Output styles change how Claude responds — not just the content. Use the built-in Explanatory, Concise, and Technical styles, or define your own, so every answer arrives in the shape that actually helps you.
Most people only ever change what they ask Claude. But you can also change how it answers. Output styles are a setting that shapes the format and posture of every response — terse or thorough, plain or technical, teaching or just doing.
It's the difference between tuning the message and tuning the messenger. Both matter.
Claude ships with three: Explanatory gives detailed, step-by-step answers that teach as they go. Concise keeps it brief and action-first. Technical is precise and lets the jargon through for when you want signal density.
You switch with a quick command and pick the voice the moment needs — Concise when you're moving fast, Explanatory when you're learning.
/output-style explanatory # teach as you go /output-style concise # brief, action-first /output-style technical # precise, jargon ok
Here's the compounding move: when you're new to something, run Explanatory and ask Claude to teach the why, not just hand you the fix. You read the reasoning, understand the move, and carry it forward — instead of pasting an answer you couldn't reproduce.
The goal isn't to be finished. It's to be more capable than you were this morning. Borrow the understanding; don't outsource it.
You can define a custom style once and reuse it forever: 'plain English, no jargon, lead with the action, one concrete example.' From then on, every answer arrives in your format without you re-asking for it.
Set the standard once and hold it every time. That consistency is its own kind of discipline.
Three copy-paste output styles — Plain Teacher, Fast Operator, and Officer Brief — plus a template to build your own.