What this is. The full, working-physicist treatment of gate SG-8 — Flavor closure (the $F^+$ chamber) on the frozen 13D K₆ branch. It expands the 30-second live popup and the brief closure-result article into the deep version: the rigorous mechanism, the insights that produced it, the honest reproducibility ledger, and — most load-bearing — a concrete, specialist-grade work plan for every open hole. Audience: the specialist who will close the gaps, and the public under the openness policy.
Binding discipline (carried verbatim from the corpus). STATUS-UPGRADES:0. The honest gate status is OPEN and this dossier never upgrades it. Frozen branch
dcc66f1b2685/ manifest metaa5b1e6f9d951is READ-ONLY. The flavor over-determination is ~4× (3.7–4.4×) — NOT the retired ~5.5× headline and NOT the over-corrected ~1.6×. $N_d/N_e/N_\nu$ are sector-scale calibration inputs, NOT derived from $N_u$ (there is no $N_d=f(N_u)$ relation anywhere in the corpus). $M_R$ is UNKNOWN/open. given-E ≠ derivation of E; AXIOM-CLOSED ≠ proven; selection ≠ derivation; dissolved ≠ solved. Every number below is traceable to a corpus source actually read; uncomputed quantities are marked OPEN.
Given the particles we observe, this geometry predicts the shape of the flavor puzzle — every within-sector fermion mass ratio as a forced power of one constant, and all CKM and PMNS mixing angles plus the CP phase from a single chamber angle — by diagonalizing frozen operators, not by inserting numbers. The mechanism's signature is that per-family tuning is structurally forbidden: the CKM matrix is the misalignment $V_{\rm CKM}=U_u^\dagger U_d$ of two frozen diagonalizations, so you cannot fit one entry without breaking the rest. That is what makes it a mechanism rather than a parameterization.
| Field | Value |
|---|---|
| Gate | SG-8 — Flavor closure (Yukawas / CKM / PMNS); GUT manuscript Gate 9 (§6.9) |
| Live chip (binding) | OPEN — ratio / mixing / phase leg DERIVED-GIVEN-E (direction: held) |
| Frozen branch | dcc66f1b2685 / manifest meta a5b1e6f9d951 (READ-ONLY) |
| Over-determination | ~4× (3.7–4.4×) — ~22 outputs from ~5–6 effective inputs |
| Anchors paid | $y_t(M_Z)$ → $N_u$ · $\lvert V_{us}\rvert$ → $\theta_F$ (two declared), plus three fitted sector scales $N_d, N_e, N_\nu$ and the uncomputed $M_R$ |
The gate is OPEN by the least-closed-residual rule: even though the ratio/mixing/phase leg is DERIVED-GIVEN-E, the gate carries live residuals (R2, R6 OPEN; R3, R7 BLOCKED) and three reduce-to-axiom posits that the 2026-06-25 atomicity sweep reopened to AXIOM-OPEN (R1, R8, R9). A gate is no stronger than its weakest residual; here the weakest are genuinely open.
Source for status: …/PER_GATE_DOSSIERS/SG8_COMPLETION_HANDOFF/02_CURRENT_STATE.md; …/07_CLOSURE_RESULT.md; the brief article articles/SG8_FLAVOR_CLOSURE_RESULT.md.
It establishes (given the observed spectrum E, the selected $F^+$ chamber, and the anchors): within-sector mass ratios are forced powers of $\kappa=e^{-\pi\sqrt3}\approx4.3286\times10^{-3}$ with the sector scale cancelling in the ratio; the CKM CP phase $\delta_{\rm CKM}=-2\pi/3$ is read off an order-three holonomy; and all CKM and PMNS magnitudes plus the Jarlskog $J$ fall out of a single chamber angle $\theta_F$ — by diagonalizing the frozen sector operators, with family-level tuning structurally banned. This is the genuine, defensible content.
It does not establish a zero-input derivation of flavor. It is explicitly not "CKM solved," not a complete flavor theory, not a derivation of the absolute mass scales. Three sector scales are fitted, the seesaw scale $M_R$ is uncomputed, the up-quark mass initially appeared to carry a forced ~4.4σ tension against PDG (a 4D-shadow comparison since resolved to +0.058σ once the full 13D Weyl-shadow transport supplies the symmetry factor $1/\sqrt6=1/\sqrt{|S_3|}$), and the numerical comparison harness is unverified-by-re-execution (AUDIT/BLOCKED). given-E is not a derivation of E. The honest ceiling everywhere in this program is serious candidate, NOT validated.
Flavor is the Standard Model's deepest unsolved input. The SM has 19 free parameters; the majority of them live in the flavor sector: nine charged-fermion masses (or equivalently nine Yukawa couplings spanning roughly six orders of magnitude, from the electron at ~0.5 MeV to the top at ~173 GeV), three CKM mixing angles, one CKM CP-violating phase, and — once neutrino mass is included — three more mixing angles, at least one Dirac CP phase, and the neutrino mass-squared splittings. Every one of these is measured and typed in by hand. No accepted theory derives the pattern: the masses, the mixings, the hierarchy, and the CP phase are simply free inputs.
The structure that cries out for explanation is striking and specific:
The textbook landscape of flavor model-building, and why each line falls short of a derivation:
The common thread: every accepted approach reorganizes the flavor freedom (into charges, into a fitted modulus, into textures, into priors) but none eliminates it down to a small structural input set with the per-entry freedom genuinely removed. That is the gap. The contribution of SG-8 is to make the within-sector ratios and all mixings/phases forced functions of a small frozen structure — with the honest concession that the absolute scales remain calibrated.
Full derivation, worked two-anchor fixing, per-observable certificate tables, and freeze records live in the published manuscript (GUT.html §6.9, §7, §8, Appendices I/J/K, R0/R1). This section is the attack-grade reconstruction, citing the corpus for every load-bearing number.
