The Claude Code Command Reference

Session & context

Start, resume, branch, and control what Claude is holding in mind.

/add-dir<path> Add a new working directory
/autocompact[auto|<tokens>] Configure the auto-compact window sizegated
/branch[name] Create a branch of the current conversation at this point
/brief Toggle brief-only mode
/btw[question] Ask a quick side question without interrupting the main conversation
/cd<path> Move this session to a new working directory
/clear[name] Start a new session with empty context; previous session stays on disk (resumable with /resume)also /reset, /new
/compact<optional custom summarization instructions> Free up context by summarizing the conversation so far
/context Show current context usagegated
/copy Copy Claude's last response to clipboard (or /copy N for the Nth-latest)
/export[filename] Export the current conversation to a file or clipboard
/focus Toggle focus view: just your prompt, summary, and response
/fork<directive> Spawn a background agent that inherits the full conversation
/goal Set a goal — keep working until the condition is metgated
/memory Edit CLAUDE.md files and memory settings
/pause-memory Pause automemory for this sessionalso /memory-pause, /toggle-memorygated
/recap Generate a one-line session recap now
/rename[name] Rename the current conversationalso /name
/resume[conversation id or search term] Resume a previous conversationalso /continue
/rewind Restore the code and/or conversation to a previous pointalso /checkpoint, /undo

Model & effort

Which model answers, and how hard it thinks.

/advisor Let Claude consult a stronger model at key momentsgated
/effort] Set effort level for model usage
/fast[on|off] Toggle fast mode: the same model, answering faster.gated
/model<model> Set the AI model for Claude Codegated
/plan[open|share|<description>] Enable plan mode or view the current session plan
/ultraplan<prompt> Draft an editable plan in Claude Code on the web.

Code & review

Reading changes and getting them reviewed.

/autofix-pr Monitor and autofix any issues with the current PRgated
/batch<instruction> Research and plan a large-scale change, then execute it in parallel across 5–30 isolated worktree agents that each open a PR.add-on
/diff View uncommitted changes and per-turn diffs
/init Initialize a new CLAUDE.md file with codebase documentationadd-on
/install-github-app Set up Claude GitHub Actions for a repository
/ultrareview Run a cloud-hosted multi-agent code review of the current branch, or of a pull request, and print the findings.gated

Background & parallel work

Send work away, run several at once, pick it up elsewhere.

/background[prompt] Send this session to the background and free the terminalalso /bg
/cloud-plugins Choose whether cloud sessions use the plugins enabled on this machine
/daemon Manage background services and routines
/desktop Continue the current session in Claude Desktopalso /appgated
/list-agents List subagents, teammates, and other Claude sessions you can messagealso /peers
/loops List, create, and delete loops
/mobile Show QR code to download the Claude mobile appalso /ios, /android
/remote-control[name] Control this session from your phone or claude.ai/codealso /rcgated
/remote-env Choose the default environment for cloud agentsgated
/session Show cloud session URL and QR codealso /remotegated
/stop Stop this background session; transcript and worktree are kept
/subtask<task> Send a subagent off with your full context; its result comes back here
/tasks View and manage everything running in the backgroundalso /bashes
/teleport Send this session to the cloud, or resume one from claude.aialso /tpgated
/workflow-launch-exec Execute a server-launched workflow handoff (workflow_launch event sessions only)gated
/workflows Browse running and completed workflows

Skills, plugins & agents

Extending what Claude can do.

/agents (removed) Ask Claude to create/manage subagents, or edit .claude/agents/
/artifacts Browse your published and shared artifacts
/designconsent | revoke Grant or revoke Claude agent access to your Design projects
/design-consent Grant Claude agent access to your Design projectsgated
/design-login Authorize design-system access for /design-sync with your claude.ai account
/design-revoke Revoke Claude agent access to your Design projectsgated
/design-sync Push your design system components to claude.ai/designadd-on
/fewer-permission-prompts Scan your transcripts for common read-only Bash and MCP tool calls, then add a prioritized allowlist to project .claude/settings.json to reduce permission prompts.add-on
/insights Generate a report analyzing your Claude Code sessionsadd-on
/plugin Manage Claude Code pluginsalso /plugins, /marketplace
/plugin-types[dir] Write claude-code.d.ts and claude-code-mcp.d.ts: the plugin API's TypeScript declarations and the inputs of the connected MCP tools, for typing a hooks module against this session
/powerup Discover Claude Code features through quick interactive lessons
/reload-plugins[--force] Activate pending plugin changes in the current session
/reload-skills Pick up skills added or changed on disk during this session
/skill-doctor Show which loaded skills are unused and costing contextgated
/skills List available skills
/team-onboarding Help teammates ramp on Claude Code with a guide from your usagegatedadd-on

Integrations

Connecting Claude to the rest of your tools.

