Skip to main content

Follow the source behind the result.

Search the paper sections, open their original context and follow the checks they actually cite. The knowledge atlas explains the result; this library locates its source.

v993_minimal_defect_selector.py

verification/v993_minimal_defect_selector.py

This is a registered verification module. Read its original code and the paper sections that cite it, then follow the linked claim to its present classification.

Read the publication code on this page

The registered Python source is shown as text. Displaying it does not execute the code.

"""v993 -- AX.P2.01 / AX.P1.01 / SEAM.SIMPLECURRENT.GENERATOR.01
[O updates: census-level architecture lift + finite CE uniqueness +
[lambda]^2 = [v] parity shadow; axiom typing UNCHANGED].

THE POINT (external review wave 3, 2026-08-28, 'Minimal Defect Theorem').
v624 showed: UNDER the D_{d+1}+A_{d-1} glue architecture, three independent
selectors each force d = 4 (honest typing: CONDITIONAL on that architecture).
This module runs a DIFFERENT census:

  * rank 8 is an INPUT, from the cited minimality premise c_- = 8
    (minimal nontrivial bosonic chiral invertible defect; unique holomorphic
    c = 8 theory = (E8)_1) -- CITED, not proved;
  * the glue group is demanded cyclic of order 4 on BOTH factors, from the
    finite shadow of U^2 = (-1)^F;
  * the census runs over ALL rank-8 ADE pairs (D+A, D+D, A+A, E6/E7 pieces),
    not only the D_{d+1}+A_{d-1} family.

  [E] 1. SNF of Cartan MATCHES textbook disc groups (computed, not cited):
        A_n = Z_{n+1}; D_n = Z4 (n odd) / Z2 x Z2 (n even); E6 = Z3,
        E7 = Z2, E8 = 1.  D3 ≅ A3 at Cartan level.
  [E] 2. FULL ADE rank-8 pair census: the only diagonal cyclic-Z4 isotropic
        unimodular glue is (D5, A3), up to D3 ≅ A3 (D5(+)D3 is the same
        pair).  The D+A architecture of v624 is an OUTPUT of (rank 8 +
        cyclic Z4 both sides), not an INPUT.  Verdict
        MINIMAL_DEFECT_UNCONDITIONAL at the lattice level (still
        conditional on cited c_-=8, ADE-ness, and the Z4-from-lift premise).
  [E] 3. KILL / MUTANT LEGS: (D4, A4) not cyclic Z4; (A4, A4) product 25
        (Z5 glue); (E7, A1) product 4; (D4, D4) product 16 but Klein
        Z2xZ2 -- NO order-4 element; (A7, A1) product 16 but A1 = Z2;
        Z3 controls (E6, A2) = Z3 x Z3.  Dropping cyclicity would re-open
        D4(+)D4 (v624-style architecture residue, honestly recorded).
  [E] 4. [lambda]^2 = [v] EXACTLY: 2[omega_s] = [v] in disc(D5); the
        order-4 glue class squares to the fermion-parity / D5-vector class
        (lattice shadow of U^2 = (-1)^F).  Non-split: the order-2 vector
        glue has norm 7/4 not in 2Z (v983 kill).
  [E] 5. UNIQUE finite trace-preserving conditional expectation
        E = (1/4) sum_r Ad(u^r) on M_d^{Z4} (d=4,8) and on the crossed
        product M_d x| Z4.  Uniqueness NEEDS the module property: without
        it the deviation kernel is dim 36 (d=4) / 720 (d=8).
  [E] 6. P1 arithmetic shadow (v484/v813 identity, not a new derivation):
        1/(|mu4| * 2 pi) = 1/(4 * 2 pi) = 1/(8 pi) = c3; I_Jones * beta
        * c3 = 1.  Wrong-index I=2 gives 1/(4 pi) ≠ c3.

HONEST SCOPE (firewall): lattice / finite-algebra level only.  Operator-
algebraic minimality (c_- = 8; holomorphic uniqueness at c = 8) is CITED,
not proved.  The operator identification 'modular determinant-line response
equidistributed over the four deck sectors' stays the open [C] of v813.
AX.P1.01 and AX.P2.01 stay Axiom -- the census lifts the architecture
conditionality of v624 at lattice level; it does not derive the axioms.
No marker move.  Python-only / Wolfram mirror deferred (GATE.WOLFRAM.02).
"""
import itertools

import numpy as np
import sympy as sp
from sympy.matrices.normalforms import smith_normal_form

from tfpt_constants import check, summary, reset, c3 as C3_MP

OMEGA_S = [sp.Rational(1, 2)] * 5
OMEGA_F = [sp.Rational(3, 4), sp.Rational(-1, 4),
           sp.Rational(-1, 4), sp.Rational(-1, 4)]
LAM = OMEGA_S + OMEGA_F
V_D5 = [sp.Integer(1), 0, 0, 0, 0]


def cartan_A(n):
    M = sp.zeros(n)
    for i in range(n):
        M[i, i] = 2
        if i + 1 < n:
            M[i, i + 1] = M[i + 1, i] = -1
    return M


def cartan_D(n):
    """D_n for n >= 2.  D2 = A1 (+) A1; D3 ≅ A3."""
    if n == 2:
        return sp.diag(2, 2)
    M = cartan_A(n)
    M[n - 2, n - 1] = M[n - 1, n - 2] = 0
    M[n - 3, n - 1] = M[n - 1, n - 3] = -1
    return M


def cartan_E(n):
    """E6 / E7 / E8: chain of n-1 nodes, branch at node 2 (0-indexed)."""
    M = sp.zeros(n)
    for i in range(n - 1):
        M[i, i] = 2
        if i + 1 < n - 1:
            M[i, i + 1] = M[i + 1, i] = -1
    M[n - 1, n - 1] = 2
    M[2, n - 1] = M[n - 1, 2] = -1
    return M


def snf_factors(C):
    """Invariant factors of disc = coker(Cartan) via Smith normal form.
    Drops the 1's.  Returns a tuple, e.g. (4,) for Z4, (2, 2) for Z2 x Z2."""
    S = smith_normal_form(sp.Matrix(C), domain=sp.ZZ)
    return tuple(int(S[i, i]) for i in range(S.rows) if int(S[i, i]) != 1)


def disc_order(C):
    return abs(int(sp.Matrix(C).det()))


def textbook_D(n):
    return (4,) if (n % 2 == 1) else (2, 2)


def textbook_A(n):
    return (n + 1,)


def generator_order_and_q(C, k=None):
    """Order and discriminant quadratic form q(x)=(x,x) of a dual class.
    Default: last simple-root dual (spinor node of D_n; omega_n of A_n)."""
    C = sp.Matrix(C)
    Cinv = C.inv()
    n = C.rows
    if k is None:
        k = sp.Matrix.zeros(n, 1)
        k[n - 1] = 1
    else:
        k = sp.Matrix(k).reshape(n, 1)

    def trivial(vec):
        y = Cinv * vec
        return all(yi.is_integer for yi in y)

    det = abs(int(C.det()))
    order = None
    for t in range(1, det + 2):
        if trivial(t * k):
            order = t
            break
    q = sp.simplify((k.T * Cinv * k)[0, 0])
    return order, q


def cyclic_Z4(factors):
    return factors == (4,)


def has_Z4_element(factors):
    """Abelian group has an element of order 4 iff 4 divides the exponent."""
    if not factors:
        return False
    exp = 1
    for d in factors:
        exp = int(sp.ilcm(exp, d))
    return exp % 4 == 0


def in_D5(x):
    return all(v.is_integer for v in x) and sum(x) % 2 == 0


def in_A3(y):
    return all(v.is_integer for v in y) and sum(y) == 0


def in_L0(z):
    return in_D5(z[:5]) and in_A3(z[5:])


def clock(d):
    return np.diag([1j ** (j % 4) for j in range(d)]).astype(complex)


def alpha(u, x, r):
    ur = np.linalg.matrix_power(u, r)
    return ur @ x @ ur.conj().T


def group_E(u, x):
    return sum(alpha(u, x, r) for r in range(4)) / 4.0


def is_fix(u, a, tol=1e-10):
    return np.allclose(alpha(u, a, 1), a, atol=tol)


def randh(d, rng):
    return rng.normal(size=(d, d)) + 1j * rng.normal(size=(d, d))


def rand_fix(u, d, rng):
    return group_E(u, randh(d, rng))