The flavor sector is generated by the finite flavor chamber $F^+_{\rm finite}$, a non-metric $\oplus$-layer object. Crucially it contributes 0 of the $D=13$ propagating dimensions and carries no Kaluza–Klein tower (CR9.3 / I.8a). It augments the SM-routing backbone $K_{\rm gauge}=K_6\times S^2\times S^1_Y/\mathbb{Z}_2$ with exactly the structure needed to turn three identical-charge families into structured Yukawa operators. It is not an extra dimension; it is an algebraic decoration that organizes the generation index.
The pipeline is a single forward chain (manuscript §7.3):
(y_t , |V_us|) --calibrate--> F+ --frozen ops--> O_{u,d,e,ν}
--Yukawa map--> Y_{u,d,e,ν} --diagonalize & RG--> {m_q, V_CKM, J, m_ℓ, U_PMNS}
The load-bearing factors, each named so a reader can attack it independently.
$$ \mathcal{G}_{\rm gen}=\mathrm{span}\{g_1,g_2,g_3\}, \qquad \dim \mathcal{G}_{\rm gen}=3. $$
The dimension 3 is inherited, not assumed: it is the spin-ℂ Borel–Weil–Bott family index $\chi(K_6,E)=-3$ established at SG-3 (Gate 4). SG-8 introduces no independent per-family multiplicity. (Attack handle: the family count is given-E; SG-8 inherits SG-3's conditionality. Per the geometry-role split this leg is generic-survives-without — the integer 3 is observed independently of the geometry.)
The chamber modulus sits at the order-three modular fixed point
$$ \tau=\omega=e^{2\pi i/3}, \qquad \text{(frozen hash 03b30a9c931a)}, $$
inherited frozen from SG-6 (stabilization; the Weyl-rigid chamber). Two consequences flow from this single algebraic fact:
(Attack handle: $\tau=\omega$ is read from a minimum in SG-6, not independently derived — it shares SG-6's soft spot; this is residual R8.)
$$ \Pi_u,\ \Pi_d,\ \Pi_e,\ \Pi_\nu, \qquad \Pi_i\Pi_j=\delta_{ij}\Pi_i, \qquad \text{(frozen hash 3b8d68559f5e)}. $$
These orthogonal projectors are determined by group theory; they decompose the chiral mode space into the four sectors. (Per the geometry-role split this leg is essential — geometry-indexed, no geometry-free analog.)
Each sector carries an integer/rational action ladder selected lex-minimally, target-blind on its declared Dynkin/ladder family:
$$ a_u=(2,1,0)\ \ (\text{hash e2ef21cecade}), \qquad a_d=(\tfrac43,\tfrac23,0)\ \ (\text{hash 989edc50b559}). $$
The up ladder is lex-min on the $A_2$ root system; the down ladder on the affine $\tilde A_2$ root system. The charged-lepton ladder is structural (derived from $a_d$ plus the leptonic charge triplet $(-1,0,+1)$ under $\mathbb{Z}_3$):
$$ a_e=(2,\tfrac43,0). $$
The neutrino operator $O_\nu$ additionally carries the second-cycle Berry phase $+2\pi/3$ on the $A_2$ root system. (Attack handle: lex-min selection over a declared family is selection-inside-a-category, not forcedness across all rational ladders — residual R9.)
Every within-sector mass step is a power of one constant:
$$ \boxed{\ \kappa=e^{-\pi\sqrt3}\approx4.3286\times10^{-3}.\ } $$
The chamber operators at $\tau=\omega$ have diagonal entries
$$ (O_s)^{aa}=N_s\,\kappa^{\,a_s^{(a)}}, \qquad s\in\{u,d,e,\nu\}, $$
so the within-sector ratios are forced:
$$ \frac{m_t}{m_c}=\frac{m_c}{m_u}=\kappa^{-1}=e^{\pi\sqrt3}\approx 231, \qquad \frac{m_b}{m_s}=\frac{m_s}{m_d}=\kappa^{-2/3}=e^{2\pi\sqrt3/3}\approx 38.5. $$
This is the genuine prediction: the hierarchies are forced, not fit. The decisive point is that the sector scale $N_s$ cancels in every ratio — it cannot be smuggled in to fix a ratio, because it appears identically in numerator and denominator. (This leg is essential, and it reproduces by hand from $\kappa$ alone.)
Source for $\kappa$, ladders, and ratios: 01_DOSSIER.md §1.1–§1.2 (W1, W2); 07_CLOSURE_RESULT.md §1.
$$ (Y_s)^{ab}=N_s\,\langle g_a\mid O_s\mid g_b\rangle, \qquad s\in\{u,d,e,\nu\}, \qquad \text{(map hash 1f20935643cf)}. $$
The normalization $N_s$ is sector-level only. Family-level normalizations $N_{s,a}$ are explicitly forbidden by rule I.4 — this is the operational anti-fitting firewall. Phases are read from holonomy and are not retunable. (This leg is essential.)
Masses and mixings come from diagonalizing the generated Yukawas:
$$ U_s^\dagger\,Y_sY_s^\dagger\,U_s=D_s^2. $$
The CKM matrix is the misalignment of two frozen diagonalizations:
$$ V_{\rm CKM}=U_u^\dagger U_d, \qquad U_u=\mathbb{1}_3 \ \text{at}\ \tau=\omega, \qquad U_d=\text{DFT-on-}\mathbb{Z}_3 \text{ rotated by the single chamber angle } \theta_F. $$
Likewise $U_{\rm PMNS}=U_e^\dagger U_\nu$. The CKM is not an inserted unitary — it is what you get from rotating two frozen operators into their eigenbases and comparing. This is the structural heart of the anti-fitting claim: there is no entry you can dial. (Frozen chamber-angle hash 1ff57f48d45a; this leg is essential.)
The order-three ($\mathbb{Z}_3$) holonomy forces the raw CKM phase to the third root of unity:
$$ \delta_{\rm CKM}=-\frac{2\pi}{3}=-120° \quad(\text{raw}). $$
In the Wolfenstein-aligned convention this maps to $+60.0°$. Both numbers must be stated honestly (W6):
The ~3.7σ raw-PDG figure is the honest hostile-reviewer number; the 0.79σ is band-relative. (Source: 07_CLOSURE_RESULT.md §2 caveat 2; 01_DOSSIER.md W6.)
All CKM magnitudes beyond the $\lvert V_{us}\rvert$ anchor, all PMNS angles, both CP phases, and the Jarlskog $J$ follow from the single calibrated chamber angle $\theta_F$. Two certified pulls are notably tight:
$$ \lvert V_{cb}\rvert: \ 0.005\sigma, \qquad J_{\rm CKM}: \ 0.21\sigma. $$
(Source: 01_DOSSIER.md §1.2 genuine-output table; 07_CLOSURE_RESULT.md §1.)
The neutrino masses arise from a Type-I seesaw:
$$ M_\nu^{\rm eff}=-M_D\,M_R^{-1}\,M_D^T, \qquad M_D=N_\nu\,\langle g_a\mid O_\nu\mid g_b\rangle. $$
The ratio $\Delta m^2_{21}/\lvert\Delta m^2_{32}\rvert$ is a genuine frozen output of $O_\nu$. The absolute splitting is set by $N_\nu^2/M_R$ — and here the construction stops: $M_R$ is asserted "fixed by the chamber's Cartan-torus modulus and spin-ℂ flux $N=1$" (manuscript K.4, L9580/L9683) but never computed — no value, no closed form, no hash. This is residual R3, and it is the single cleanest open object in the gate.
Stated precisely (and this is the authoritative figure that retires both the ~5.5× overclaim and the ~1.6× over-correction):
Why the two wrong numbers are wrong:
Source: GATE9_FLAVOR_INPUT_LEDGER_FINDING_2026-06-24.md (the authoritative FITTED_RELABEL adjudication); 01_DOSSIER.md W9 / R4.
| Genuine output (frozen-ladder / frozen-holonomy prediction) | Mechanism |
|---|---|
| Within-sector mass ratios, all four sectors | $\kappa^{a_s^{(a)}}$ powers; $N_s$ cancels; no per-family knob |
| All CKM magnitudes except the $\lvert V_{us}\rvert$ anchor | $U_u^\dagger U_d$ misalignment, one angle $\theta_F$ |
| $\delta_{\rm CKM}$, Jarlskog $J_{\rm CKM}$ | order-three holonomy, read not fit |
| PMNS angles + leptonic CP phase $\delta_{CP}^\ell$ | $U_e^\dagger U_\nu$; second-cycle Berry phase |
| $\Delta m^2_{21}/\lvert\Delta m^2_{32}\rvert$ ratio | frozen $O_\nu$ |
| Quantity | Honest reality (calibration in disguise, per I.0a.2) |
|---|---|
| $m_b$ absolute | anchor-consistency check — $N_d$ is pinned to it |
| $m_\tau$ absolute | anchor-consistency check — $N_e$ is pinned to it |
| $\Delta m^2$ absolute scale | anchor-consistency check — set by $N_\nu^2/M_R$; $M_R$ UNKNOWN |
| $m_t$ absolute | anchor-consistency check — $m_t=y_t v/\sqrt2$, the $y_t$ anchor as a mass |
So the precise statement of closure: the chamber genuinely predicts within-sector hierarchies + all mixing/phase data from two anchors; the absolute mass scale of each sector is a tuned normalization, one scale per sector — legitimate standard physics, but NOT "derived from $N_u$" and NOT "one knob per observable."
These are the moves that produced the progress, now shared at working depth so a reader can both check and reuse them.
The central insight is that the empirical near-equality $m_t/m_c\approx m_c/m_u$ (and its down-sector analog) is not a coincidence to be fit but the signature of a single geometric step. If consecutive-generation ratios are equal, the spectrum is a geometric progression $1:\kappa:\kappa^2$ — one constant, not three masses. The construction makes this exact: the operator eigenvalues are literally $N_s\kappa^{a_s^{(a)}}$ with integer/rational ladder exponents. The reason this is a prediction and not a fit is the cancellation in §3.6: because $N_s$ divides out of every ratio, no amount of scale-tuning can move a ratio. The ratio is whatever $\kappa$ and the ladder say it is.
$\kappa$ is the single constant from which every within-sector step is built; it is read from the chamber geometry, not adjusted to match a mass. The falsification test discipline (the κ³/π rule — a proposed constant counts only if it would be written without knowing the target) is applied here: $\kappa$ enters as $e^{-\pi\sqrt3}$, a geometric quantity, and reproduces ~231 and ~38.5 with no further input. That it lands on the observed ratios given it was written geometrically is the content of the win.
Most flavor models carry the CP phase as a continuous free parameter. Here the order-three modular fixed point forces the phase to a discrete value, $-2\pi/3$. A discrete prediction is the strongest possible kind: there is no nearby value to slide to. The honest edge is that the discrete prediction ($+60.0°$ aligned) sits ~3.7σ from the raw PDG central value — which is exactly why both the band-relative and raw numbers are published.
The deepest structural insight is that flavor mixing is a derived misalignment, not an inserted matrix. By computing $V_{\rm CKM}=U_u^\dagger U_d$ from the eigenbases of two frozen operators, and by banning family-level normalizations (I.4), the construction removes the per-entry freedom that every fitting approach relies on. This is the difference between a mechanism and a parameterization: in a parameterization you can always absorb a discrepancy into a coefficient; here you cannot, because the coefficients do not exist as free objects.
A program-level insight surfaced repeatedly in the SG-8 audits: the over-determination figure is itself a place where dishonesty hides in both directions. Overclaiming (5.5×) pretends the sector scales are not inputs; over-correcting (1.6×) pretends the forced ratios are independently injected. The discipline of counting effective inputs honestly — landing at ~4× — is what makes the genuine win believable. The same band-honesty rule (report raw pulls; never let an oversized theory band manufacture a sub-1σ "pass") is a cross-gate rule shared with SG-7 and BG-10 (CONCERN 2(f) / FL-03).
