recordKernel_eq

In the forced D=3 eight-tick cell (the cube $2^3$), a configuration places one recognition bit on each of eight vertices and packs them as an element of $\mathrm{Fin},256$. The boundary record is the six face-closure parities (one per face). The record kernel is the filter of configurations that share the empty cell's record: the bulk states invisible to the ledger.

The module runs the cell-injection test from the entropy-fork program: flip interior bits and ask whether the boundary record must change. The verdict is a dichotomy. Every single-vertex flip posts (and hits exactly the three incident faces), yet the record map is non-injective. Its blind set is a rank-4 subgroup of sixteen moves, generated by whole-face flips and including the global complement $255$ and the two inscribed tetrahedra $105={0,3,5,6}$ and $150={1,2,4,7}$.

One-line computational proof: decide exhaustively checks that the filter defining the kernel equals the listed sixteen-element Finset. Decidable equality on face records and the finite type $\mathrm{Fin},256$ make the identity a pure decision problem; no algebraic lemmas are applied.

This is the explicit naming of the blind set that the module's rank-nullity story rests on: $|\mathrm{kernel}|=16=2^4$, $|\mathrm{image}|=16=2^4$, and $16\cdot 16=256$. It underwrites the claims that blindness is only global (any two same-record configs differ on at least four vertices) and that a whole-face flip of weight four is invisible from every base. Framework landmarks: T7 (eight-tick octave) and T8 ($D=3$) force the cube substrate on which the test runs. The declaration sits in the holography entropy-fork program that asks whether recognition complementarity stays an axiom or becomes a theorem; used_by is currently empty, so it is a documented leaf for later complementarity arguments.