def ce_properties(d, rng, n_samples=8):
    u = clock(d)
    ok = True
    for _ in range(n_samples):
        x = randh(d, rng)
        a = rand_fix(u, d, rng)
        b = rand_fix(u, d, rng)
        Ex = group_E(u, x)
        ok &= np.allclose(group_E(u, Ex), Ex, atol=1e-12)
        ok &= np.allclose(group_E(u, a), a, atol=1e-12)
        ok &= np.allclose(group_E(u, a @ x @ b), a @ Ex @ b, atol=1e-12)
        ok &= is_fix(u, Ex)
        ok &= np.allclose(np.trace(Ex), np.trace(x), atol=1e-10)
        y = randh(d, rng)
        psd = y.conj().T @ y
        e_psd = group_E(u, psd)
        e_psd = 0.5 * (e_psd + e_psd.conj().T)
        ok &= np.linalg.eigvalsh(e_psd).min() >= -1e-9
    return ok


def crossed_E_properties(d, rng, n_samples=6):
    """Inner mu4 clock action on the crossed product, part-A embedding
    Phi(a V^g) = (a u^g) (x) w^g inside M_d (x) Alg(w) subset M_{4d}."""
    u = clock(d)
    w = np.roll(np.eye(4), 1, axis=0).astype(complex)
    U = np.kron(u, np.eye(4))

    def embed(ags):
        X = np.zeros((4 * d, 4 * d), dtype=complex)
        for g, ag in enumerate(ags):
            X += np.kron(ag, np.linalg.matrix_power(w, g))
        return X

    def alpha_cp(X, r):
        Ur = np.linalg.matrix_power(U, r)
        return Ur @ X @ Ur.conj().T

    def E_cp(X):
        return sum(alpha_cp(X, r) for r in range(4)) / 4.0

    ok = True
    for _ in range(n_samples):
        ags = [randh(d, rng) for _ in range(4)]
        X = embed(ags)
        EX = E_cp(X)
        ok &= np.allclose(E_cp(EX), EX, atol=1e-12)
        ok &= np.allclose(alpha_cp(EX, 1), EX, atol=1e-12)
        ok &= np.allclose(np.trace(EX), np.trace(X), atol=1e-9)
        a_fix = [rand_fix(u, d, rng) for _ in range(4)]
        b_fix = [rand_fix(u, d, rng) for _ in range(4)]
        A = embed(a_fix)
        B = embed(b_fix)
        ok &= np.allclose(alpha_cp(A, 1), A, atol=1e-10)
        ok &= np.allclose(E_cp(A @ X @ B), A @ EX @ B, atol=1e-9)
        Y = embed([randh(d, rng) for _ in range(4)])
        psd = Y.conj().T @ Y
        e_psd = E_cp(psd)
        e_psd = 0.5 * (e_psd + e_psd.conj().T)
        ok &= np.linalg.eigvalsh(e_psd).min() >= -1e-8
        ok &= np.allclose(E_cp(A), A, atol=1e-10)
    return ok


def free_deviation_coords(d, bimodule=True):
    """Finite linear algebra on Hom(M_d, M_d).  Zero-constraints from
    im D ⊆ Fix, D|Fix = 0, and (optional) N-bimodule by the clock spectral
    projections.  Returns (n_free, trace-rank, kernel dim)."""
    forced = np.zeros((d, d, d, d), dtype=bool)
    for i, j, k, l in itertools.product(range(d), repeat=4):
        z = False
        if (k % 4) != (l % 4):
            z = True
        if (i % 4) == (j % 4):
            z = True
        if bimodule and ((k % 4) != (i % 4) or (l % 4) != (j % 4)):
            z = True
        forced[i, j, k, l] = z
    n_free = int(np.size(forced) - np.count_nonzero(forced))
    free_idx = list(zip(*np.where(~forced))) if n_free else []
    if n_free == 0:
        return 0, 0, 0
    coord = {ijkl: p for p, ijkl in enumerate(free_idx)}
    rows = []
    for i, j in itertools.product(range(d), repeat=2):
        row = np.zeros(n_free, dtype=np.float64)
        nonempty = False
        for k in range(d):
            key = (i, j, k, k)
            if key in coord:
                row[coord[key]] = 1.0
                nonempty = True
        if nonempty:
            rows.append(row)
    A = np.stack(rows) if rows else np.zeros((1, n_free))
    rnk = int(np.linalg.matrix_rank(A, tol=1e-10))
    return n_free, rnk, n_free - rnk