/chrome Open Claude in Chrome settings
/ide[open] Manage IDE integrations and show status
/import[codex|gemini] [--dry-run] Import config from another AI coding agentgated
/install-slack-app Install the Claude Slack app
/mcp[reconnect|enable|disable [<server>|all]] Manage MCP serversgated
/setup-bedrock Reconfigure Amazon Bedrock authentication, region, or model pinsgated
/setup-vertex Reconfigure Google Vertex AI authentication, project, region, or model pinsgated
/web-setup Set up Claude Code on the web with your GitHub accountgated

Configuration

Settings, appearance, keybindings, diagnostics.

/auto-mode-setup[--request-id <uuid>] (--wizard posture=… scope=… depth=… --propose | --expect-sha256 <64-hex> --apply-file <path>) Teach auto mode about your environment, plus optional rule tweaksgated
/color Set the prompt bar color for this session
/configkey=value Set a setting by keyalso /settingsgated
/debug[issue description] Enable debug logging for this session and help diagnose issuesadd-on
/doctor Health-check your setup and fix issues: installation, unused extensions, duplicated or bloated memory files, slow hooks, updates, permissionsalso /checkup
/heapdump Dump the JS heap to ~/Desktopgated
/hooks View hook configurations for tool events
/install[options] Install Claude Code native build
/keybindings Open your keyboard shortcuts file
/permissions Manage allow and deny tool permission rulesalso /allowed-tools
/privacy-settings View and update your privacy settings
/radio Listen to Claude FM lo-fi radio
/release-notes View release notes
/sandboxcommand pattern Show and change sandboxing for this session: whether commands run sandboxed, and what is allowed without asking.gated
/scroll-speed Adjust mouse wheel scroll speed
/statusline Set up Claude Code's status line UIadd-on
/stickers Order Claude Code stickers
/terminal-setup Set up the key binding for newlines in your terminal, and quiet the audible bell.
/theme Change the theme
/tui[default|fullscreen] Set the terminal UI renderer (default | fullscreen)
/update Switch to the latest version (conversation continues)also /restartgated
/version Show this session's version (autoupdate may have a newer one)
/voice[hold|tap|off] Toggle voice modegated
/wellbeing Configure optional break reminders and quiet-hours nudgesalso /breaks, /break-reminder, /downtime

Account & usage

Signing in, what you have spent, what is left.

/bug[report] Report a bug or share your conversationalso /share
/exit End the session.also /quit
/extra-usage Renamed to /usage-creditsgated
/feedback[report] Send feedback to Anthropic or report a bug
/help Show help and available commands
/login Sign in with your Anthropic account
/logout Sign out from your Anthropic account
/passes Share a free week of Claude Code with friends and earn usage credits.gated
/pro-trial-expired Options shown when the Pro plan Claude Code trial has endedgated
/rate-limit-options Show options when rate limit is reachedgated
/status Show Claude Code status including version, model, account, API connectivity, and tool statuses
/upgrade Upgrade to Max for higher rate limits and more Opus
/usage Show session cost, plan usage, and what's contributing to your limitsalso /cost, /statsgated
/usage-credits Configure usage credits or request them from your admin when you hit a limitgated
/cost now an alias of /usage
/stats now an alias of /usage
/extra-usage renamed to /usage-credits

Shortcuts & symbols

Keys and characters, not commands.

Esc Stop what Claude is doing, without ending the session.
Esc Esc Open the rewind menu.
Shift + Tab Cycle permission modes: normal, auto-accept, plan.
Tab Toggle extended thinking.
Ctrl + R Search back through your command history.
Ctrl + V Paste an image from the clipboard.
Ctrl + L Clear the terminal without clearing the conversation.
Ctrl + C Interrupt. Twice to exit.
Ctrl + D Exit the session.
@filename Reference a file or directory in your prompt.
#content Add a line to CLAUDE.md without leaving the session.
!command Run a shell command directly and keep its output in context.
"think harder" Raise effort for a single turn.
"ultrathink" Maximum reasoning depth for a single turn.

Starting Claude from the terminal

Options and subcommands you type before the session begins.

