The Atom Search Optimization (ASO) algorithm is a recent advancement in metaheuristic optimization inspired by principles of molecular dynamics. It mathematically models and simulates the natural behavior of atoms, with interactions governed by forces derived from the Lennard-Jones potential and constraint forces based on bond-length potentials. Since its inception in 2019, it has been successfully applied to various challenges across diverse fields in technology and science. Despite its notable achievements and the rapidly growing body of literature on ASO in the metaheuristic optimization domain, a comprehensive study evaluating the success of its various implementations is still lacking. To address this gap, this article provides a thorough review of half a decade of advancements in ASO research, synthesizing a wide range of studies to highlight key ASO variants, their foundational principles, and significant achievements. It examines diverse applications, including single- and multi-objective optimization problems, and introduces a well-structured taxonomy to guide future exploration in ASO-related research. The reviewed literature reveals that several variants of the ASO algorithm, including modifications, hybridizations, and multi-objective implementations, have been developed to tackle complex optimization problems. Moreover, ASO has been effectively applied across various domains, such as engineering, healthcare and medical applications, Internet of Things and communication, clustering and data mining, environmental modeling, and security, with engineering emerging as the most prevalent application area. By addressing the common challenges researchers face in selecting appropriate algorithms for real-world problems, this study provides valuable insights into the practical applications of ASO and offers guidance for designing ASO variants tailored to specific optimization problems.