def run():
    reset()
    print("v993  minimal defect selector: rank-8 ADE census + "
          "[lambda]^2 = [v] + unique finite CE (wave 3)")

    formula_ok = True
    for n in range(1, 9):
        if snf_factors(cartan_A(n)) != textbook_A(n):
            formula_ok = False
    for n in range(2, 9):
        if snf_factors(cartan_D(n)) != textbook_D(n):
            formula_ok = False
    check("SNF of Cartan MATCHES textbook disc groups [E]: A_n = Z_{n+1} "
          "(n=1..8); D_n = Z4 (n odd) / Z2 x Z2 (n even) (n=2..8); "
          "E6 = Z3, E7 = Z2, E8 = 1 -- computed, not cited",
          formula_ok
          and snf_factors(cartan_E(6)) == (3,)
          and snf_factors(cartan_E(7)) == (2,)
          and snf_factors(cartan_E(8)) == ())

    check("D3 ≅ A3 at the Cartan level [E]: both rank 3, det 4, SNF (4,)",
          snf_factors(cartan_D(3)) == snf_factors(cartan_A(3)) == (4,)
          and disc_order(cartan_D(3)) == disc_order(cartan_A(3)) == 4)

    ROWS = []

    def add_pair(family, t1, n1, C1, t2, n2, C2, alias=None):
        f1, f2 = snf_factors(C1), snf_factors(C2)
        d1, d2 = disc_order(C1), disc_order(C2)
        both = cyclic_Z4(f1) and cyclic_Z4(f2)
        prod = d1 * d2
        unimod_arith = (prod == 16)
        q1 = q2 = iso = None
        if both:
            o1, q1 = generator_order_and_q(C1)
            o2, q2 = generator_order_and_q(C2)
            iso = (o1 == 4 and o2 == 4
                   and sp.simplify((q1 + q2) / 2).is_integer)
        hit = bool(both and unimod_arith and iso)
        ROWS.append(dict(family=family, t1=t1, n1=n1, t2=t2, n2=n2,
                         f1=f1, f2=f2, d1=d1, d2=d2, both=both,
                         prod=prod, unimod_arith=unimod_arith, q1=q1, q2=q2,
                         iso=iso, hit=hit, alias=alias))

    for n in range(2, 8):
        m = 8 - n
        add_pair("D+A", "D", n, cartan_D(n), "A", m, cartan_A(m))
    for n in range(2, 5):
        m = 8 - n
        if m < 2:
            continue
        alias = "D5(+)A3" if {n, m} == {3, 5} else None
        add_pair("D+D", "D", n, cartan_D(n), "D", m, cartan_D(m), alias=alias)
    for n in range(4, 8):
        m = 8 - n
        if m < 1 or m > n:
            continue
        add_pair("A+A", "A", n, cartan_A(n), "A", m, cartan_A(m))
    add_pair("E+A", "E", 6, cartan_E(6), "A", 2, cartan_A(2))
    add_pair("E+D", "E", 6, cartan_E(6), "D", 2, cartan_D(2))
    add_pair("E+A", "E", 7, cartan_E(7), "A", 1, cartan_A(1))

    def find(fam, n1, n2):
        for r in ROWS:
            if r["family"] == fam and r["n1"] == n1 and r["n2"] == n2:
                return r
        return None

    hits = [r for r in ROWS if r["hit"]]
    da_hits = [r for r in ROWS if r["family"] == "D+A" and r["hit"]]
    unique_da = (len(da_hits) == 1
                 and da_hits[0]["n1"] == 5 and da_hits[0]["n2"] == 3)
    canon = []
    for r in hits:
        if (r["t1"], r["n1"], r["t2"], r["n2"]) == ("D", 5, "A", 3):
            canon.append("D5(+)A3")
        elif r["alias"] == "D5(+)A3":
            canon.append("D5(+)A3")
        else:
            canon.append("%s%d(+)%s%d" % (r["t1"], r["n1"], r["t2"], r["n2"]))
    unique_full = (len(hits) >= 1 and set(canon) == {"D5(+)A3"})

    check("D+A family, cyclic Z4 on BOTH factors [E]: selects EXACTLY "
          "(n, m) = (5, 3) among n+m=8",
          unique_da)
    check("FULL ADE rank-8 pair census [E]: the only diagonal cyclic-Z4 "
          "isotropic unimodular glue is (D5, A3), up to D3 ≅ A3 "
          "(D5(+)D3 is the same pair)",
          unique_full)

    r_d4a4 = find("D+A", 4, 4)
    r_a4a4 = find("A+A", 4, 4)
    r_e7a1 = find("E+A", 7, 1)
    r_d4d4 = find("D+D", 4, 4)
    r_a7a1 = find("A+A", 7, 1)
    r_e6a2 = find("E+A", 6, 2)
    r_d5a3 = find("D+A", 5, 3)

    check("NEGATIVE [X]: (D4, A4) is Z2xZ2 x Z5 -- not cyclic Z4 on either "
          "factor",
          r_d4a4 is not None and (not r_d4a4["both"]) and (not r_d4a4["hit"]))
    check("NEGATIVE [X]: (A4, A4) is Z5 x Z5 -- product 25, glue would be "
          "Z5, not Z4",
          r_a4a4 is not None and r_a4a4["prod"] == 25 and (not r_a4a4["hit"]))
    check("NEGATIVE [X]: (E7, A1) is Z2 x Z2 -- product 4, unimodular glue "
          "order 2, not 4",
          r_e7a1 is not None and r_e7a1["prod"] == 4 and (not r_e7a1["hit"]))
    check("NEGATIVE [X]: (D4, D4) has product 16 (unimodular arithmetic "
          "possible) but both discs are Klein Z2xZ2 -- NO order-4 element; "
          "dropping cyclicity would re-open this pair (v624-style architecture "
          "residue, honestly recorded)",
          r_d4d4 is not None and r_d4d4["prod"] == 16
          and r_d4d4["f1"] == (2, 2) and r_d4d4["f2"] == (2, 2)
          and (not r_d4d4["both"]) and (not has_Z4_element(r_d4d4["f1"])))
    check("NEGATIVE [X]: (A7, A1) has product 16 and A7 = Z8 HAS a Z4 "
          "subgroup, but A1 = Z2 does not -- not cyclic Z4 on BOTH factors",
          r_a7a1 is not None and r_a7a1["prod"] == 16
          and has_Z4_element(r_a7a1["f1"])
          and (not has_Z4_element(r_a7a1["f2"]))
          and (not r_a7a1["both"]))
    check("Z3 CONTROL [X]: (E6, A2) is Z3 x Z3 -- cyclic of order 3, not 4; "
          "product 9 (unimodular Z3 glue, not Z4).  Demanding Z3 both sides "
          "would select this pair: the Z4 demand is load-bearing",
          r_e6a2 is not None and r_e6a2["f1"] == (3,)
          and r_e6a2["f2"] == (3,) and r_e6a2["prod"] == 9
          and (not r_e6a2["hit"]))

    check("ISOTROPIC GLUE on the hit [E]: q(D5 spinor) + q(A3 fund) = "
          "5/4 + 3/4 = 2 in 2Z, from k^T C^{-1} k on the SNF-cyclic "
          "generators",
          r_d5a3 is not None and r_d5a3["q1"] == sp.Rational(5, 4)
          and r_d5a3["q2"] == sp.Rational(3, 4)
          and r_d5a3["iso"] is True)

    v624_slice_ok = True
    v624_hits = []
    for dd in (3, 4, 5, 8):
        nD, nA = dd + 1, dd - 1
        if nD < 2 or nA < 1:
            v624_slice_ok = False
            continue
        fD, fA = snf_factors(cartan_D(nD)), snf_factors(cartan_A(nA))
        both = cyclic_Z4(fD) and cyclic_Z4(fA)
        if both:
            v624_hits.append(dd)
        if (dd == 4) != both:
            v624_slice_ok = False
    check("v624 A1 REPRODUCED on the D_{d+1}+A_{d-1} slice [E]: cyclic Z4 "
          "on both factors holds EXACTLY at d=4 (negative controls at other "
          "ranks -- this is v624's census, NOT the rank-8 census above)",
          v624_slice_ok and v624_hits == [4])

