This study explores the identification of key components and indicators for designing effective prompts used by Large Language Models (LLMs) in Software Engineering (SE) learning. In contrast to other learning domain, SE Learning integrates systematic thinking, architectural reasoning and iterative development cycles. This characteristic of SE where requires engineered prompts that can bridge transition between abstract understanding and practical implementation to ensuring students can navigate analytical inquiry and constructive practice. In recent years, SE education increasingly incorporates generative AI, it is essential to tailor prompts in ways that align with SE specific objectives, such as problem solving, theory application, and software design. The research investigates the structural components of prompt design to assess how well they support self-directed learning in SE and facilitate AI interactions that enhance student engagement and understanding. Through A Systematic Literature Review (SLR) and Content Analysis, the study synthesizes existing research to identify effective practices and indicators for prompt engineering. The findings emphasize the need for clear, contextual, and iterative prompt structures that ensure AI outputs remain relevant and aligned with SE curriculum objectives. This paper contributes to the field by providing a framework for creating more effective prompts that drive deeper learning outcomes in Software Engineering education using LLMs. © 2026 Copyright held by the owner/author(s)