The stimulator of interferon genes (STING) is a key signaling adaptor in the cGAS-STING pathway of the innate immune system, playing a significant role in autoimmune diseases, viral infections, and cancer, thus representing a promising target for small-molecule inhibitor therapies. This study presents an integrated multi-dimensional computer-aided drug design (CADD) approach that utilizes machine learning (ML), molecular docking, molecular dynamics (MD) simulations, and ADMET prediction to efficiently discover new STING inhibitors from natural products. We developed a precise ML-based STING classification model with 98.5% accuracy and a robust STING inhibitor activity regression model demonstrating strong predictive capabilities, evidenced by an R² of 0.826, MAE of 0.357, and RMSE of 0.452. Virtual screening across multiple traditional Chinese medicine (TCM) compound libraries (Tao Shu L6810, TCMIO, TCMBank, and HERB) yielded 1,596 compounds with predicted pIC50 ≥ 7.00. After rigorous multi-step screening, seven compounds were selected for ADMET evaluation and experimental validation. Notably, two natural compounds, Cassiaside and Plantaginin, showed STING pathway-suppressive activity in THP-1-derived macrophages, and MD simulations further validated their stable binding to the STING protein. Collectively, this study provides a robust and accurate ML-driven strategy for STING inhibitor discovery and identifies two promising TCM-derived lead compounds, offering valuable structural scaffolds for the rational design of STING-targeted therapeutics against immune and inflammatory diseases.