Setting up Claude Code

I've recently been utilising and customising Claude Code more - rather than just sticking with what comes out of the box. With more and more of my colleagues and friends adopting AI (and, specifically, Claude), I thought I would share my go-to config when setting it all up. TERMINAL Claude Code on the terminal is where I spend most of my time - I find it is more powerful than the Code element of the desktop app. Once Claude is installed, these are the settings I reach for: * /model: Sonnet - I've found Sonnet to be a good reliable workhorse without burning all the tokens * /effort: High - Effort was explained to me like "motivation" - High is a good, solid balance * /advisor: Opus - This is a killer feature - it does a lot of the thinking in Sonnet and then checks with Opus * /config - Some settings I change from the default * Default permission mode: Plan - Means I don't have to remember to switch, sometimes the plan gives you a solution without needing to burn more tokens…