Purpose

Explain one thing, once, in genuinely plain words. The object can be a concept, a piece of code, an error, an architecture, or the assistant's own previous output. The core move is an altitude drop: land the explanation at the lowest useful altitude first — a concrete analogy, zero jargon — then layer altitude back up only when the user asks for it. Michael Scott's "Why don't you explain this to me like I'm five" (The Office US, S5E9 "The Surplus") is the tone anchor; the Feynman technique is the method.

Use when: the user signals they didn't follow something — I don't get it, explain simply, ELI5, what does this actually mean, rephrase that. Skip when: the user wants ongoing, multi-session coaching with persistent state (hand off to /education:teach); a plain inline answer already lands (just answer).

This skill auto-invokes (no disable-model-invocation) because "I don't get it" should reach it without the user naming a command. Auto-trigger is best-effort; /education:explain is the guaranteed path.