multiset + no-extra-summand check, but that check is referenced, not independently re-executed; per the validator the input file is absent on disk (BLOCKED). (Confirmed: a corpus-wide search for a G02 certificate folder returned nothing on disk.)

(b) Why it's hard / traps. Not hard — it is mechanical and fail-closed. No new physics. Trap to avoid: do not let "claimed certificate pass" drift into "machine-verified" until the file is actually mounted and re-run — that would be a phantom-citation overclaim (the program has caught this exact failure mode elsewhere, e.g. SG-9's phantom certificate). The status is BLOCKED, not VERIFIED, until the re-run happens.

(c) Exactly what closes it. (1) Mount certificates/G02_gauge_recovery/. (2) Re-run the multiset test on R1.2/R1.4 isometry data: confirm the surviving simple-summand multiset is exactly {8,3,1}, rank 4, with the no-extra-summand predicate passing. (3) Re-hash the carrier objects, confirm dcc66f1b2685 / a5b1e6f9d951 recompute byte-equal. (4) Confirm the D.4 exotics ledger marks every candidate Absent / Massive. Success criterion: VERIFIED — "claimed certificate pass" becomes machine-real. Refuting result (valid close): any extra/missing summand → Gate 2 → Open / not claimed (per the §6.12 falsification map) — a downgrade, which is a legitimate honest outcome.

(d) Machinery & inputs. No new math — execution + verification only. Inputs: R1.2/R1.4 isometry data; the frozen carrier hashes; Appendix D.4 exotics ledger. Start from: closure-attack dossier §4.7; hole queue SG-2 item 4. Status flow: BLOCKED_INPUTS → (mount) → VERIFIED or REFUTED.

(e) Leverage. LOW-MEDIUM on the gate's meaning but highest value-per-effort — converts the gate's central asserted predicate into a machine-checked one with no new physics. Best first task for an owner-artifact session.

6.5 The disclosed residuals (correct as stated — keep stated, never promote)

Three further residuals are not holes to close but honest claim-boundaries to keep correctly stated:

6.6 Attack-plan roll-up

Hole / residual Technique Named axiom (κ³/π-clean) Specialist target / artifact Realistic endpoint
Hole 1 — C1 neutrality (W_R2) axiom-floor + selector AXIOM-CLEAN-CARRIER CSDR-centralizer uniqueness over SU(3) subgroup lattice DERIVED-CLOSED-inside-CSDR (bounded)
Hole 2 — shelf completeness (W_R3) eliminative enumeration AXIOM-SU3-CARRIER-SHELF completeness ledger over homogeneous shelf DERIVED for homogeneous shelf
Hole 3 — F1 generalization axiom-floor + no-go AXIOM-NONABELIAN-NEEDS-NONABELIAN-CARRIER abelian-isometry ⟹ abelian-gauge no-go AXIOM-CLOSED → DERIVED (bounded)
Hole 4 — G02 certificate machine-lane (none) mount + re-run multiset / no-extra-summand BLOCKED → VERIFIED (best value/effort)
R1 OUTCOME-TIE firewall (framing) AXIOM-GAUGE-OUTCOME-TIE wording audit DISCLOSED (honesty spine)
R6 given-E guarded restatement AXIOM-GIVEN-E-RECOVERY keep conditional stated DISCLOSED-CONSISTENT
R8 grammar axiom-floor AXIOM-FORCES-ARE-ISOMETRIES name in axiom ledger AXIOM-CLOSED

REDUCE-vs-RELOCATE verdict. The plan does not turn one hard problem into three harder ones, and — critically — it does not try to "close" the things that cannot be closed (R1 tie; R6 given-E). Each genuine-physics path is a bounded finite Lie-theory computation on a small object (Hole 1: the six-class SU(3) subgroup lattice; Hole 2: the short classified coset shelf; Hole 3: a one-line no-go corollary). None introduces a tunable real; there is nothing to target-fit in SG-2 except the framing. The honest expected outcome of a full campaign: Hole 1 → DERIVED-CLOSED-inside-CSDR (the biggest real gain); Hole 2 → DERIVED for the homogeneous shelf; Hole 3 → AXIOM-CLOSED→DERIVED; Hole 4 → VERIFIED; R1/R6 stay DISCLOSED; R8 AXIOM-CLOSED. No promotion of the gate's label is implied — SG-2 stays DERIVED-GIVEN-E — but its carrier-forcedness moves from category-relative assertions to category-relative theorems, and the OUTCOME-tie stays correctly disclaimed. That is the only kind of "more closure" SG-2 admits.


