Everyone Writes About AI Generating Code. Nobody Writes About AI Testing It.

I play Magic the Gathering (MTG) Arena while background agents file bug reports. I expected AI-assisted development to help me write code faster. It did. The part I didn't expect was how much it changed testing and debugging. Manasight is a desktop overlay for MTG Arena, the free-to-play digital version of Magic: The Gathering. It's a Tauri app with a Rust backend, TypeScript frontend, and a companion Astro website. I'm one developer. The project has roughly 70,000 lines of code across four repositories, with over 2,400 tests. Every line was written with Claude Code. Every line was tested with it too. This post is about the testing side. Claude Code, slash commands, and agents If you haven't used Claude Code : it's an AI coding assistant that runs in your terminal. You give it access to your codebase, and it can read files, write code, run shell commands, and use tools, all within a conversation. Two Claude Code features matter for this post. Slash commands are reusable prompt scripts.