This study developed a web-based Capstone Project Repository and Similarity Detection System with AI-Assisted Features for the College of Computer Studies (CCS) of Colegio de San Gabriel Arcangel, Inc. The system was designed to provide a centralized platform for managing capstone project records and supporting the evaluation of proposed titles and abstracts. It integrates role-based access for administrators, advisers, and students; repository management; proposal submission; chapter and final-document workflows; defense-result recording; notifications; and AI- assisted functions. For similarity detection, the system uses Term Frequency-Inverse Document Frequency (TF-IDF) and cosine similarity to compare proposed titles and abstracts with existing records. AI-assisted features using Gemini serve as supplementary support and do not independently determine plagiarism, originality, or final academic approval. The system was developed using the Waterfall Software Development Life Cycle and evaluated by 27 CCS student respondents using selected ISO/IEC 9126 software quality characteristics. The results obtained weighted means of 4.34 for Functionality, 4.04 for Reliability, 4.12 for Usability, and 4.20 for Efficiency, with an overall weighted mean of 4.20, interpreted as Very Good. The findings indicate that the developed system provides a functional, reliable, usable, and efficient platform that can support the centralized management of capstone projects and assist advisers and students in reviewing project similarity. Keywords: Capstone Project Repository, Similarity Detection, Artificial Intelligence, TF-IDF, Cosine Similarity, Gemini, Web- Based System, ISO/IEC 9126