Mycobacterium tuberculosis uracil-DNA glycosylase (MtUng), a key DNA repair enzyme, represents an attractive target for the design of new antimycobacterial agents. However, only a limited number of weak MtUng inhibitors are reported, primarily based on the uracil ring, and hence, lack diversity. We report the first structure-based virtual screening (SBVS) using three separate libraries consisting of uracil and non-uracil small molecules, together with the FDA-approved drugs. Twenty diverse virtual hits with the highest predicted binding were procured and screened using a fluorescence-based assay to evaluate their potential to inhibit MtUng. Several of these molecules were found to inhibit MtUng activity at low mM and µM levels, comparable to or better than several other reported Ung inhibitors. Thus, these molecules represent a diverse set of scaffolds for developing next-generation MtUng inhibitors. The most active uracil-based compound 5 (IC<sub>50</sub> = 0.14 mM) was found to be ∼ 15-fold more potent than the positive control, uracil. The binding stability and conformation of compound 5 in complex with the enzyme were further confirmed using molecular dynamics simulation.