Answers are grounded in this paper and cite their sources.
Companion paper I, version 1.1 (finite-Fourier model, machine-checked). In a finite-Fourier model of vector fields on T^3 we consider the vector-valued tethered operation {a,b} = [a,b] + T(a,b), where [·,·] is the vector-field Lie bracket and T(a,b) = −(κw)⟨Π_u a, Π_u b⟩u. For every divergence-free axis u along e_x with ⟨u,u⟩ ≠ 0 (in particular u = (sin y, 0, 0)) the Jacobiator vanishes whenever all three arguments are divergence-free and point along e_x. On this class the Lie bracket vanishes; T need not vanish, but it takes values in ℂu, which lies in the class, so every nested term vanishes. Lean proves that the hypotheses are inhabited with a nonzero bracket inside the class (non-vacuity) and that ⟨u,u⟩ ≠ 0 cannot be dropped: for the isotropic axis eⁱʸe_x the Jacobiator on (2cos y, 0, 0) is 6(κw)^2 eⁱʸe_x. This is an elementary, machine-checked boundary case, not a new Jacobi theorem. Companion paper II shows the e_x-direction hypothesis cannot be dropped. What Lean checks. shear_jacobi_holds_of_pr_ne_zero, shearY_jacobi_holds, paperI_nonvacuous, shear_jacobi_fails_isotropic and the supporting lemmas in FTVerified/May/FieldJacobi.lean and FTVerified/Companion/JacobiCompanion.lean. The paper's hypothesis-to-declaration table (README.md) matches every Lean hypothesis to a stated assumption; no mismatch was found. The isotropic guard relies on Lean's convention x/0 = 0, which the paper states. Open gap (not addressed here). No identification of this finite-Fourier operation with the state-dependent geometric functional bracket has been proved, in Lean or on paper. The geometric bracket in BenFrohman/frohmanian-tether-geometric has a state-dependent axis u, a state-dependent projection Π_u and an enstrophy prefactor with κ = 3/2 and no free parameter w; the operation here has a frozen axis, free complex κ and w, a complex-bilinear pairing and Lean's division convention. Neither result is claimed to transport to the geometric setting, and nothing here bears on Navier–Stokes or any Clay problem. Reproducibility. Full Lean source: BenFrohman/frohmanian-tether-verified at a761aed99fba58ed4ac9bd5970577b2fc089824d (tip of main on 11 November 2026; a git archive zip of that commit is attached). The paper cites code commit cf04afd; the Lean sources, scripts, lakefile, manifest and toolchain are byte-identical between that commit and a761aed. Lean leanprover/lean4:v4.30.0-rc1, Mathlib 0c154d67103f74be3a0f2c509f72ccbf5be9f2a7. One-command rebuild from the repository root: lake exe cache get && lake build && python3 scripts/check_jacobi_axioms.py. GitHub Actions on main at a761aed passed (run 42932429909). On 11 November 2026 the pinned commit was rebuilt locally from scratch for the project modules (attached build.log); the attached axiom output (axiom-output.txt: PASS, 49 declarations) shows only propext, Classical.choice and Quot.sound (or a subset), and the attached scan found no sorry, admit, native_decide or axiom declarations. SHA256SUMS lists every attached file. Limits. A kernel check shows that the stated Lean propositions follow from the stated Lean definitions. It does not show that those definitions model any geometric or physical object.
No takes yet. Share an insight, caveat, or question.
Benjamin Frohman (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: