Claude Code: Slash Command Reference Cheat Sheet
Every built-in slash command available in Claude Code, Anthropic's command-line coding tool, organized by what you're trying to do, with plain-language explanations of each one.
Claude Code is Anthropic's agentic coding tool, used from the terminal, an IDE extension, or the desktop app. Inside an active session, slash commands (typed shortcuts starting with /) control the session itself—clearing context, switching models, managing permissions, kicking off a code review, and more. This is different from a normal prompt, which tells Claude what to build or fix.
Type / at the start of a message to open the command menu, or type / followed by a few letters to filter it. A command is only recognized when it appears at the very start of a message; anything typed after the command name is passed to it as an argument.
Tip Not every command below appears for every user. Availability depends on platform, plan, and Claude Code version—for example, /desktop only shows on macOS or Windows with a Claude subscription, and /upgrade only shows on Pro and Max plans. When in doubt, type /help in your own session to see what's actually available to you.
Session & Conversation Management
| Command | What it does |
|---|---|
| /clear [name] | Starts a new conversation with an empty context window while keeping project memory. The old conversation is still reachable through /resume. Aliases: /reset, /new. |
| /compact [instructions] | Frees up context space by summarizing the conversation so far, optionally focused on what to keep, without starting over. |
| /context [all] | Shows a visual breakdown of what is filling the context window, plus suggestions for freeing it up. |
| /resume [session] | Reopens a previous conversation by name or ID or opens a picker of past sessions. Alias: /continue. |
| /branch [name] | Forks the current conversation into a new branch so you can try a different approach without losing the original. |
| /fork <directive> | Hands a side task to a background subagent that inherits the full conversation, so you can keep working in the main thread. |
| /rewind | Rolls the conversation and/or code back to an earlier checkpoint. Aliases: /checkpoint, /undo. |
| /rename [name] | Renames the current session so it is easier to find later in /resume. |
| /export [filename] | Saves the current conversation as a plain-text file. |
| /copy [N] | Copies the latest (or Nth-latest) assistant response to the clipboard. |
| /recap | Produces a quick one-line summary of the current session on demand. |
| /btw <question> | Asks a quick side question without adding it to the main conversation history. |
Model, Effort & Cost Controls
| Command | What it does |
|---|---|
| /model [model] | Switches the active AI model and saves it as the default for new sessions. |
| /effort [level|auto] | Sets how much reasoning depth the model applies, from low up to max. |
| /fast [on|off] | Toggles a lighter, faster response mode. |
| /advisor [model|off] | Turns on a second model that offers guidance at key decision points during a task. |
| /usage | Shows session cost, plan usage limits, and a usage breakdown. Aliases: /cost, /stats. |
| /usage-credits | Configures extra usage credits so work can continue after hitting a plan limit. |
Planning & Code Quality
| Command | What it does |
|---|---|
| /plan [description] | Switches into plan mode, where Claude proposes an approach before touching any files. |
| /code-review [level] [--fix] [--comment] [target] | Reviews the current diff for bugs and cleanup opportunities; can auto-apply fixes or post inline PR comments. |
| /security-review | Scans pending changes for security issues such as injection risks, auth problems, or exposed credentials. |
| /review [PR] | Gives a read-only review of a specified (or auto-detected) GitHub pull request. |
| /simplify [target] | Runs a parallel, cleanup-only review covering reuse, simplification, and efficiency, then applies the fixes. |
| /diff | Opens an interactive viewer of uncommitted changes and per-turn diffs. |
| /verify | Confirms a change works by actually running the app, rather than relying only on tests. |
| /run | Launches and drives the project's app to demonstrate that a change works. |
| /run-skill-generator | Teaches /run and /verify how to build and launch a specific project from a clean environment. |
Background & Parallel Work
| Command | What it does |
|---|---|
| /background [prompt] | Detaches the session to keep running in the background, freeing up the terminal. Alias: /bg. |
| /tasks | Lists everything currently running in the background. Alias: /bashes. |
| /batch <instruction> | Breaks a large codebase change into independent units and runs each one in parallel in its own workspace. |
| /schedule [description] | Sets up a recurring routine that runs on Anthropic-managed infrastructure. Alias: /routines. |
| /workflows | Opens a view to watch, pause, resume, or save background workflow runs. |
| /goal [condition|clear] | Sets a target outcome that Claude keeps working toward across multiple turns. |
| /stop | Stops a background session that is currently attached. |
Project Setup & Memory
| Command | What it does |
|---|---|
| /init | Scans the project and generates a starter CLAUDE.md file. |
| /memory | Edits CLAUDE.md memory files and manages auto-memory entries. |
| /add-dir <path> | Grants file access to an additional working directory for the session. |
| /cd <path> | Moves the session to a different working directory while preserving the conversation. |
| /permissions | Manages which tool actions are allowed, require approval, or are denied. Alias: /allowed-tools. |
| /hooks | Views configured hooks that run automatically on tool events. |
| /agents | Points to creating or managing subagents for specialized tasks. |
| /skills | Lists available skills and lets you toggle their visibility. |
| /reload-skills | Rescans skill and command directories so changes made mid-session become available without restarting. |
| /claude-api [migrate|managed-agents-onboard] | Loads Claude API reference material for the project's language, or helps migrate existing API code to a newer model. |
