Registered Before the Fact: The Exact Re-Grade Every LARES-2 Outcome Owes This Frame

This paper is a registration, written in the desk's seal-before-the-fact register: dated, unconditional, and constructed so that no hedging after the fact is possible. Paper A (#756) priced the existing frame-dragging bound: LAGEOS / LAGEOS 2 / LARES satellite laser ranging measures μ = 0.9910 ± 0.02 (systematic) of the general-relativistic value (Ciufolini et al. 2019), which in Einstein–Cartan theory proper prices the torsion contribution at **exactly zero** — torsion is algebraic, sourced by intrinsic spin density, and absent in the vacuum a satellite flies through — and bounds every *propagating or anomalous* gravitomagnetic addition at ε = μ − 1 = −0.009 ± 0.020, the interval [−0.029, +0.011], |ΔΩ̇| ≲ 1.0 mas/yr on the ~50.16 mas/yr combined observable. LARES-2, in orbit since 13 July 2022 on a supplementary inclination to LAGEOS, is built to tighten that bound. A sub-percent LARES-2 figure circulates in secondary coverage; it is **not verified against a primary paper, and it is not used anywhere in this registration** — not as a premise, not as an expectation, not as a target. What this paper does instead is compute, now, the exact re-grade the house frame will owe for each candidate *verified* outcome: confirmation of general relativity at 1%, confirmation at 0.1%, a confirmation at the present central value with a tightened error, and deviations of stated size in either direction. The re-grade is stated as a mechanical rule (§3) plus worked cases with the arithmetic shown (§4), followed by the tighten / die / untouched ledger (§5) and, in the last section, what would falsify, void, or break this registration itself (§7). **Untouched, at full size: most of this corpus.** LARES-2 measures one number in one channel around one planet; the cosmological body of the frame never routed through it, and no outcome registered here reaches it.