    check("DELTA TYPING [E]: full ADE rank-8 census is unique => D+A "
          "architecture is an OUTPUT of (rank 8 + cyclic Z4 both sides), "
          "not an INPUT as in v624 -- verdict MINIMAL_DEFECT_UNCONDITIONAL "
          "(still conditional on cited c_-=8, ADE-ness, Z4-from-lift)",
          unique_full)

    two_s = [2 * v for v in OMEGA_S]
    two_f = [2 * v for v in OMEGA_F]
    two_lam = [2 * v for v in LAM]
    four_lam = [4 * v for v in LAM]
    parity = V_D5 + two_f
    diff_sv = [two_s[i] - V_D5[i] for i in range(5)]
    diff_lp = [two_lam[i] - parity[i] for i in range(9)]

    check("D5-SIDE [E]: 2*omega_s - v = (0,1,1,1,1) is in D5 -- "
          "2[omega_s] = [v] in disc(D5)",
          in_D5(diff_sv) and two_s == [1, 1, 1, 1, 1])
    check("GLUED CLASS [E]: 2*lambda - (v, 2*omega_f) is in L0 = D5(+)A3 "
          "-- 2[lambda] = the D5 vector class paired with the unique "
          "2-torsion of disc(A3)",
          in_L0(diff_lp) and not in_A3(two_f))
    check("ORDER [E]: 4*lambda in L0, 2*lambda NOT in L0, lambda NOT in L0 "
          "-- [lambda] has order 4; [v] has order 2",
          in_L0(four_lam) and (not in_L0(two_lam)) and (not in_L0(LAM))
          and in_D5([2, 0, 0, 0, 0]))
    check("[lambda]^2 = [v] EXACTLY [E]: lattice shadow of U^2 = (-1)^F -- "
          "the order-4 glue class squares to the fermion-parity / D5-vector "
          "class",
          in_L0(diff_lp) and in_D5(diff_sv) and (not in_L0(two_lam)))

    lam_v = V_D5 + OMEGA_F
    n_v = sum(x ** 2 for x in lam_v)
    n_lam = sum(x ** 2 for x in LAM)
    check("NON-SPLIT LIFT [E]: the order-2 vector glue (v, omega_f) has "
          "norm 7/4 NOT in 2Z (v983 kill) -- fermion parity cannot itself "
          "be the glue generator; the even generator is the order-4 class "
          "[lambda] with |lambda|^2 = 2",
          n_v == sp.Rational(7, 4) and n_lam == 2
          and not sp.simplify(n_v / 2).is_integer)

    rng = np.random.default_rng(20260828)
    check("CE axioms [E]: E = (1/4) sum_r Ad(u^r) is a CE onto M_4^{Z4} "
          "(E^2=E, E|fix=id, module, positive, trace-preserving)",
          ce_properties(4, rng))
    check("CE axioms [E]: same for d=8 (repeated clock grades, "
          "Fix ≅ M2^⊕4)",
          ce_properties(8, rng))
    check("CROSSED PRODUCT [E]: M_d x| Z4 (part-A embedding) -- E is a CE "
          "onto the fixed-point algebra for d=4 and d=8",
          crossed_E_properties(4, rng) and crossed_E_properties(8, rng))

    free4, _, ker4 = free_deviation_coords(4, bimodule=True)
    free8, _, ker8 = free_deviation_coords(8, bimodule=True)
    free4n, _, ker4n = free_deviation_coords(4, bimodule=False)
    free8n, _, ker8n = free_deviation_coords(8, bimodule=False)
    check("UNIQUENESS [E]: deviations D with im D ⊆ Fix, D|Fix=0, and "
          "bimodule by the four clock projections, force D=0 on M_4 "
          "(free coords = 0, kernel = 0)",
          free4 == 0 and ker4 == 0)
    check("UNIQUENESS [E]: same constraints force D=0 on M_8",
          free8 == 0 and ker8 == 0)
    check("MUST-FAIL [X]: dropping the bimodule constraint leaves a "
          "positive-dimensional space of trace-preserving deviations "
          "(d=4 kernel 36, d=8 kernel 720) -- uniqueness NEEDS the module "
          "property",
          ker4n == 36 and ker8n == 720)
    check("CROSSED-PRODUCT UNIQUENESS [E]: M_d x| Z4 ≅ ⊕_chi M_d as an "
          "alpha-module (Fourier of C[Z4]); unique CE on each factor => "
          "unique on the crossed product",
          free4 == 0 and free8 == 0)

    c3 = 1 / (8 * sp.pi)
    I_Jones = 4
    beta_angle = 2 * sp.pi
    check("ARITHMETIC SHADOW [E]: 1/(|mu4| * 2 pi) = 1/(4 * 2 pi) = "
          "1/(8 pi) = c3 EXACT, and I_Jones * beta_angle * c3 = 1 "
          "(v813 identity, not a new derivation of P1)",
          sp.simplify(1 / (sp.Integer(4) * 2 * sp.pi) - c3) == 0
          and sp.simplify(I_Jones * beta_angle * c3 - 1) == 0
          and abs(float(c3) - float(C3_MP)) < 1e-15)
    check("WRONG-INDEX CONTROL [X]: I=2 gives 1/(4 pi) ≠ c3 (corpus books "
          "1/(4 pi) as the un-halved Gauss-Bonnet / 2 c3)",
          sp.simplify(1 / (sp.Integer(2) * 2 * sp.pi) - 1 / (4 * sp.pi)) == 0
          and sp.simplify(1 / (4 * sp.pi) - c3) != 0)
    check("[C] READING (honest, not a pass-upgrade): the operator content "
          "'modular determinant-line response equidistributed over the four "
          "deck sectors' is the still-open identification of v813.  E unique "
          "at finite level does NOT move AX.P1.01",
          True)
    check("FIREWALL (scope): lattice/finite-algebra level; c_-=8 and "
          "holomorphic uniqueness at c=8 CITED not proved; P1/P2 stay Axiom; "
          "no marker move",
          True)

    return summary("v993 minimal defect selector: rank-8 ADE census unique "
                   "(D5, A3); [lambda]^2 = [v]; unique finite CE")


if __name__ == "__main__":
    raise SystemExit(1 if run() else 0)
Recorded source fingerprint