MCP, Integrations & Plugins
| Command | What it does |
|---|---|
| /mcp | Manages MCP server connections and authentication. |
| /plugin [subcommand] | Manages installed plugins — list, install, enable, or disable. |
| /reload-plugins [--force] | Reloads active plugins to pick up changes without restarting the session. |
| /install-github-app | Walks through installing the Claude GitHub App on a repository. |
| /install-slack-app | Installs the Claude Slack app through an OAuth flow. |
| /web-setup | Connects a GitHub account for use with Claude Code on the web. |
| /chrome | Configures Claude in Chrome settings. |
| /design-sync [hint] | Syncs a repo's React design system to Claude Design so generated designs reuse real components. |
| /design-login | Authorizes design-system access for /design-sync. |
Remote, Web & Cloud Sessions
| Command | What it does |
|---|---|
| /teleport | Pulls a Claude Code on the web session into the local terminal. Alias: /tp. |
| /remote-control | Makes the current session controllable from claude.ai. Alias: /rc. |
| /remote-env | Chooses the default environment used for cloud agents. |
| /desktop | Continues the current session in the Claude Code desktop app. Alias: /app. |
| /mobile | Shows a QR code to download the Claude mobile app. Aliases: /ios, /android. |
| /ultraplan <prompt> | Drafts a plan in a browser-reviewable session, then executes it remotely or sends it back to the terminal. |
| /ultrareview [PR] | Runs a deep, multi-agent code review in a cloud sandbox (alias for /code-review ultra). |
| /autofix-pr [prompt] | Spawns a cloud session that watches a pull request and pushes fixes when CI fails or reviewers comment. |
| /deep-research <question> | Fans out web searches on a question and synthesizes a cited report. |
Account, Settings & Interface
| Command | What it does |
|---|---|
| /login / /logout | Signs in or out of an Anthropic account. |
| /config [key=value] | Opens settings for theme, model, and output style, or sets a value directly. Alias: /settings. |
| /theme | Changes the color theme, including light, dark, and colorblind-accessible options. |
| /color [color|default] | Sets the prompt bar color for the current session. |
| /statusline | Configure a custom status line. |
| /keybindings | Opens the keyboard shortcuts file. |
| /terminal-setup | Configures terminal-specific key bindings, shown only where needed. |
| /scroll-speed | Adjusts mouse-wheel scroll speed interactively. |
| /tui [default|fullscreen] | Sets the terminal UI renderer. |
| /focus | Toggles a condensed view showing only the latest prompt and response. |
| /sandbox | Toggles sandbox mode on supported platforms. |
| /voice [hold|tap|off] | Toggles voice dictation. |
| /privacy-settings | Views and updates privacy settings (Pro and Max plans only). |
| /upgrade | Opens the plan upgrade page. |
| /passes | Shares a free week of Claude Code with a friend, when eligible. |
| /status | Shows version, model, account, and connectivity details. |
Diagnostics & Support
| Command | What it does |
|---|---|
| /doctor | Diagnoses installation and settings issues, with an option to auto-fix them. |
| /debug [description] | Enables debug logging and helps troubleshoot an issue. |
| /heapdump | Writes a memory snapshot for diagnosing high memory usage. |
| /feedback [report] | Submits feedback or reports a bug with session context attached. Aliases: /bug, /share. |
| /fewer-permission-prompts | Scans past sessions and builds an allowlist to reduce repeated permission prompts. |
| /insights | Generates a report on usage patterns and friction points across sessions. |
| /release-notes | Shows the changelog with a version picker. |
| /help | Shows help text and the full command menu. |
| /powerup | Interactive lessons that introduce Claude Code features. |
| /team-onboarding | Generates a teammate onboarding guide from a user's own session history. |
| /stickers | Orders Claude Code stickers. |
| /radio | Opens a lo-fi radio stream in the browser. |
| /exit | Exits the CLI, or detaches from a background session. Alias: /quit. |
Cloud Provider Setup
| Command | What it does |
|---|---|
| /setup-bedrock | Configures Amazon Bedrock authentication, region, and model pins. |
| /setup-vertex | Configures Google Cloud's Agent Platform authentication, project, and region. |
Custom Commands & Skills
Beyond the built-in commands above, you can create your own. Custom commands used to live as markdown files in .claude/commands/, and that format still works, but it's now considered legacy.
The current recommended format is .claude/skills/<name>/SKILL.md. Skills support the same /name invocation as custom commands, but Claude can also invoke them automatically when a task matches the skill's description — no slash needed. As of Claude Code v2.1.199, you can chain up to six skills in one message (e.g. /skill-a /skill-b do the task), and the trailing text is passed to all of them as arguments.
Project-level commands and skills apply only to the repository they live in; personal ones placed under your home directory (~/.claude/) are available across every project you open.
MCP Prompts
Connected MCP servers can expose their own prompts as commands, discovered automatically and named using the pattern /mcp__<server>__<prompt>. For example, a connected GitHub or Slack MCP server might surface commands like /mcp__github__create-issue. These appear alongside built-in commands in the / menu once a server is connected, and disappear if the server is disconnected.
If you need further assistance, Bluehost Chat Support is available 24 hours a day, 7days a week while Bluehost Phone Support is available 7 days a week from 7 am-12 midnight EST.
- Chat Support - While on our website, you should see a CHAT bubble in the bottom right-hand corner of the page. Click anywhere on the bubble to begin a chat session.
- Phone Support -
- US: 888-401-4678
- International: +1 801-765-9400
You may also refer to our Knowledge Base articles to help answer common questions and guide you through various setup, configuration, and troubleshooting steps.
Did you find this article helpful?