Debugging is a crucial skill in programming education and software development, yet it is often overlooked in CS curricula. To address this, we introduce an AI-powered debugging assistant integrated directly inside an IDE. The tool provides real-time debugging support by analyzing code, suggesting tailored breakpoints, and offering contextual hints. It combines RAG with LLMs to generate solutions and uses program slicing and custom heuristics for breakpoint recommendation, ensuring efficiency by reducing LLM calls and enhancing accuracy. A three-level evaluation, including technical analysis, a UX study, and classroom tests, showed the tool’s potential as an effective debugging assistant for programming education.