We present RF Quantum SCYTHE, a modular
framework for reproducible passive RF geolocation research. Unlike siloed studies that focus on individual algorithms, SCYTHE
provides a comprehensive suite of interoperable demonstrations
covering trajectory recovery, sensor fusion, adaptive denoising, reinforcement learning, and hybrid triangulation. Each
module produces standardized JSON summaries and LaTeXready figures/tables, enabling direct integration into publications
and downstream analysis pipelines. Together, these components
form a systems-level testbed for evaluating geolocation algorithms across diverse modalities, noise conditions, and decision
policies. We demonstrate four core modules—AoA sequence
recovery, AoA+TDoA fusion, policy-driven denoising, and hybrid triangulation—achieving 25-45% error reductions in multisensor tracking and up to 91.6% RMSE improvements in triangulation accuracy. The framework lowers barriers to entry for
reproducible RF research and provides a standardized baseline
for future extensions, including machine-learned policies and
multi-emitter geolocation scenarios.