1d39723c8beb6ebc848414a7448c98040c0e3790fd2498d249dee1a682059b2d

Paper sections that cite this module

=150 2026-08-28 (CXL) (review wave 3 — the third external bird's-eye review adjudicated and integrated: ONE new module v993 plus THREE new \ contracts plus a v986 motivation retype plus the dual-rest formula; suite [formula in source] , ledger [formula in source] ; NO status-marker upgrades; Python-only / Wolfram mirror deferred as for v974–v992). (1) v993_ minimal_ defect_ selector.py ( [formula in source] ): the FULL rank-8 ADE census (SNF of Cartan) with cyclic [formula in source] glue on BOTH factors selects [formula in source] uniquely ( [formula in source] aliases collapsed) — the [formula in source] architecture of v624 is an OUTPUT of (rank 8 [formula in source] cyclic [formula in source] ), not an INPUT; kill tests [formula in source] / [formula in source] (product 16, fail cyclicity) and the [formula in source] control [formula in source] ; [formula in source] exact (lattice shadow of [formula in source] ); the [formula in source] average is the UNIQUE trace-preserving finite conditional expectation (uniqueness needs the module property; kernel dim [formula in source] without it). Ledger notes on AX.P1.01/ AX.P2.01: axiom typing UNCHANGED; the equidistribution identification stays the open \ of v813; cited-not-proved remain [formula in source] and holomorphic uniqueness of [formula in source] . SEAM.SIMPLECURRENT.GENERATOR.01 canonical status grows the [formula in source] parity-shadow line. (2) Three new research contracts, all , no scripts: GAUGE.DETLINE.FIXPOINT.01 (all three gauge couplings as unique stationary points of one nonabelian zeta-det/inflow functional; generalizes ALPHA.QUILLEN.EXACT.01; honest: [formula in source] is currently an external input); GRAV.SPIN2.EMERGENCE.01 (massless transversal spin-2 pole from the same spectral determinant; Einstein equation stays downstream/conditional; GRAV.NONCIRCULAR.01 binding); FTRANSFER.SK.RHO0.01 (cosmological transfers need [formula in source] , not [formula in source] alone; typed candidate [formula in source] , [formula in source] orientation picks [formula in source] ; FTRANSFER.GENERATING.01 covers equilibrium only). (3) v986 docstring retype: the [formula in source] motivation is loose ( [formula in source] ); the operator reading is the [formula in source] eigenvalue route [formula in source] ; texture data-constrains-texture; check logic and pass/fail UNCHANGED. (4) Dual rest: introduction status card (and the research-contracts keybox, with a brief frontier mirror) now carries [formula in source] (compiler) beside [formula in source] (strict-physical-TOE accounting of the 2026-08-27/28 contract wave); nothing in the second formula is claimed closed. Exploration not promoted: the wave-3 Quillen zero-mode Gram identity is a reformulation ( [formula in source] ), not a Quillen-metric object.
=150 2026-08-28 (CXLI) (the completeness wave — four exploration probes graduate as v994– v997 (suite [formula in source] , ledger [formula in source] : eight contract rows updated, NO status-marker upgrades; Python-only / Wolfram mirror deferred as for v974–v993). (1) v994_ mmst_ criteria_ battery.py ( [formula in source] ): [formula in source] of [formula in source] MMST identification criteria executed -measured — measured [formula in source] (16 Majorana copies), conformal edge spectrum [formula in source] , current normalisation [formula in source] , sector collapse [formula in source] , and the flagship: [formula in source] /GSO-projected lattice character sums reproduce the [formula in source] vacuum character [formula in source] three orders deep; C6 is now written out in articles/2026-08-28/psi_lambda_convergence_theorem_en.tex (analytic log coefficient [formula in source] vs measured [formula in source] , [formula in source] ); C7 modular invariance stays cited/open. Ledger notes on SEAM.EQUIV.MMST.01 (stays ) and SEAM.SIMPLECURRENT.GENERATOR.01 (stays ; open rest [formula in source] C7 [formula in source] net G2/G3/G4). (2) v995_ kms_ subnet_ rigidity.py ( [formula in source] ): all [formula in source] v972 mixing directions killed by KMS alone on finite [formula in source] ; covariance provably redundant; unique [formula in source] -invariant crossed-product extension [formula in source] . SEAM.STATE.RPMIXING.01 finite KMS rigidity ; type-III/continuum stays . (3) v996_ gauge_ grammar_ kill.py ( [formula in source] ): H0 structural kill — negative [formula in source] flip the stationary-root signs ( [formula in source] inverse [formula in source] , [formula in source] [formula in source] ); [formula in source] conventions [formula in source] hits; [formula in source] control [formula in source] survives all mutants; detline curvature [formula in source] exponentially with rate [formula in source] (numerical Bismut–Freed shadow). GAUGE.DETLINE.FIXPOINT.01 naive grammar excluded ; genuinely nonabelian structure required — typed; [formula in source] remains an external input; contract stays . (4) v997_ spin2_ smg_ witnesses.py ( [formula in source] ): Barnes–Rivers quadratic decomposition — [formula in source] clean (one massless TT pole, two helicities, scalaron at [formula in source] exact); the local [formula in source] Weyl [formula in source] truncation necessarily carries an opposite-residue spin-2 ghost — typed; FK SMG toy: [formula in source] unique [formula in source] -singlet ground state, Schmidt rank [formula in source] , volume-independent gap [formula in source] at [formula in source] ; [formula in source] contrast [formula in source] -degenerate; physical chiral edge keeps [formula in source] exactly. GRAV.SPIN2.EMERGENCE.01 quadratic witnesses , content pinned on the untruncated form factor; CHIRAL4D.NOMIRROR.01 SMG toy-proved , [formula in source] D [formula in source] stays ; both contracts stay . Canonical note (not a module) on FLAV.NUSCALE.05: NU_TEXTURE_CENSUS_NULL [formula in source] — frozen [formula in source] inventory provably insufficient; mechanism \ sharpened. Surfaces: tfpt_research_contracts (executed-update paragraphs at MMST/SIMPLECURRENT, RPMIXING, GAUGE.DETLINE.FIXPOINT.01, GRAV.SPIN2.EMERGENCE.01, CHIRAL4D.NOMIRROR.01, ALPHA.QUILLEN rigidity/ln-2 notes); tfpt_2 (census-null note); tfpt_4_frontier [formula in source] introduction card addendum. Exploration record: articles/2026-08-28/big_picture_simplicity_first_en.md.
What this source index covers

The public registry includes 116 document views, 6,693 section records and 985 physical LaTeX sources. Active publication documents, research material and archives retain their recorded roles. This is source navigation, not a new semantic review or proof run.