Claude Code cheat sheet

Every command, flag and slash command for claude — Anthropic's agentic coder — a terminal REPL that reads, edits and runs your repo. 21+ commands, verified July 2026.

Install

curl -fsSL https://claude.ai/install.sh | bash     # macOS / Linux / WSL
irm https://claude.ai/install.ps1 | iex            # Windows (PowerShell)
brew install --cask claude-code                    # Homebrew
npm install -g @anthropic-ai/claude-code           # npm (needs Node 22+)

claude --version   ·   claude doctor   ·   claude update

Needs a Claude Pro / Max / Team / Enterprise or Console account (the free plan does not include Claude Code). Native installs auto-update; brew/npm do not.

Start & run

claude                        # interactive REPL in the current dir
claude "explain this repo"    # REPL seeded with a prompt
claude -p "query"             # headless: print the answer and exit
claude -c                     # continue the most recent conversation here
claude -r "<session>" "…"     # resume a specific session by id or name
cat file | claude -p "…"      # pipe content in, headless

Key flags

Flag	What it does
`-p, --print`	Non-interactive: print the response and exit (scripting / CI)
`-c, --continue`	Load the most recent conversation in this directory
`-r, --resume`	Resume a specific session by id or name
`--model <m>`	Alias `sonnet` / `opus` / `haiku` / `fable`, or a full model id
`--permission-mode <m>`	`default`, `acceptEdits`, `plan`, or `bypassPermissions`
`--dangerously-skip-permissions`	Skip every permission prompt (= `bypassPermissions`)
`--output-format <fmt>`	`text` (default), `json`, or `stream-json`
`--add-dir <path>`	Grant read/edit access to extra working directories
`--append-system-prompt <text>`	Append text to the default system prompt

In-session slash commands

Command	What it does
`/clear`	Start a new conversation with empty context
`/compact`	Summarize the conversation to free up context
`/context`	Visualize current context-window usage
`/model`	Switch model and save it as the default
`/init`	Generate a `CLAUDE.md` project guide
`/memory`	Edit `CLAUDE.md` memory files
`/review [PR]`	Fast single-pass review of a GitHub PR
`/agents`	Manage subagents (edit `.claude/agents/`)
`/mcp`	Manage MCP server connections
`/plan`	Enter plan mode for large changes
`/permissions`	Manage allow / ask / deny tool rules
`/cost` · `/status` · `/help`	Usage & cost · session status · help

Config

~/.claude/settings.json          # user, all projects
.claude/settings.json            # project (git-shared)
.claude/settings.local.json      # local overrides (gitignored)
~/.claude/CLAUDE.md   ·   ./CLAUDE.md    # user / project memory
.mcp.json                        # MCP servers (project)

# env: ANTHROPIC_API_KEY · ANTHROPIC_MODEL · DISABLE_AUTOUPDATER=1
# backends: Amazon Bedrock · Google Vertex AI · Microsoft Foundry

Common one-liners

claude -p "summarize what this service does"
git diff | claude -p "review these changes for bugs"
claude -c                                              # resume last session
claude --model opus -p "refactor utils.py for readability"

More CLI cheat sheets

Stop memorizing flags

Run Claude Code — and every other agent — from one app

TheVibeFather orchestrates Claude Code alongside every other coding CLI in one native macOS app. Bring the models you already pay for; keep your keys.

Get started See how it works