ESD Framework: Cosmological Benchmarks A zero-parameter audit across 66 reproduction studie

We present the cumulative observational audit of the Energy–Space–Displacement (ESD)framework defined in Higginson (2026), Gravity, Electromagnetism, and the Dark Sector from aSingle Displacement Action with Zero Free Parameters (Zenodo DOI: 10.5281/zenodo.19283596;hereafter the foundational ESD paper ), and consolidated in Higginson (2026), The Energy–Space–Displacement Framework: A Relational Admissibility Net for Cross-Sector Physics (Zen-odo DOI: 10.5281/zenodo.20059909; hereafter the ESD Framework synthesis). A single covariantaction with five φ-locked closure constants assembles one screening kernel R(u) that carries thesame grammar from galactic dynamics to cosmological background expansion. With G and H0fixed and no per-sector tuning, the framework is confronted with 66 published datasets acrossseven physical regimes: galactic dynamics (12 studies), solar-system and lab gravity (6), gravi-tational waves and multi-messenger (11), clusters, halos, and compact objects (8), cosmologicalbackground (9), CMB / LSS / linear regime (14), and empirical anomalies (6). Of the 66 studies,64 land at PASS against their published gates, one is a partial closure (G05, cosmic-void lensing,where the kernel-level prediction passes but the analytic void profile saturates), and one is anopen falsifier under active tension (G02, CMB cosmic birefringence, where ESD predicts β = 0exactly against a 3.6σ Planck PR4+WMAP signal). For each study we report the description,the locked prediction, the headline results table, the audit figure, and a one-line conclusion.The companion replication package (esd-cosmology) (Zenodo DOI: 10.5281/zenodo.20444359)regenerates every table and figure from the locked constants. This paper is the empirical faceof the framework: no fits, no tuned amplitudes, no per-study parameters.