V89: Chemical-Sector Window Shift Test — Window-Accounting Closure as a Specification Layer for V85–V87

We propose the Chemical-Sector Window Shift Test (V89) as a minimal, controlled probe to specify what an "observation window" means in the YAGC/V85–V87 framework. Chemical reactions exhibit extremely small mass deficit per event (Δm/m ≪ 1) yet occur with large multiplicity and primarily manifest through radiation, dissipation, and history-like records rather than mass density. This makes the chemical sector a uniquely sensitive tool for exposing projection residuals—components that appear "dark" when the accounting is performed in an incomplete window. We deliberately inject a chemical sector into the V85–V87 consistency structure in two ways: (A) naive addition without window transfer and (B) an accounting insertion enforcing a cross-window closure condition. Insertion A breaks energy accounting and produces structured residuals correlated with chemical activity, revealing a dominant failure mode: FM-A (insufficient window accounting). Insertion B restores closure and eliminates the residual–chemical correlation, while two remaining issues are promoted from "failures" to specifications: (CT3) chemical contributions must not be injected directly into mass density due to degeneracy with dark-matter-like terms, and (CT5) Σ-history accumulation requires normalization to prevent scale-law breakdown. We conclude that the chemical sector is not a dark-matter candidate, but a diagnostic that specifies observation windows and stabilizes the V85–V87 bridge when window accounting is explicitly enforced. Crucially, V85–V87 did not "fail"—what failed was the implicit assumption that a single window (mass) suffices.