The geometry-role split (from the verified gates-without-geometry analysis) is itself an insight worth sharing: it tells you exactly which legs are the gate's claimed achievement. Essential legs (τ=ω diagonality, the projectors, the κ-ladders, the Yukawa map, diagonalization-not-insertion, and the forced $m_u$) are geometry-indexed and have no geometry-free analog. Generic legs (the per-sector calibration scales, the ~4× count, the diagnostic labels, the reproducibility harness, the family count) survive without the geometry — they are inputs or scorekeeping, not the achievement. Holding this split prevents the gate from being credited for textbook physics or debited for the geometry's own bookkeeping.
| # | Witness | Asserts | Grade | Reproduces? |
|---|---|---|---|---|
| W1 | Frozen ladders $a_u=(2,1,0)$, $a_d=(4/3,2/3,0)$, $a_e=(2,4/3,0)$ | within-sector ratios are powers of $\kappa$, no per-family knob | hand-checkable | Yes — $e^{\pi\sqrt3}\!\approx\!231$ (up), $e^{2\pi\sqrt3/3}\!\approx\!38.5$ (down) recompute by hand |
| W2 | Chamber operators $O_{u,d,e,\nu}$ diagonal at $\tau=\omega$ | $(O_s)^{aa}=N_s\kappa^{a_s^{(a)}}$ | symbolic | Yes for diagonal entries given $N_s$ (frozen R1.6 values) |
| W3 | Quark pull table J.6 | certified rows sub-band | machine-lane | AUDIT — CSV 661bbe085fc5 referenced, not independently re-run |
| W4 | Lepton/ν pull table K.3/K.5 | charged-lepton ratios + PMNS + $\delta_{CP}^\ell$ in band | machine-lane | AUDIT — CSV 6959d274dfe2; not re-run here |
| W5 | Two-anchor economy | strict inputs < outputs | hand-checkable | Yes (strict inequality); headline corrected to ~4× (see W9) |
| W6 | $\mathbb{Z}_3$ holonomy → $\delta_{\rm CKM}=-2\pi/3$ | CP phase forced | hand-checkable | Yes — $-120°$ raw; $+60.0°$ aligned vs PDG $65.5°$ (0.79σ band / 3.7σ raw) |
| W7 | Flavor Lock Table (I.0a.1) | no output selected chamber structure | symbolic/audit | Conditional — auditable claim, not a theorem |
| W8 | Freeze hashes + meta-hash a5b1e6f9d951 |
every chamber object content-addressed | machine-lane | Yes in principle; not re-run here |
| W9 | I.0.2 honest-margin caveat | compression ≈ 4× (3.7–4.4×) | hand-checkable | Yes — 22 / ~5.5 ≈ 4×; authoritative figure |
The J.6 "Full Numerical Certificate" reports several rows as sub-1σ "certificate-complete," but the raw PDG pulls are materially larger — the sub-1σ figures come from oversized theory bands, not agreement:
| Observable | Raw PDG pull | Band-relative |
|---|---|---|
| $m_u$ | ~4.4σ (raw 4D-shadow comparison; resolved to +0.058σ under full 13D Weyl-shadow transport, $1/\sqrt6=1/\sqrt{|S_3|}$) | ~1.26σ (vs propagated theory band $\sigma_{\rm th}=1.5$) |
| $\lvert V_{td}\rvert$ | ~13.7σ | sub-1σ (band) |
| $\delta_{\rm CKM}$ | ~3.7σ | ~0.79σ (band) |
The honest read is the raw pulls; the J.6 "certificate-complete" wording is a reporting/false-closure defect, not a number change. The gate stays OPEN; W3/J.6 is annotated as a band-inflation false-closure leg. Source: 01_DOSSIER.md "Cross-gate propagation note (2026-06-29)" / CONCERN 2(f) (FL-03).
reproduce_all.py regenerating the byte-equal CSVs is the load-bearing executable check, and it has not been independently re-run. Until the CSVs are mounted and re-run target-blind, "every J.6/K.5 value regenerates from the frozen hashes" is declared-not-independently-verified.N_d = 0.024 (defined to set m_b to its target value at M_Z) (GUT.md L4467) and N_e = 0.0102 (chosen such that m_τ matches its target value) (L4468). These are now correctly relabeled "sector-scale calibration inputs."A fail-closed forcing packet (SG8_R7_R3_R6_forcing_packet) was built and independently verified to fail-closed: its three verifiers correctly report BLOCKED_MISSING_ARTIFACTS (R7), BLOCKED_MISSING_INPUTS (R3), and BLOCKED_MISSING_TOP_INPUT_OR_TRANSPORT (R6) — with no false PASS. Dropping the real files into certificates/G09_flavor/ forces each row to VERIFIED / REFUTED / sharper-OPEN with no narrative wiggle room. Source: 07_CLOSURE_RESULT.md §5.
appendix_I_quark_outputs.csv (661bbe085fc5) and appendix_J_lepton_neutrino_outputs.csv (6959d274dfe2), run reproduce_all.py against the four R1.8 anchors + R1.6 frozen chamber, confirm byte-equality with J.6/K.5, re-hash R1.6 rows, confirm the meta-hash recomputes to a5b1e6f9d951.Branch dcc66f1b2685 · manifest meta a5b1e6f9d951 · $\tau=\omega$ 03b30a9c931a · $\Pi_s$ 3b8d68559f5e · $a_u$ e2ef21cecade · $a_d$ 989edc50b559 · $O_u$ 07be17dd8a1c · $O_d$ 50ef768bb146 · $O_e$ 08ff25117d00 · $O_\nu$ 495ddbdcedb9 · Yukawa map 1f20935643cf · $\theta_F$ 1ff57f48d45a · $N_{u,d,e}$ 20dc4e0b8220 · $\eta_{BK}$ 84e94518d3f5 · $K_{tb}^{\rm crit}$ c15d00c6f664 · RG transport f531205a9159 · $M_Z$ a6852c7a6b00 · uncertainty rule 61b0d93507e7 · anchor $y_t$ 548d7099ef18 · anchor $\lvert V_{us}\rvert$ a1bc510bc7cd · FCNC no-go fff4b433b7b3 (op-class 551488d06011) · quark CSV 661bbe085fc5 · lepton/ν CSV 6959d274dfe2. $M_R$: no hash — UNKNOWN/open.
This is the most load-bearing section. Each open hole is a self-contained work-package: (a) precise statement, (b) why it is hard + traps to avoid, (c) exactly what closes it (with the refuting outcome named), (d) machinery & inputs, (e) leverage. The cardinal discipline throughout: the κ³/π falsification test — a proposed axiom or constant counts only if it would be written without knowing the target. A closure that introduces a new tuning has relocated the input, not removed it.
The residual register and current dispositions (from 02_CURRENT_STATE.md / 07_CLOSURE_RESULT.md):
| ID | Residual | Current disposition |
|---|---|---|
| R1 | $F^+$ chamber: derivation or compressed fit? | strong version OPEN (no witness); weak lock AXIOM-OPEN |
| R2 | $N_d, N_e, N_\nu$ fitted sector scales | OPEN (no $N_d=f(N_u)$) |
| R3 | seesaw $M_R$ | BLOCKED/OPEN (uncomputed) |
| R6 | $m_u$: raw ~4.4σ, resolved to +0.058σ | RESOLVED (13D Weyl-shadow transport → $1/\sqrt6=1/\sqrt{|S_3|}$; sharp prediction that passes) |
| R7 | J.6/K.5 numerical harness | BLOCKED/AUDIT |
| R8 | $\tau=\omega$ read from a minimum | AXIOM-OPEN |
| R9 | lex-min ladders category-relative | AXIOM-OPEN |
| R10 | PMNS octant / $\delta_{CP}^\ell$ | EXPORTED/Diagnostic (experiment-gated) |
Attack order by leverage: R3 (cheapest genuine count-shrink) → R7 (cheapest machine-reality win) → R2 (largest count-shrink) → R8, R9 (atomicity proofs) → R6 (falsifier confirmation) → R10 (experiment-gated).
(a) Precise statement. Three sector normalizations are full free inputs, each pinned to one absolute mass: $N_d$ must yield $m_b(M_Z)=2.89$ GeV (i.e. $\lvert y_t/y_b\rvert\approx58$); $N_e$ must yield $m_\tau(M_Z)=1746$ MeV; $N_\nu^2/M_R$ must yield $\Delta m^2_{21}=7.42\times10^{-5}\,\mathrm{eV}^2$. There is no $N_d=f(N_u)$ relation anywhere in the corpus. Closing R2 means deriving at least one scale from geometry + already-counted anchors, with no new tuning.
(b) Why it's hard / traps. The "derived from $N_u$ after freeze" label (GUT.md L9147/L9093/L13849) is a relabel, not a derivation — the L13675 promise "it arrives from the geometry" is never discharged. The defining lines are self-incriminating (N_d = 0.024 (defined to set m_b to its target); N_e = 0.0102 (chosen such that m_τ matches its target)). Trap to avoid: renaming the scales "geometric" without a target-blind computation is RELABEL_FAIL — the corpus correctly refused AXIOM-SECTOR-SCALE-INFLOW precisely because Route B was not computed. Do not bank an axiom for an unrun computation.
(c) Exactly what closes it.
- Route A — between-sector determinant. The manuscript already routes $\lvert y_t/y_b\rvert$ through the frozen Wilson-line finite determinant via $\eta_{BK}$ (84e94518d3f5) and $K_{tb}^{\rm crit}=e^{-\pi\sqrt3/16}$ (c15d00c6f664), landing at $\lvert y_t/y_b\rvert(M_Z)=57.50\approx58$. Success criterion: prove $N_d/N_u=g(\eta_{BK}, K_{tb}^{\rm crit})$ target-blind — i.e. that $\eta_{BK}$ and $K_{tb}^{\rm crit}$ were frozen from Gate-8 geometry before and independently of the $m_b$ value, and that $g$ returns $0.024$ (= $N_d$ at $N_u=1$) with $m_b$ not in its inputs. κ³/π status: SUSPECT — $K_{tb}^{\rm crit}=e^{-\pi\sqrt3/16}$ is suspiciously the right magnitude; if it was reverse-engineered from $\lvert y_t/y_b\rvert\approx58$, Route A RELOCATES (the κ³/π-manifest true-by-construction cautionary pattern). Refuting outcome: the provenance audit shows reverse-engineering → Route A dead, R2 stays three calibration inputs.
- Route B — chiral inflow. Conjecture (κ³/π-clean): AXIOM-SECTOR-SCALE-INFLOW — "the sector normalization $N_s$ equals the ratio of the chiral-index density of $\Pi_s E$ to that of $\Pi_u E$." Writable with no flavor number in sight (it references only bundle indices). Success criterion: the computed index ratios yield $0.024$ (down), $0.0102$ (lepton) target-blind. Refuting outcome: index ratios computed target-blind and they miss → axiom REFUTED, R2 confirmed as fits (a clean, valid negative).
(d) Machinery & inputs. Route A: recompute $\eta_{BK}$ and $K_{tb}^{\rm crit}$ from the Gate-8 Wilson-line finite determinant from scratch, target-blind (hand to the Gate-8/threshold specialist). Route B: compute the spin-ℂ chiral-index density on each $\Pi_s E$ sub-bundle and form the ratios; compare to 0.024 / 0.0102. Files: 01_DOSSIER.md §4.2; frozen hashes 84e94518d3f5, c15d00c6f664; manuscript App I/J.
(e) Leverage. HIGH. Each scale removed shrinks the input count by 1 (improving the ~4×). Deriving even one is the largest count-shrink available in the gate.