7. Honest ceiling & scope

Overall disposition: DERIVED-GIVEN-E — and it stays there. The genuine content (algebra equality {8,3,1}, rank 4, no extra/missing; carrier-forcedness C1/C2/C3, including the CP² end-to-end break) is rigid within the declared grammar. The gauge-group OUTCOME is a rival TIE; cross-geometry uniqueness is not claimed. Hardening the open holes moves the carrier-forcedness from category-relative assertions to category-relative theorems — it does not promote the gate's label. Ceiling, as everywhere: serious candidate, NOT validated.

Dissolved ≠ solved. Three "unicorns" are dissolved as shared ceilings — limits on all knowledge, not gaps in ours — and must never be printed as open weaknesses or claimed as proven:

Selection ≠ derivation; given-E ≠ derivation of E. SHAPE is selected-not-forced absolutely — forced only inside the declared grammar + MDL metric, at a cost of ~4 anchors + ~9–10 injected reals (~13–14 total). SG-2 consults none of the numeric anchors {ℏ, M_Pl, α_i(M_Z), y_t, |V_us|}: it reads no coupling and no UV number. It owes no new measured invariant — the only objects it owes are theorems and one certificate re-run, not measurements. Its dependence is on existing structural/spectral anchors: the observed SM gauge group as given-E content, and the LEP/SLD light-species count 2.984 ± 0.008 for the no-mirror requirement.

Explicitly NOT claimed by this dossier: (1) that recovering SU(3) × SU(2) × U(1) discriminates this framework from rivals — it is a tie; (2) that this gauge group is forced across all geometries — given-E only; (3) that F1/F2/C1 are architecture-neutral theorems — they are theorems inside the category (two terminal walls); (4) any coupling-unification numeric — that is SG-7; (5) the representation/charge/ℤ₆ structure — that is Gates 3–5.

The anchors paid. Two measured invariants (the observed spectrum E including the gauge group as given-E content; the LEP/SLD count) and two grammar-anchors ("forces = isometries" + the CSDR centralizer rule). The gauge-group OUTCOME is measured-but-shared → DISCLOSED, carrying no framework-discriminating anchor.

Closing honest statement. SG-2 is DERIVED-GIVEN-E. Its genuine, defensible content is real and rigid: the surviving 4D isometry algebra equals 𝔰𝔲(3)_c ⊕ 𝔰𝔲(2)_L ⊕ 𝔲(1)_Y (equality, not containment; multiset {8,3,1}, rank 4, no extra/missing), and — the framework-internal part — the carriers are forced within the declared grammar: K₆ = SU(3)/T² is the unique clean SU(3) carrier by abelian-isotropy uniqueness (C_SU(3)(T²)=T², so the CSDR centralizer adds no gauge; the cheaper CP² = SU(3)/U(2) over-produces gauge via its non-abelian U(2) isotropy and is killed — the 11D CP² build wsmnjvt55 BREAKS), S² is forced for weak by F1, and the folded S¹_Y/ℤ₂ is forced for hypercharge by F2. Its honest open surface is equally clear: the gauge-group OUTCOME is a rival TIE — a filter string/M/F/NCG/lattice all pass, not a framework discrimination; the recovery is given-E; the carrier-forcedness results are theorems inside the "forces = isometries" category (architecture-neutrality asserted, not proven — two terminal walls); SU(3)-carrier completeness over all carriers is established for the clean class but audit-grade as a full enumeration; and the G02 multiset certificate is BLOCKED (input absent on disk). The attack plan reduces these to named, target-blind axioms and bounded finite Lie-theory computations, with the κ³/π falsification test guarding the framing (never bank the OUTCOME tie as a framework win). The realistic ceiling is a uniqueness theorem inside CSDR for the clean carrier (Hole 1 → DERIVED-CLOSED-inside-CSDR) plus a verified certificate (Hole 4) over a named grammar axiom (R8) — a real strengthening of the carrier-forcedness from assertion to theorem, not a promotion of the gate's label and not a closure of the OUTCOME tie. STATUS-UPGRADES:0; frozen branch dcc66f1b2685 / a5b1e6f9d951 READ-ONLY; given-E ≠ derivation of E; the gauge-group OUTCOME is and remains a tie — the discrimination is the carrier-forcedness; nothing applied, nothing deployed.

Dossier built 2026-06-29. Frozen 13D K₆ branch only. Synthesized + expanded from the corpus closure-attack dossier, the validator report, the 2026-06-29 re-grade ledger and hole queue, and GUT.md — common material referenced to source, not invented.