claude agents [options] Manage background agents
claude attach <id> Open a background session in this terminal. <id> is the short id that `claude --bg` prints and `claude agents` lists
claude auth Manage authentication
claude auto-mode Inspect or reset auto mode classifier configuration
claude doctor Check the health of your Claude Code installation. Reads settings files in the current directory without a trust prompt. For a full checkup that can also fix issues, run /doctor in a session.
claude gateway [options] Run the enterprise auth/telemetry gateway
claude import [options] [source] Import config from another AI coding agent into Claude Code
claude install [options] [target] Install Claude Code native build. Use [target] to specify version (stable, latest, or specific version)
claude logs <id> Print a background session's recent terminal output
claude mcp Configure and manage MCP servers
claude plugin|plugins Manage Claude Code plugins
claude project Manage Claude Code project state
claude respawn [options] [id] Restart a background session, or all of them with --all, so it runs the current Claude Code version
claude rm <id> Delete a background session, and its worktree when that is safe. Works on sessions that have already exited
claude setup-token Set up a long-lived authentication token (requires Claude subscription)
claude stop|kill <id> Stop a background session. Its conversation is kept: `claude attach <id>` opens it again, `claude --resume` works once it is stopped
claude ultrareview [options] [target] Run a cloud-hosted multi-agent code review of the current branch (or a PR number / base branch) and print the findings
claude update|upgrade Check for updates and install if available
--add-dir <directories...> Additional directories to allow tool access to
--agent <agent> Agent for the current session. Overrides the 'agent' setting.
--agents <json> JSON object defining custom agents (e.g. '{"reviewer": {"description": "Reviews code", "prompt": "You are a code reviewer"}}')
--allow-dangerously-skip-permissions Enable bypassing all permission checks as an option, without it being enabled by default. Recommended only for sandboxes with no internet access.
--allowedTools, --allowed-tools <tools...> Comma or space-separated list of tool names to allow (e.g. "Bash(git *) Edit")
--append-system-prompt <prompt> Append a system prompt to the default system prompt
--autocompact <auto|tokens> Auto-compact window size (auto, or 100k–1M tokens)
--ax-screen-reader Render screen-reader friendly output (flat text, no decorative borders or animations).
--bg, --background Start the session in the background and return immediately. Prints the id that `claude attach`, `logs`, `stop` and `rm` take; `claude agents` lists them. With --resume <session-id>, continues that session in the background under the same ID, or starts a copy and says so when the session is already running
--bare Minimal mode: skip hooks, LSP, plugin sync, attribution, auto-memory, background prefetches, keychain reads, and CLAUDE.md auto-discovery. Sets CLAUDE_CODE_SIMPLE=1. Anthropic auth is strictly ANTHROPIC_API_KEY or apiKeyHelper via --settings (OAuth and keychain are never read). 3P providers (Bedrock/Vertex/Foundry) use their own credentials. Skills still resolve via /skill-name. Explicitly provide context via: --system-prompt[-file], --append-system-prompt[-file], --add-dir (CLAUDE.md dirs), --mcp-config, --settings, --agents, --plugin-dir.
--betas <betas...> Beta headers to include in API requests (API key users only)
--brief Enable SendUserMessage tool for agent-to-user communication
--chrome Enable Claude in Chrome integration
--cloud [description|session_id|url] Create a cloud session with the given description, or attach to an existing one by session ID or claude.ai/code URL
-c, --continue Continue the most recent conversation in the current directory
--dangerously-skip-permissions Bypass all permission checks. Recommended only for sandboxes with no internet access.
-d, --debug [filter] Enable debug mode with optional category filtering (e.g., "api,hooks" or "!1p,!file")
--debug-file <path> Write debug logs to a specific file path (implicitly enables debug mode)
--disable-slash-commands Disable all skills
--disallowedTools, --disallowed-tools <tools...> Comma or space-separated list of tool names to deny (e.g. "Bash(git *) Edit")
--effort <level> Effort level for the current session (low, medium, high, xhigh, max)
--environment <environment_id> Create a new cloud session that runs on the given self-hosted environment (ccpool_...).
--exclude-dynamic-system-prompt-sections Move per-machine sections (cwd, env info, memory paths, git status) from the system prompt into the first user message. Improves cross-user prompt-cache reuse. Only applies with the default system prompt (ignored with --system-prompt). (default: false)
--fallback-model <model> Enable automatic fallback to specified model(s) when the default model is overloaded or not available. Accepts a comma-separated list to try each in order. Re-tries the primary at the start of each user turn. (only works with --print)
--file <specs...> File resources to download at startup. Format: file_id:relative_path (e.g., --file file_abc:doc.txt file_def:img.png)
--fork-session When resuming, create a new session ID instead of reusing the original (use with --resume or --continue)
--forward-subagent-text Forward subagent text and thinking blocks as assistant/user messages with parent_tool_use_id set (only works with --print and --output-format=stream-json)
--from-pr [value] Resume a session linked to a PR by PR number/URL, or open interactive picker with optional search term
-h, --help Display help for command
--ide Automatically connect to IDE on startup if exactly one valid IDE is available
