The ever-accelerating growth in scientific literature presents a formidable challenge for researchers and students aiming to stay abreast of the most recent findings. Previous solutions, which include content-based, collaborative-based, and graph-based filtering recommendation systems, have their own limitations, primarily their inability to efficiently manage time-consuming search engine queries, a frequent issue for students. To address these constraints, we introduce a novel tool-a multi-agent system with an intelligent filtering mechanism. This system automates the literature search and filtering process, generating search queries independently and conducting comprehensive online searches. The system comprises autonomous agents that collectively gather and analyze data from a myriad of sources. Utilizing sophisticated techniques, the intelligent filtering mechanism leverages user preferences, interests, and contextual information. Continuous learning from user feedback allows the system to iteratively refine its recommendations, providing a personalized user experience. A user-friendly interface has been developed to streamline the configuration of the search procedure, offering users an easy way to fine-tune their preferences. Evaluations indicate that our approach delivers superior performance, significantly improving the process of scientific literature recommendation. Our tool is designed to assist researchers and students by minimizing the manual effort required in literature search and filtering, thereby ensuring efficient access to pertinent information. By automating these labor-intensive tasks, our tool enables users to keep pace with the latest scientific discoveries with increased ease.