Bayesian Atmospheric Radiative Transfer (BART, Harrington et al. 2020, Cubillos et al. 2020, Blecic et al. 2020) is an open-source, reproducible-research (RR) code for atmospheric composition and structure retrieval. Its Bayesian sampler (MCCubed, Cubillos et al. 2017) proposes atmospheric models and compares them to data via a line-by-line radiative-transfer (RT) code. Auxiliary codes initialize to thermochemical equilibrium (TEA, Blecic et al. 2016), produce plots and diagnostics, calculate contribution functions, etc. The BARTTest module checks validity against known-correct calculations or community consensus results, depending on the test. BART has modes for eclipse, transit, and isolated objects (such as imaged exoplanets, solar-system atmospheres, or brown dwarfs). We developed a criterion for sampling sufficiency that applies to any Bayesian posterior distribution. RR is a research methodology that accelerates the pace of science by publishing the codes, settings, and data used to support a paper’s scientific claims. Researchers can then answer their own questions about a paper’s methodology details and can verify the work. RR enables fast resolution of differences between groups’ work that otherwise can take years to resolve. BART’s license requires that users follow RR practice in reviewed publications and BART produces the required compendium material. This research was supported by the NASA Fellowship Activity under NASA Grant 80NSSC20K0682 and NASA Exoplanets Research Program grant NNX17AB62G.