Direct experimentation with chemical warfare agents (CWA) to study important problems such as their permeation across protective barrier materials, decontamination of equipment and facilities, or the environmental transport and fate of CWAs is not feasible because of the obvious toxicity of the CM/As and associated restrictions on their laboratory use. The common practice is to use "simulants," namely, analogous chemicals that closely resemble the CWAs but are less toxic, with the expectation that the results attained for simulants can be correlated to how the CWAs would perform. Simulants have been traditionally chosen by experts, by means of intuition, using similarity in one or more physical properties (such as vapor pressure or aqueous solubility) or in the molecular structural features (such as functional groups) between the stimulant and the CWA. This work is designed to automate the simulant identification process backed by quantitative metrics, by means of chemical similarity search software routinely used in pharmaceutical drug discovery. The question addressed here is: By the metrics of such software, how similar are traditional simulants to CWAs? That is, what is the numerical "distance" between each CWA and its customary simulants in the quantitative space of molecular descriptors? The answers show promise for finding close but less toxic simulants for the ever-increasing numbers of CWAs objectively and fast. Published by Elsevier B.V.