Slash command

A slash command is a named shortcut you type as /name that Claude Code expands into a longer, predefined instruction. Instead of retyping the same multi-step prompt every time you deploy, review code, or scaffold a new feature, you capture that prompt once as a slash command and trigger the whole thing with a few keystrokes. It is the difference between re-explaining your deployment checklist on every release and just typing /deploy.

Slash commands live as plain Markdown files in your project's .claude/commands directory (or in your home directory for ones you want everywhere). The filename becomes the command name, and the file's contents become the instruction that gets injected when you invoke it. Because they are just files in your repo, they are version-controlled, reviewable in pull requests, and shared automatically with everyone who clones the project - so your whole team runs the exact same playbook instead of each person improvising.

The most useful slash commands encode workflows that are repeated and easy to get subtly wrong: a release checklist, a code-review pass that checks your specific conventions, a 'write tests the way we write tests' routine, or a project-specific scaffolding step. A good rule of thumb: the third time you find yourself typing roughly the same prompt, turn it into a slash command. Combined with CLAUDE.md for standing context and skills for reusable know-how, slash commands are how a team turns its tribal knowledge into something Claude Code executes consistently.