--include-hook-events Include all hook lifecycle events in the output stream (only works with --output-format=stream-json)
--include-partial-messages Include partial message chunks as they arrive (only works with --print and --output-format=stream-json)
--input-format <format> Input format (only works with --print): "text" (default), or "stream-json" (realtime streaming input) (choices: "text", "stream-json")
--json-schema <schema> JSON Schema for structured output validation. Example: {"type":"object","properties":{"name":{"type":"string"}},"required":["name"]}
--max-budget-usd <amount> Maximum dollar amount to spend on API calls (only works with --print)
--mcp-config <configs...> Load MCP servers from JSON files or strings (space-separated)
--model <model> Model for the current session. Provide an alias for the latest model (e.g. 'fable', 'opus', or 'sonnet') or a model's full name (e.g. 'claude-fable-5').
-n, --name <name> Set a display name for this session (shown in the prompt box, /resume picker, and terminal title)
--no-chrome Disable Claude in Chrome integration
--no-session-persistence Disable session persistence - sessions will not be saved to disk and cannot be resumed (only works with --print)
--output-format <format> Output format (only works with --print): "text" (default), "json" (single result), or "stream-json" (realtime streaming) (choices: "text", "json", "stream-json")
--permission-mode <mode> Permission mode to use for the session (choices: "acceptEdits", "auto", "bypassPermissions", "manual", "dontAsk", "plan")
--permission-prompts <target> Who answers permission prompts with --print: "host" (the SDK host or --permission-prompt-tool) or "none" (nobody: anything that would prompt is denied automatically; the permission mode still decides everything else) (choices: "host", "none", default: "host")
--plugin-dir <path> Load a plugin from a directory or .zip for this session only (repeatable: --plugin-dir A --plugin-dir B.zip) (default: [])
--plugin-url <url> Fetch a plugin .zip from a URL for this session only (repeatable: --plugin-url A --plugin-url B) (default: [])
-p, --print Print response and exit (useful for pipes). Note: The workspace trust dialog is skipped when Claude is run in non-interactive mode (via -p, or when stdout is not a TTY, e.g. piped or redirected output). Only use this in directories you trust. Settings files that fail validation are silently ignored in this mode (no error dialog is shown).
--prompt-suggestions [value] Enable prompt suggestions. In print/SDK mode, emits a prompt_suggestion message after each turn with a predicted next user prompt (choices: "true", "false", "1", "0", "yes", "no", "on", "off", preset: "true")
--remote-control [name] Start an interactive session with Remote Control enabled (optionally named)
--remote-control-session-name-prefix <prefix> Prefix for auto-generated Remote Control session names (default: hostname)
--replay-user-messages Re-emit user messages from stdin back on stdout for acknowledgment (only works with --input-format=stream-json and --output-format=stream-json)
--restricted Restricted mode: removes the built-in tools that run commands or code (Bash, PowerShell, REPL and the other code-running tools) and WebFetch unless --tools names them, and ignores user, project and local settings files (managed settings and --settings still apply; add --strict-mcp-config to skip MCP servers too). Also confines the file tools to the working directories (--add-dir included), refuses bypassPermissions, and lets only a person or the configured permission handler approve writes to settings, git and tool-configuration files.
-r, --resume [value] Resume a conversation by session ID, or open interactive picker with optional search term
--safe-mode Start with all customizations (CLAUDE.md, skills, plugins, hooks, MCP servers, custom commands and agents, output styles, workflows, custom themes, keybindings, and more) disabled — useful for troubleshooting a broken configuration. Admin-managed (policy) settings still apply. Auth, model selection, built-in tools, and permissions work normally. Sets CLAUDE_CODE_SAFE_MODE=1.
--session-id <uuid> Use a specific session ID for the conversation (must be a valid UUID)
--setting-sources <sources> Comma-separated list of setting sources to load (user, project, local).
--settings <file-or-json> Path to a settings JSON file or a JSON string to load additional settings from
--strict-mcp-config Only use MCP servers from --mcp-config, ignoring all other MCP configurations
--system-prompt <prompt> System prompt to use for the session
--system-prompt-snapshot <on|off> Record the system prompt once per conversation and reuse it verbatim on every request and resume (recommended: on). By default it is on for the built-in prompt, and passing --system-prompt or --append-system-prompt turns it off so the given text applies fresh each launch. on: an existing record in the conversation is sent as-is (a later launch's different --system-prompt/--append-system-prompt is ignored until compaction); otherwise the prompt is rendered with the append included, sent, and recorded. off: never record. No effect where system-prompt recording is not yet enabled. (choices: "on", "off")
--teleport [session] Resume a teleport session, optionally specify session ID
--tmux Create a tmux session for the worktree (requires --worktree). Uses iTerm2 native panes when available; use --tmux=classic for traditional tmux.
--tools <tools...> Specify the list of available tools from the built-in set. Use "" to disable all tools, "default" to use all tools, or specify tool names (e.g. "Bash,Edit,Read").
--verbose Override verbose mode setting from config
-v, --version Output the version number
-w, --worktree [name] Create a new git worktree for this session (optionally specify a name)

Every command, flag and shortcut in Claude Code 2.1.259.

The twelve worth learning first.