26 Claude Code Shortcuts You Should Know

I stumbled across a post that made me stop scrolling instantly. A LinkedIn creator compiled all 26 slash commands available in Claude Code, and honestly, I’ve been using Claude Code daily and still didn’t know half of these existed.

If you’ve ever felt like you’re only scratching the surface of what Claude Code can do, this list is about to change that. The original poster laid out every single shortcut command, and I’m going to break each one down so you actually know when and why to use them.

Here’s the key insight: these aren’t prompts. They’re slash commands you type at the very start of your input inside Claude Code. Think of them as power-user controls that sit right under the hood.

Session and Context Management

1. /CLEAR wipes your entire conversation history and resets context. Perfect when you’re switching between unrelated tasks and don’t want leftover context bleeding into your next prompt.
2. /COMPACT [instructions] compresses your conversation while letting you specify exactly what to keep. This is huge when you’re deep into a long session and the context window is getting crowded. You tell it what matters, it trims the rest.
3. /RESUME [session] reopens a previous conversation by ID or name. No more losing that brilliant thread you had going yesterday. Just call it back up.
4. /BRANCH [name] creates a new branch from your current conversation. Want to explore a different direction without losing your original thread? Branch it, just like you would in Git.
5. /REWIND rolls the conversation and any code changes back to an earlier point. Made a wrong turn three messages ago? Rewind and try again.
6. /RENAME [name] gives your current session a custom name. Skip the name and Claude generates one for you. Simple but incredibly useful for staying organized.
7. /EXPORT [filename] exports the full conversation as plain text. Great for documentation, sharing with teammates, or keeping records of important problem-solving sessions.

Model and Usage Controls

1. /MODEL [model] switches between models mid-session: Sonnet, Opus, or Haiku. Need faster responses for a quick task? Switch to Haiku. Need deeper reasoning? Jump to Opus. All without leaving your session.
2. /COST shows token usage for your current session. Handy for keeping an eye on how much context you’re burning through.
3. /USAGE displays your plan’s usage limits and current rate-limit status. Know exactly where you stand before hitting a wall.
4. /EXTRA-USAGE enables extra usage when you’ve hit your standard limits. When you’re in the middle of something critical and can’t afford to stop, this keeps you going.

Project Setup and Memory

1. /INIT initializes your project with a CLAUDE.md guide file. This is one of the first things you should do with any new project. It gives Claude persistent context about your codebase.
2. /MEMORY edits CLAUDE.md memory files and controls auto-memory settings. This is where Claude stores what it learns about your project across sessions. Managing this well is the difference between a helpful assistant and one that truly understands your work.
3. /ADD-DIR <path> adds a working directory for file access during the session. Working across multiple folders? Give Claude access to all of them.

Code Review and Security

1. /DIFF opens a diff viewer for uncommitted changes and per-turn diffs. See exactly what Claude changed, line by line, before you commit anything.
2. /SECURITY-REVIEW checks pending changes for security issues. Running this before committing is a habit worth building. It catches things you might miss during a fast iteration cycle.
3. /PLAN [description] enters plan mode, optionally starting with a specific task description. Instead of jumping straight into code, Claude maps out an approach first. This is how the expert recommends tackling complex multi-step tasks.

Permissions, Plugins, and Configuration

1. /PERMISSIONS manages allow, ask, and deny rules for tool access. Control exactly what Claude can and can’t do. Want it to read files but not write them? Set it here.
2. /AGENTS manages agent and sub-agent configurations. If you’re running multi-agent workflows, this is your command center.
3. /SKILLS lists all available skills, both built-in and custom. Think of skills as specialized capabilities you can bolt onto Claude Code. This command shows you everything available.
4. /PLUGIN manages Claude Code plugins for extending functionality.
5. /RELOAD-PLUGINS reloads active plugins without restarting your session. Made changes to a plugin? Reload it live.
6. /MCP manages MCP server connections and OAuth authentication. MCP (Model Context Protocol) lets Claude Code connect to external tools and services. This command handles those connections.

Look and Feel

1. /CONFIG opens settings for theme, model, output style, and general preferences. Your one-stop shop for customizing how Claude Code behaves.
2. /THEME changes the color theme, including light, dark, and accessible options. Work at night? Switch to dark. Need higher contrast? Accessible themes have you covered.
3. /COLOR [color] sets the prompt bar color for your current session. A small touch, but when you’re running multiple sessions, color-coding them is a surprisingly effective way to keep track.

The real value here isn’t memorizing all 26 commands. It’s knowing they exist so you can reach for the right one when you need it. Bookmark this list, and next time you’re deep in a Claude Code session, try one you haven’t used before. You’ll be surprised how much faster your workflow gets.

For the full original breakdown, check out the LinkedIn post linked below.