Approximately 3.9 billion individuals are vulnerable to dengue infection, a prevalent cause of tropical diseases worldwide. Currently, no drugs are available for preventing or treating Flavivirus diseases, including Dengue, West Nile, and the more recent Zika virus. The highly conserved Flavivirus NS2B-NS3 protease, crucial for viral replication, is a promising therapeutic target. This study employed in-silico methodologies to identify novel and potentially effective anti-dengue small molecules. A pharmacophore model was constructed using an experimentally validated NS2B-NS3 inhibitor, with the Gunner Henry score confirming the model's validity. The Natural Product Activity and Species Source (NPASS) database was screened using the validated pharmacophore model, yielding a total of 60 hits against the NS2B-NS3 protease. Furthermore, the docking finding reveals that our newly identified compounds from the NPASS database have enhanced binding affinities and established significant interactions with allosteric residues of the target protein. MD simulation and post-MD analysis further validated this finding. The free binding energy was computed in terms of MM-GBSA analysis, with the total binding energy for compound 1 (−57.3 ± 2.8 and − 52.9 ± 1.9 replica 1 and 2) indicating a stronger binding affinity for the target protein. Overall, this computational study identified these compounds as potential hit molecules, and these findings can open up a new avenue to explore and develop inhibitors against Dengue virus infection.