(a) Precise statement. $M_R$ in $M_\nu^{\rm eff}=-M_D M_R^{-1}M_D^T$ is asserted "fixed by the chamber's Cartan-torus modulus and spin-ℂ flux $N=1$" (K.4) but never computed: no value, no closed form, no hash. The absolute neutrino scale leg is BLOCKED until $M_R$ exists.
(b) Why it's hard / traps. $M_R$ has no laboratory witness — the only thing that would pin it directly is $\eta_B$ (the baryon asymmetry), and using $\eta_B$ would be target-fitting (excluded by the κ³/π falsification test). Trap to avoid: a candidate $M_R=\kappa M_U$ is ~231× off the corpus $\kappa^0$ — flagged as a candidate-in-tension. Do not bank any $\kappa$-power identity before running the inversion. The corpus correctly refused AXIOM-MR-IS-MU because the inversion/coincidence test has not been run.
(c) Exactly what closes it — the decisive inversion (do this first). Given $N_\nu$ pinned by the $\Delta m^2$ ratio (a genuine output) and the measured absolute splitting, invert for the required scale:
$$ M_R^{\rm req}=\frac{N_\nu^2}{(\text{measured abs }\Delta m^2)}. $$
Then test coincidence: does $M_R^{\rm req}$ land — parameter-free, within an order of magnitude — on a scale the geometry already committed: $M_U$ (the SG-7 threshold-unification scale, $\sim10^{16}$ GeV), $R_0^{-1}$ ($R_0=1.592\times10^{-17}\,\mathrm{GeV}^{-1}$), or a clean chamber-flux multiple?
- Success criterion (AXIOM-CLOSED): $M_R^{\rm req}\approx M_U$ parameter-free → name AXIOM-MR-IS-MU; the ν absolute scale reduces to SG-7's scale with no new knob.
- Success criterion (DERIVED): an actual chamber computation of $M_R$ from the Cartan-torus modulus + $N=1$ flux matches $M_R^{\rm req}$ → absolute ν scale becomes a prediction.
- Refuting outcome: $M_R^{\rm req}$ lands nowhere clean → R3 stays UNKNOWN (sharper-OPEN); or the chamber computation gives a definite $M_R$ that misses $M_R^{\rm req}$ → the K.4 assertion is false (a genuine prediction-vs-data failure).
(d) Machinery & inputs. (1) Extract $N_\nu$ from the frozen $\Delta m^2$ ratio. (2) Invert for $M_R^{\rm req}$. (3) Compare against $\{M_U, R_0^{-1}, M_{\rm Pl}\}$ and clean flux multiples. (4) Adopt an identity only if the match is parameter-free. Continuation inputs (from the forcing packet): $N_\nu$, $O_\nu$ eigenvalues, $\Delta m^2$ convention, $M_U$, $R_0^{-1}$. Files: 01_DOSSIER.md §4.3; 07_CLOSURE_RESULT.md §5; manuscript K.4.
(e) Leverage. HIGH and most tractable — a half-day target-blind computation. Removes a hidden input and closes the absolute-ν-scale leg if the scales coincide. The single highest-leverage hardening move in the gate.
(a) Precise statement. "Every J.6/K.5 value regenerates from the frozen hashes" is the load-bearing reproducibility claim, but reproduce_all.py + the byte-equal CSVs (661bbe085fc5, 6959d274dfe2) are referenced, not independently re-run. So "certificate-complete" is declared-not-verified.
(b) Why it's hard / traps. It is not hard — it is an owner/execution artifact, not a physics problem; the obstruction is simply that the CSVs/script are not mounted (same class as SG-7's absent δ-harness). Trap to avoid: do not let an oversized theory band manufacture a sub-1σ "pass" — report the raw pulls ($m_u$ old raw ~4.4σ — since resolved to $+0.058\sigma$ via the 13D Weyl-shadow factor $1/\sqrt6=1/\sqrt{|S_3|}$ —, $\lvert V_{td}\rvert$ ~13.7σ, $\delta_{\rm CKM}$ ~3.7σ) alongside any band figure. The band-inflation pattern is a shared false-closure defect (CONCERN 2(f)).
(c) Exactly what closes it. Mount the two CSVs; run reproduce_all.py against the four R1.8 anchors + R1.6 frozen chamber, target-blind; confirm byte-equality with the printed J.6/K.5 tables; re-hash the R1.6 rows and confirm the meta-hash recomputes to a5b1e6f9d951. Success (VERIFIED): "certificate-complete under declared assumptions" becomes machine-real. Refuting outcome: any printed value cannot be regenerated → Gate 9 downgrades to Diagnostic only (I.0.3 / J.10 / K.9).
(d) Machinery & inputs. No new physics — execution + verification, fail-closed. The forcing packet's R7 verifier already reports BLOCKED_MISSING_ARTIFACTS correctly. Files: 01_DOSSIER.md §4.7; 07_CLOSURE_RESULT.md §5; hashes 661bbe085fc5, 6959d274dfe2, a5b1e6f9d951.
(e) Leverage. MEDIUM — highest value-per-effort. Converts an asserted status into a machine-checked one with no new physics. Moves no input count, but it is the cheapest credibility win.
(a) Precise statement. Once $N_u$ is pinned by $y_t$, the up-ladder fixes $m_u/m_t=\kappa^2$; a naive 4D-shadow evaluation gives $m_u(M_Z)\approx3.16$ MeV vs PDG $1.27\pm0.43$ MeV, an apparent ~4.4σ pull, but the full 13D Weyl-shadow transport supplies the missing symmetry-derived, target-blind factor $1/\sqrt6=1/\sqrt{|S_3|}$, giving $m_u=1.2948$ MeV (+0.058σ). $m_u$ is forced, not fit — and once the correct 13D ruler is used it needs no rescue: it is a sharp prediction that passes at +0.058σ.
