Most developers using AI coding assistants have no idea why their prompts sometimes work and sometimes don't. AITutor is a new open-source project that tries to fix that with an interactive terminal tutorial modeled after vimtutor, the classic Vim learning tool.
The tutorial walks through foundational concepts that AI coding tools depend on but rarely explain: context windows (the amount of text an AI model can "see" at once, which determines when it forgets earlier parts of your conversation), tool use, the Model Context Protocol (MCP, a standard for connecting AI assistants to external data sources), hooks, subagents, and prompt structure.
It's built for developers who already use tools like Cursor, Claude Code, or GitHub Copilot daily but treat them as black boxes. The interactive format lets you learn by doing rather than reading documentation, with each lesson building on the previous one.
The project fills a gap that official tool docs mostly ignore. Cursor's documentation tells you what buttons to click. AITutor explains why the AI behind those buttons behaves the way it does - and how to adjust your workflow when it misbehaves. For developers spending hours a day pair-programming with AI, understanding these mechanics is the difference between fighting the tool and actually directing it.