The growing potential of generative AI in software development has inspired research on how to integrate a library information system with an AI model. The primary goal is to recognize how intelligent support for personalized book recommendations, ratings, and summaries; can enhance traditional library functionalities. Therefore, this research focuses on user experience and service quality as a measure of success. Unlike other approaches that involve training custom models, the approach proposed here uses prompt engineering to guide pre-trained AI responses, reducing development complexity and resource usage. Key features include structured forms that direct user input into targeted API prompts, ensuring relevant, domain- specific output. This method proved to be effective in limiting off-topic responses and improving the user experience without the need for model fine-tuning or custom datasets. Additional functionalities, such as spatial localization, extend the usefulness of the system. The results show that integrating generative AI into information systems can significantly improve interaction quality and service personalization. The approach offers a scalable and cost-effective solution for future smart library applications, demonstrating the transformative role of AI in traditional system architectures.