<p><i>Artemisia annua</i> is well known for its antimalarial properties; however, its diverse phytochemicals have also attracted attention for their antiviral potential. In this study, an integrated in silico approach—combining molecular docking, ADMET profiling, and 200-ns molecular dynamics (MD) simulations coupled with MM/GBSA free-energy calculations—was applied to evaluate A. annua–derived compounds against the proteases of HSV-2, MERS-CoV, and SARS-CoV-2. Among the screened metabolites, Chrysosplenol D (Mol007395) was identified as a promising candidate based on a combined assessment of docking performance, favorable drug-likeness (full compliance with Lipinski’s rule of five), and the highest quantitative estimate of drug-likeness (QED = 0.608). While PASS predictions suggested potential antiviral activity, these results were considered supportive and remain computational in nature. MD simulations and MM/GBSA analyses demonstrated that Chrysosplenol D forms a strong and dynamically stable complex with the HSV-2 protease, driven primarily by favorable van der Waals and electrostatic interactions. In contrast, its interaction with the MERS-CoV 3CLpro was characterized by a more flexible yet moderately stable binding mode, whereas the SARS-CoV-2 protease complex exhibited weak and unstable binding behavior. Overall, these findings highlight Chrysosplenol D as a selective and energetically favorable in silico inhibitor of the HSV-2 protease, supporting its potential as a lead scaffold for further experimental validation and structure–activity optimization toward the development of anti-HSV-2 antiviral agents.</p>