AI-driven drug discovery relies on multiscale data integration. However, the veterinary drug development field currently lacks systematic integration of drug-disease-target information. To address this gap, we constructed the Veterinary DrugBank (https://ys.yuhoutech.com/vp/#/home, VDB). VDB integrates multisource authoritative resources (PubChem, VSDB, FDA Green Book, TTD, and Papich Handbook of Veterinary Drugs) to systematically characterize 891 clinical veterinary drugs. This includes their fundamental attributes, structural/physicochemical properties, and 271 unique therapeutic targets, with the drugs demonstrating efficacy across >50 anmial species. The database annotates 956 ATCvet-coded indications and 904 ICD-11-classified diseases, thereby providing a robust foundation for AI-driven drug discovery. Porcine epidemic diarrhea virus (PEDV) causes high mortality in neonatal piglets and substantial economic losses, threatening global swine production. To address the urgent need for anti-PEDV therapeutics, we demonstrate VDB's drug repurposing utility using PEDV as a case study. Given the genomic and target similarities between PEDV and Coronaviridae family members, we conducted a similarity assessment between veterinary drugs and over 400 anticoronavirus compounds. We identified two antiparasitic agents, Eprinomectin (VDB00307) and Selamectin (VDB00726), which exhibited antiviral activity against PEDV. This study demonstrates the potential of drug repurposing in veterinary medicine and provides data support for AI-driven drug design, thereby accelerating the development of novel veterinary drugs.