Claude Code cheatsheet

curl -fsSL https://claude.ai/install.sh | bash  # native
brew install --cask claude-code                 # Homebrew
npm install -g @anthropic-ai/claude-code        # npm, Node 22+
claude install stable                           # stable channel

On Windows: irm https://claude.ai/install.ps1 | iex in PowerShell, or winget install Anthropic.ClaudeCode.

See: Setup

claude                       # first run opens a browser
claude auth login            # sign in again
claude auth login --console  # bill to a Console account
claude auth status           # JSON status; --text for plain

See: Authentication

claude                        # start interactive session
claude "explain this repo"    # start with a prompt
claude -p "summarize README"  # print mode: answer and exit
git diff | claude -p "review this diff"

Use claude -c to continue the last conversation, or claude -r "auth-refactor" to resume one by name. See: Quickstart

Sessions

Command	Description
claude	Start interactive session
claude "query"	Start with an initial prompt
claude -p "query"	Print response and exit
claude -c	Continue most recent conversation
claude -r <session>	Resume a session by ID or name
claude --bg "task"	Start a background agent

Setup

Command	Description
claude update	Update to latest version
claude install	Install or pin a version
claude doctor	Check install and settings
claude mcp	Configure MCP servers
claude plugin	Install and manage plugins
claude auth status	Show authentication status
claude setup-token	Generate a token for CI

See: CLI reference

Flag	Description
-c, --continue	Continue most recent conversation
-r, --resume	Resume a session by ID or name
--model sonnet	Choose the model
--effort high	Set reasoning effort
--fallback-model sonnet	Fall back when overloaded
--permission-mode plan	Start in a permission mode
--add-dir ../lib	Add extra working directories
--verbose	Show full turn-by-turn output

See: CLI reference

Flag	Description
--debug	Enable debug mode
--settings file.json	Load settings for this session
--ide	Auto-connect to your IDE
-w, --worktree	Start in an isolated git worktree
--dangerously-skip-permissions	Skip all permission prompts

See: CLI reference

Flag	Description
-p, --print	Print the response and exit
--output-format json	text, json, or stream-json
--json-schema schema	Constrain output to a schema
--max-turns 3	Limit agentic turns
--max-budget-usd 5	Stop the run at a spend cap
--bare	Skip hooks, plugins, and memory

See: Headless mode

Conversation

Command	Description
/help	Show available commands
/clear	Start a new conversation
/compact	Summarize to free up context
/context	Show context window usage

Planning

Command	Description
/plan	Enter plan mode
/goal <condition>	Keep working until it is met
/btw <question>	Ask a side question

History

Command	Description
/resume	Resume a previous conversation
/rewind	Restore code and conversation
/branch	Branch the conversation here
/export	Export the conversation as text

Usage

Command	Description
/status	Show version, model, and account
/usage	Show cost and plan limits

See: Commands

Settings

Command	Description
/config	Open settings or set a value
/model	Switch the AI model
/effort	Set reasoning effort
/memory	Edit memory files

Project

Command	Description
/init	Generate a starter CLAUDE.md
/permissions	Allow, ask, and deny rules
/mcp	Manage MCP connections
/hooks	View hook configurations
/doctor	Run a setup checkup

Account

Command	Description
/login	Sign in to your account
/logout	Sign out
/exit	Exit the CLI (/quit)

See: Commands

Changes

Command	Description
/diff	Review working-tree changes
/code-review	Review the diff, a PR, or a path
/security-review	Scan changes for vulnerabilities

/review is an alias of /code-review, which takes an effort level, --fix, and a PR number.

See: Review locally with /code-review

Command	Description
/background	Detach this session as a background agent
/fork	Copy it into a new background session
/subtask <task>	Run a task in a forked subagent
/tasks	View and manage background work

See: Agent view

Session

Shortcut	Description
Ctrl+C	Interrupt, clear input, or exit
Ctrl+D	Exit Claude Code
Esc	Interrupt Claude, or close dialogs
Esc Esc	Clear the draft, or rewind
Ctrl+B	Background running tasks
Ctrl+X Ctrl+K	Stop all background subagents; press twice

Prompt

Shortcut	Description
Shift+Tab	Cycle permission modes
Alt+P	Switch model
Alt+T	Toggle extended thinking
Alt+O	Toggle fast mode
Ctrl+V	Paste an image from the clipboard
Ctrl+G	Edit the prompt in $EDITOR
Ctrl+S	Stash or restore the prompt

Display

Shortcut	Description
Ctrl+L	Redraw the screen
Ctrl+R	Search prompt history
Ctrl+O	Toggle the transcript viewer
Ctrl+T	Toggle Claude's task list

On macOS, use Option instead of Alt; some chords need Option mapped to Meta.

See: Interactive mode

Input	Effect
/	Slash commands and skills
!	Run a shell command directly
@	Mention a file path
:	Insert an emoji by shortcode
?	Show the shortcut help panel

For a newline, \ + Enter or Ctrl+J works anywhere; Shift+Enter needs /terminal-setup in some terminals.

See: Interactive mode

~/.claude/settings.json                # user settings
.claude/settings.json                  # project settings, committed
.claude/settings.local.json            # personal project settings
/etc/claude-code/managed-settings.json # managed policy (Linux)

Run /config to edit settings, or pass --settings file.json for one session. See: Settings

Variable	Purpose
ANTHROPIC_API_KEY	API key used instead of a subscription
ANTHROPIC_MODEL	Default model name
ANTHROPIC_BASE_URL	Route requests through a gateway
API_TIMEOUT_MS	API timeout (default 600000)
BASH_DEFAULT_TIMEOUT_MS	Bash tool timeout (default 120000)
CLAUDE_CODE_DISABLE_AUTO_MEMORY	Turn auto memory off
DISABLE_AUTOUPDATER	Turn background updates off

Settings files can also set env vars under an env key. See: Environment variables

CLAUDE.md             # project instructions, committed
CLAUDE.local.md       # personal notes, not committed
.claude/rules/*.md    # topic rules, optionally path-scoped
~/.claude/CLAUDE.md   # instructions for all your projects

/init generates a project CLAUDE.md; /memory edits memory files and auto memory. Import files with @path/to/file.md. See: Memory

Mode	What runs without asking
default	Reads only (labeled Manual)
acceptEdits	Reads, file edits, filesystem commands
plan	Reads, until you approve a plan
auto	Everything, with safety checks
dontAsk	Pre-approved tools only; rest denied
bypassPermissions	Everything

auto is the built-in starting mode; Shift+Tab cycles modes in a session. A classifier can still block actions, and explicit ask rules still prompt. See: Permission modes

git diff | claude -p "review this diff for bugs"
claude -p "check src/ for security issues"

In a session, /code-review reviews the diff, /code-review 1234 a pull request, and /security-review scans for vulnerabilities.

See: Review locally with /code-review

# One-shot JSON output for scripts
claude -p "list the failing tests" --output-format json

# Pipe logs in
cat build.log | claude -p "explain this failure"

--bare skips hooks, plugins, and auto-discovered config so scripted calls start faster. It ignores OAuth credentials, so set ANTHROPIC_API_KEY.

See: Headless mode

claude --bg "migrate the auth tests"  # detach immediately
claude agents                         # monitor and dispatch

Inside a session, /background detaches it and /batch fans work out across worktrees.

See: Agent view