Background. Prostate cancer (PCa) is one of the common malignant tumors of the urological system, and metastasis often occurs in advanced stages. Chemotherapy is an effective treatment for advanced PCa but has limitations in terms of efficacy, side effects, multidrug resistance, and high treatment costs. Therefore, new treatment modalities for PCa need to be explored and improved. Methods. R language and GEO database were used to obtain differentially expressed genes for PCa single-cell sequencing. TCMSP, STITCH, SwissTargetPrediction, and PubChem databases were used to obtain the active ingredients and targets of Pueraria lobata (PL). Next, Cytoscape software was used to draw the interactive network diagram of “drug–active component–target pathway.” Based on the STRING database, the protein–protein interaction network was constructed. Gene Ontology and the Kyoto Encyclopedia of Genes and Genomes were applied for the genes. Molecular docking was used to visualize the drug–target interaction via AutoDock Vina and PyMOL. Finally, prognosis-related genes were found by survival analysis, and Protein Atlas was used for validation. Results. Four active components and 31 target genes were obtained through the regulatory network of PL. Functional enrichment analysis showed that PL played a pharmacological role in the treatment of PCa by regulating the metabolic processes of reactive oxygen species, response to steroid hormones, and oxidative stress as well as IL-17 signaling pathway, PCa, and estrogen signaling pathway. Single-cell data showed that AR, MIF, HSP90B1, and MAOA genes were highly expressed, and molecular docking analysis showed that representative components had a strong affinity with receptor proteins. Survival analysis found that APOE, CA2, IGFBP3, MIF, F10, and NR3C1 could predict progression-free survival (PFS), and some of them could be validated in PCa. Conclusion. In this paper, a drug–active ingredient–target pathway network of PL at the single-cell level of PCa was constructed, and the findings revealed that it acted on genes such as AR, MIF, HSP90B1, and MAOA to regulate several biological processes and related signaling pathways to interfere with the occurrence and development of PCa. APOE, CA2, IGFBP3, MIF, F10, and NR3C1 were also important as target genes in predicting PFS.