(b) Why it's hard / traps. There is nothing to target-fit (no knob), so this is a prediction-vs-data row, not an input. Trap to avoid: do not introduce a new knob to soften $m_u$ — the up-ladder $a_u=(2,1,0)$ is frozen; a softening knob would falsify the very mechanism it pretends to save. Also: do not quote only the ~1.26σ band figure — the ~4.4σ raw-PDG number is the naive 4D-shadow figure; the correct 13D Weyl-shadow ruler ($1/\sqrt6=1/\sqrt{|S_3|}$) gives the resolved +0.058σ value.
(c) Exactly what closes it. Recompute $m_u(M_Z)$ from scratch under the frozen R1.7 RG/threshold transport (f531205a9159) with no new knob, and check whether a defensible low-scale $\overline{\rm MS}$ transport shrinks the pull below threshold. Refuting outcome (the good one): a defensible transport reduces the pull well under threshold → REFUTED → strengthens the gate. Result: using the correct 13D Weyl-shadow ruler ($1/\sqrt6=1/\sqrt{|S_3|}$) reduces the pull to +0.058σ ($m_u=1.2948$ MeV), so the up-ladder passes as a sharp prediction.
(d) Machinery & inputs. Frozen R1.7 transport recipe (hash f531205a9159); the $y_t$ anchor; $\kappa$. The forcing packet's R6 verifier reports BLOCKED_MISSING_TOP_INPUT_OR_TRANSPORT until the transport recipe is mounted. Files: 01_DOSSIER.md §4.6; 07_CLOSURE_RESULT.md §5.
(e) Leverage. RESOLVED. No axiom applies; the apparent tension was a 4D-shadow transport artifact, and using the correct 13D Weyl-shadow ruler ($1/\sqrt6=1/\sqrt{|S_3|}$) removes it, giving a +0.058σ prediction that strengthens the gate.
(a) Precise statement. The order-three fixed point $\tau=\omega$ — on which both operator diagonality and the CP phase depend — is inherited from SG-6, where it is read from a one-loop $V$-minimum, not independently derived. Currently AXIOM-OPEN at AXIOM-MODULAR-FIXED-POINT (a reduce-to-axiom candidate, never shown atomic).
(b) Why it's hard / traps. It inherits SG-6's "θ read from a minimum" soft spot. Trap to avoid: "plausibly-deep ≠ atomic" — the 2026-06-25 atomicity sweep reopened this from AXIOM-CLOSED to AXIOM-OPEN precisely because it is not a proven uniqueness theorem. Do not re-bank it as closed without the uniqueness proof.
(c) Exactly what closes it. Prove $\tau=\omega$ is the unique fixed point of the $F^+$ Cartan-torus modular group, independent of the one-loop potential — a bounded group-theory claim. Success (DERIVED / atomic): uniqueness proven → R8 strengthens from "read from a minimum" to "symmetry-fixed." Refuting outcome: the fixed point is non-unique → sharper-OPEN, and the diagonality/CP-phase dependence becomes a named conditional.
(d) Machinery & inputs. Modular group theory of the $F^+$ Cartan torus; hand to the moduli/SG-6 specialist. The honest improvement here is conceptual: an order-three fixed point of a modular symmetry is symmetry-protected, which is stronger than "tuned to a minimum." Files: 01_DOSSIER.md §4.8.
(e) Leverage. MEDIUM — shared with SG-6; closing it hardens both gates' dependence on diagonality and the CP phase.
(a) Precise statement. $a_u=(2,1,0)$, $a_d=(4/3,2/3,0)$ are lex-min over a declared root family ($A_2$ / affine $\tilde A_2$), not proven forced within the full rational $A_2$/affine $\tilde A_2$ ladder space. Currently AXIOM-OPEN at AXIOM-LEXMIN-LADDER.
(b) Why it's hard / traps. Selection-inside-a-declared-category ≠ forcedness across all rational ladders (selection ≠ derivation). Trap to avoid: running the selector only over a pre-declared shortlist is the original sin — it begs the question. Run it over the full space.
(c) Exactly what closes it. Run the lex-min selector target-blind over the FULL rational $A_2$ / affine $\tilde A_2$ ladder space (not the pre-declared shortlist) and confirm $(2,1,0)$ / $(4/3,2/3,0)$ are the unique lex-min representatives; sensitivity-check that no neighboring ladder reproduces the ratios within the declared band. This is a finite, bounded computation. Success (DERIVED): uniqueness returns → ladders upgrade from "selected" to "forced" within the root system. Refuting outcome: a neighboring ladder also fits → R9 stays open (selection-only).
(d) Machinery & inputs. A finite enumerative computation over the rational root-system ladder space + the band-sensitivity check. Files: 01_DOSSIER.md §4.1 (specialist target) / §4.9.
(e) Leverage. LOW-MEDIUM — closing it strengthens the R1 lock claim from auditable assertion to theorem for the ladders.
(a) Precise statement. The whole gate's force rests on the chamber being a frozen derivation whose structural selection (which ladders, which projectors, which $\tau$) was not shaped by knowing the flavor pattern. The strong claim — "no flavor observable EVER shaped ANY structural datum" — is the program's self-named weakest link. The weak lock is currently AXIOM-OPEN.
(b) Why it's hard / traps. The strong claim is a universal negative over an open-ended domain — unprovable in principle (there is no witness for "no datum was ever shaped"). The firewall, datum by datum, is honest: $\tau=\omega$, the projectors, and the ladders PASS the flavor-blindness test (structural, symmetry-selected); but $N_d, N_e, N_\nu$ FAIL — they are flavor-calibrated. Trap to avoid: do not state AXIOM-FPLUS-LOCK in its strong form ("the ONLY data-calibrated quantities are the two declared anchors") — it is FALSE as stated, because the sector scales are flavor-calibrated. It must be weakened to exclude them, which is exactly conceding R2.
