Section 1
The 5 From the Video
These are the ones I called out. They're worth knowing well before you go hunting for the rest.
/voice Toggles voice dictation. Hold to talk, release to send. Claude transcribes and responds like normal — you just never touch the keyboard.
Pro tip: Use this when you're thinking out loud. Speaking is 3x faster than typing a vague prompt, and the unfiltered version is usually better anyway. Requires a Claude.ai account.
/plan Puts Claude into read-only plan mode. It can explore your codebase and think through the problem — but it can't write or edit anything until you approve.
Pro tip: Run this before any significant change. It stops Claude from charging ahead and breaking something before you've agreed on the approach. You can describe the task inline: /plan refactor the auth module.
/insights Generates a report on your recent Claude Code sessions — what areas you work in most, where you hit friction, and patterns in how you use the tool.
Pro tip: Run this weekly. It's a fast way to notice which parts of your workflow are still manual when they don't need to be.
/statusline Configures what information shows in your terminal status bar. Run it without arguments and Claude auto-configures it from your shell setup.
Pro tip: Describe what you want to see — current branch, token count, model name — and Claude sets it up for you. No config file editing required.
/rewind Rolls back the conversation and your code to a previous point. Pick any message in the thread and jump back to exactly how things were at that moment.
Pro tip: Use this the moment something breaks and you're not sure what caused it. It's faster than git reset and undoes the conversation context too.
Section 2
Session Management
These commands control your working session — context, history, and how you move between projects.
/clear Wipe the conversation and start completely fresh
Use this between unrelated tasks. Context carries over whether you want it to or not — /clear makes sure it doesn't.
/compact Compress the conversation to free up context without losing the thread
When you're deep in a long session and things start getting slow, this summarizes what's happened so you can keep going without starting over. Pass instructions to tell Claude what to preserve.
/branch Fork the conversation and try a different approach
Creates a copy of your session at the current point. Explore a risky idea in the branch without touching your original. If it works, great. If not, you didn't lose anything.
/resume Pick up a previous session by name or open the session picker
Useful when you're juggling multiple projects. Each session keeps its full context — you don't have to re-explain anything.
/rename Give the current session a name you'll actually remember
Claude auto-generates session names by default. They're not great. Name your sessions something real so /resume is actually useful.
/export Save the full conversation as a plain text file
Useful for documenting decisions, sharing session output, or archiving a research thread. Pass a filename to write directly — no dialog.
Section 3
Configuration
Personalize how Claude Code looks and behaves. Most of these persist across sessions once set.
/config Open the full settings panel
Theme, model, output style, everything in one place. Also accessible as /settings.
/theme Switch between color themes
Light, dark, colorblind-accessible (daltonized), and ANSI themes that inherit from your terminal colors. Worth trying ANSI if you've customized your terminal.
/color Set the prompt bar color for the current session
Options: red, blue, green, yellow, purple, orange, pink, cyan. Use different colors for different projects — it's the fastest way to know at a glance which session you're in.
/vim Toggle Vim editing mode
If you live in Vim keybindings, this makes the prompt feel like home. Toggle off just as easily if you don't.
/keybindings Open your keybindings file to customize shortcuts
Creates the file if it doesn't exist. Add chord bindings, change submit keys, remap whatever feels wrong.
Section 4
Model and Performance
Control which model you're using and how hard it's thinking. These have a real effect on speed, quality, and cost.
/model Switch Claude model mid-conversation
Use arrow keys to adjust. Takes effect immediately — no restart. Switch to Sonnet for quick tasks, Opus when the problem actually needs it.
/effort Set reasoning depth: low, medium, high, max, or auto
Low/medium/high persist across sessions. Max is session-only and requires Opus 4.6 — use it when you want Claude to really think through a hard problem. Auto resets to default.
/fast Toggle fast mode on or off
Optimizes for speed. Good for quick iterations where you don't need deep reasoning. Turn it off when accuracy matters more than pace.
Section 5
Context and Cost
Know what you're spending before you find out the hard way.
/context See exactly how full your context window is
Renders as a colored grid. Green means you're fine. Yellow means start thinking about /compact. Red means do something now. Also surfaces optimization suggestions.
/cost Token usage breakdown and running cost tracker
Shows input, output, and cache tokens with subscription-specific details. If you're on an API plan, check this before long sessions.
/usage Rate limit status and plan limits
Check this before starting a big project. Easier than hitting a wall 40 minutes in.
Section 6
Code and Dev Workflow
Commands for reviewing changes, catching problems, and setting up projects correctly from the start.
/diff Interactive diff viewer — git changes and per-turn changes side by side
Use arrow keys to flip between git diff and individual Claude turns. Good for reviewing exactly what changed in each step, not just the final result.
/security-review Scan pending changes for injection, auth, and data exposure issues
Runs against your git diff. Good habit to run this before committing anything that touches auth, user data, or external APIs.
/pr-comments Pull GitHub PR review comments into your session
Auto-detects the PR for your current branch, or pass a URL/number directly. Lets Claude see what reviewers said and address it inline. Requires the gh CLI.
/init Set up a new project with a CLAUDE.md guide
First thing to run in a new project. CLAUDE.md is how you give Claude persistent context — what this project is, how to navigate it, what not to touch. Run it once, benefit every session.
/memory Edit CLAUDE.md memory files directly
Opens your memory configuration. Manage what Claude remembers, enable or disable auto-memory, and view what's been saved across sessions.
Quick Reference
All Commands at a Glance
Bookmark this. Every command in one place.
The 5 From the Video
/voice /plan /insights /statusline /rewind Session Management
/clear /compact /branch /resume /rename /export Configuration
/config /theme /color /vim /keybindings Model and Performance
/model /effort /fast Context and Cost
/context /cost /usage Code and Dev Workflow
/diff /security-review /pr-comments /init /memory Other Useful Commands
/mcp /doctor /permissions /copy /btw /stats /help Zero to Automated
Get the full AI operating system
Guides are a great start. But inside Zero to Automated, you get a done-for-you AI system deployed within 48 hours, plus everything you need to keep building:
- ✓Claude Code 101 — 8 lessons, zero to automating real work by Sunday
- ✓Ready-to-install Claude Skills — a growing library with 2+ new skills every week
- ✓Done-for-you workflows — copy, paste, run. New ones every week.
- ✓Weekly live builds — watch real projects get built, ask questions, follow along
- ✓Small group coaching — direct access to both founders
$50/mo. Price locked for life at whatever you join at.