Cybersecurity concepts are cumulative, yet conventional learning platforms commonly report only aggregate scores and provide uniform remediation. This paper proposes a rule-based cybersecurity mastery learning platform that detects concept-level knowledge gaps and uses OpenAI-assisted generation to deliver curriculum-aligned remedial support. Assessment items are mapped to learning outcomes and knowledge components through a validated Table of Specifications and Q-matrix. A transparent rule engine computes module and component mastery, applies an initially proposed 80% criterion, ranks detected gaps, and controls progression. OpenAI is limited to generating explanations, examples, hints, guided exercises, and self-check items from faculty-approved evidence; it does not determine grades or progression. The planned descriptive-developmental study uses incremental system development, expert validation, pilot deployment, pretest remediation-reassessment measures, an AI-response rubric, and an adapted ISO/IEC 25010 quality instrument. The proposed architecture aims to make remediation personalized, auditable, and consistent while preserving instructor oversight, data privacy, cybersecurity safety, and academic integrity.

Keywords—adaptive learning, cybersecurity education, generative artificial intelligence, knowledge-gap detection, mastery learning, rule-based systems