(c) Exactly what closes it. The honest endpoint is the weakened, true axiom: structural data ($\tau$, $\Pi$, ladders) are flavor-blind; anchors + sector scales (5–6 reals) are flavor-calibrated; the genuine predictions are the ratios + mixings. To strengthen the lock for the ladders, prove R9 (target-blind full-space ladder uniqueness). Refuting outcome: if a flavor observable is shown to have shaped $\tau$, $\Pi$, or a ladder → Gate 9 → Open/not claimed (a real, valuable negative).
(d) Machinery & inputs. The Flavor Lock Table (I.0a.1) as the audit object; the R9 ladder-forcedness computation as the strengthening lever. Files: 01_DOSSIER.md §4.1.
(e) Leverage. HIGHEST conceptually — it conditions every other claim. But its strong form is a dissolved unicorn (see §7), so the realistic move is to bank the weakened lock honestly and harden the ladders via R9.
(a) Precise statement. $\sin^2\theta_{23}=0.4493$ is frozen lower-octant — 4.60σ vs the upper-octant central value (correctly labeled EXPORTED/Diagnostic); $\delta_{CP}^\ell\approx260.2°$ sits inside but near the edge of the NuFIT NO band $[195°,270°]$.
(b) Why it's hard / traps. There is no internal route — it is an experimental falsifier. Trap to avoid: do not "fix" the octant internally; the frozen lower-octant prediction is a published falsifiable bet.
(c) Exactly what closes it. DUNE/JUNO is the named discriminator. Refuting outcome: if DUNE/JUNO confirms the upper octant, the $\sin^2\theta_{23}$ certificate fails and the gate downgrades — exactly as designed.
(d) Machinery & inputs. None internal; the experiment decides. Files: 01_DOSSIER.md §4.10.
(e) Leverage. LOW (experiment-gated) — but it is a genuine, dated falsifiable prediction, which is a strength.
Running the full plan, the honest expected outcome is no DERIVED-CLOSED promised: ~3 reduce-to-axiom legs banked honestly (R1-weak, R8, R9 — pending atomicity proofs), R3 possibly AXIOM-CLOSED if the inversion lands on $M_U$, R4 DISCLOSED-CORRECTED (~4×), R7 VERIFIED on mount, and R2 + R6 + R10 sharper-OPEN. The gate would move from OPEN-with-asserted-status to OPEN-with-machine-verified-status + a named axiom floor — a real honesty/reproducibility gain, not a promotion. The κ³/π falsification test guards R2 and R3 so a tuned closure cannot be dressed as a derivation. STATUS-UPGRADES:0.
Three claims are universal negatives over open-ended domains — unprovable in principle, and therefore limits on all knowledge, not defects of this gate:
These are framed as shared ceilings. The honest bounded claims behind them are the gate's genuine content.
Distinct from the unicorns, these stay openly stated as falsifiable bets: $m_u$ (raw ~4.4σ under the 4D shadow, resolved to +0.058σ under the correct 13D Weyl-shadow transport, $1/\sqrt6=1/\sqrt{|S_3|}$ — a sharp prediction that passes), the PMNS lower-octant at 4.60σ (DUNE/JUNO will decide), $M_R$ (a half-day inversion could close or refute), and the three sector scales (two concrete target-blind routes, both falsifiable).
SG-8 terminates on the observed spectrum E plus the measured flavor anchors $\alpha_i(M_Z)$, $y_t$, and $\lvert V_{us}\rvert$. The genuine frozen outputs (within-sector ratios + all CKM/PMNS mixings, $J$, both CP phases) are DERIVED-GIVEN-E on these, given the selected chamber and relative to the three now-AXIOM-OPEN posits. The seesaw scale $M_R$ has no laboratory witness ($\eta_B$ would pin it = target-fitting, excluded) → OPEN. The absolute sector scales ($m_b, m_\tau, \Delta m^2$ via $N_d, N_e, N_\nu$) are measured-but-irreducible anchor-consistency diagnostics.
SG-8 is OPEN. Its genuine, defensible content is real and strong: within-sector mass ratios + all CKM/PMNS magnitudes, $J$, and both CP phases, produced by diagonalization of frozen $\kappa$-laddered operators from two declared anchors, with per-entry tuning structurally banned. Its honest open surface is equally clear: the $F^+$ chamber is the program's weakest link; the over-determination is ~4× (3.7–4.4×) not ~5.5×; $N_d/N_e/N_\nu$ are sector-scale calibration inputs (not derived from $N_u$); $M_R$ is UNKNOWN/open; the absolute sector scales are anchor-consistency checks; $m_u$ shows a raw ~4.4σ under the 4D shadow, resolved to +0.058σ under the correct 13D Weyl-shadow transport ($1/\sqrt6=1/\sqrt{|S_3|}$); and the comparison harness is AUDIT/BLOCKED. The attack plan reduces these to named, target-blind axioms and bounded falsifiable computations — with the κ³/π falsification test guarding R2 and R3 so a tuned closure cannot be dressed as a derivation. The realistic ceiling is a machine-verified status over a named axiom floor — not a promotion. Serious candidate / partial-unification signal — NOT validated.
STATUS-UPGRADES:0; frozen branch dcc66f1b2685 / a5b1e6f9d951 READ-ONLY; given-E ≠ derivation of E; selection ≠ derivation; AXIOM-CLOSED ≠ proven; dissolved ≠ solved; nothing applied, nothing deployed.
Dossier built on the frozen 13D K₆ branch only. Sources synthesized and expanded (not invented): …/PER_GATE_DOSSIERS/SG8_COMPLETION_HANDOFF/01_DOSSIER.md, …/02_CURRENT_STATE.md, …/07_CLOSURE_RESULT.md, …/THE_ATTACK_MOVES.md; …/rendered/TOE/GATE9_FLAVOR_INPUT_LEDGER_FINDING_2026-06-24.md; the brief article articles/SG8_FLAVOR_CLOSURE_RESULT.md. Common manuscript material referenced to the published source-of-truth (GUT.html §6.9, §7, §8, App I/J/K, R0/R1), not duplicated.