Claude Code Cheat Sheet

TL;DR Highlight

A cheat sheet for developers who use Claude Code daily but keep forgetting commands — covering everything from keyboard shortcuts to MCP configuration, memory management, and CLI flags, on one page. With auto-update to always stay current.

Who Should Read

Developers who use Claude Code in the terminal daily but frequently forget commands or shortcuts — especially those who want to properly leverage advanced features like MCP server configuration, agent workflows, and context management.

Core Mechanics

The cheat sheet covers four main areas: keyboard shortcuts (Esc escaping, Ctrl+C canceling, etc.), slash commands (/clear, /compact, /memory), CLI flags (--continue, --resume, --print), and MCP configuration syntax.
The memory management section is particularly useful: it documents the difference between /memory (view current memory), /memory add (add a memory), and the CLAUDE.md file for persistent project context.
The MCP section documents the JSON config format for adding MCP servers, which is underdocumented in the official docs and frequently asked about.
Auto-update is implemented by fetching the latest version from a remote source, so the cheat sheet reflects new Claude Code features as they're released.
The project is open-source and accepts contributions — several community members have added tips for specific use cases (monorepos, Python projects, etc.).

Evidence

The cheat sheet became a popular bookmark in developer communities within days of release — many developers reported learning shortcuts they didn't know existed.
Commenters frequently mentioned the MCP configuration section as the most valuable part, as the official documentation on MCP setup is sparse.
Several developers noted they'd been using /clear when they should have been using /compact — a common mistake the cheat sheet clarifies.
The auto-update feature was noted as particularly thoughtful since Claude Code releases updates frequently and cheat sheets for rapidly-evolving tools go stale fast.

How to Apply

Bookmark or pin the cheat sheet URL in your browser for quick access during Claude Code sessions.
Pay special attention to /compact vs /clear — /compact summarizes context to free up space while preserving memory, while /clear wipes everything.
For team adoption, share the MCP configuration section specifically — it answers the most common setup questions.
If you're using Claude Code with a specific stack (Python, TypeScript, monorepo), check the community contributions section for stack-specific tips.

Code Example

snippet

# Analyze a file in non-interactive mode and receive JSON results
cat myfile.py | claude -p "Find bugs in this code" --output-format json --max-budget-usd 2

# Resume by session name
claude -r "my-session-name"

# Continue from the last conversation
claude -c

# Start in plan mode (review before execution)
claude --permission-mode plan

# Minimal headless mode (no hooks/LSP/plugins)
claude --bare

# Set default effort level via environment variable
export CLAUDE_CODE_EFFORT_LEVEL=high
export MAX_THINKING_TOKENS=0  # Disable thinking
export IS_DEMO=1  # Remove welcome banner

# Isolated work per feature using Git Worktree
claude --worktree feature/new-api

Terminology

/compactA Claude Code slash command that summarizes the current conversation context to reduce token usage while preserving key information.

/memoryA slash command for viewing and managing Claude Code's persistent memory across sessions.

CLAUDE.mdA project-level file that Claude Code reads as persistent context, useful for documenting architecture, conventions, and project-specific instructions.

MCP ServerAn external tool server connected to Claude Code via the Model Context Protocol, extending its capabilities.