The Aurora kinases (serine/threonine kinases) were discovered in 1995 during studies of mutant alleles associated with abnormal spindle pole formation in Drosophila melanogaster. They soon became the focus of much attention because of their importance in human biology and association with cancer. Aurora kinases are essential for cell division and are primarily active during mitosis. Following their identification as potential targets for cancer chemotherapy, many Aurora kinase inhibitors have been discovered, and are currently under development. The binding modes of Aurora kinase inhibitors to Aurora kinases share specific hydrogen bonds between the inhibitor core and the back bone of the kinase hinge region, while others parts of the molecules may point to different parts of the active site via noncovalent interactions. Currently there are about 30 Aurora kinase inhibitors in different stages of pre-clinical and clinical development. This review summarizes the characteristics and status of Aurora kinase inhibitors in preclinical, Phase I, and Phase II clinical studies, with particular emphasis on the mechanisms of action and resistance to these promising anticancer agents. We also discuss the validity of Aurora kinases as oncology targets, on/off-target toxicities, and other important aspects of overall clinical performance and future of Aurora kinase inhibitors.