Objective: Medical libraries face growing demands related to digital transformation in healthcare and to teaching hospital accreditation requirements. However, traditional manual consultation services are often constrained by limited staffing and service hours and struggle to meet the 24/7 demands of clinical settings. This study developed an intelligent consultation system based on generative artificial intelligence (AI) to optimize service performance in medical libraries.Methods: This study was conducted in a library setting at Kaohsiung Medical University Gangshan Hospital, Kaohsiung, Taiwan. A cross-functional team identified four primary service gaps: availability, response speed, workload, and technical application. By using Claude and retrieval-augmented generation within the Claude Artifacts low-code environment, the team integrated institutional regulations, electronic resources, and academic integrity guidelines to develop LibraryChatbot. This responsive system features a dual-module interface that combines a “topic menu” with “natural language dialogue” and includes builtin error prevention and human handover mechanisms to mitigate AI hallucinations.Results: A 3-month postlaunch evaluation revealed that the system successfully handled approximately 420 consultations per month. It reduced the number of simple consultations handled manually from 500 to 80 cases per month (an 84% reduction), which allowed librarians to focus on advanced academic inquiries. It also increased service availability from 47.5 hours per week to 24/7 coverage (168 hours per week). Testing confirmed the system’s high accuracy in factual consistency, logical reasoning, and resource integration.Conclusion: This study constructed a cost-effective and replicable AI model for medical libraries that effectively frees librarians’ time and ensures 24/7 service availability. Expanding system functions through the application programming interface (API) is expected to further strengthen these digital capabilities and provide a practical reference for the digital transformation of other administrative units.