██████╗██╗ █████╗ ██╗ ██╗██████╗ ███████╗ ██████╗ ██████╗ ██████╗ ███████╗ ██╔════╝██║ ██╔══██╗██║ ██║██╔══██╗██╔════╝ ██╔════╝██╔═══██╗██╔══██╗██╔════╝ ██║ ██║ ███████║██║ ██║██║ ██║█████╗ ██║ ██║ ██║██║ ██║█████╗ ██║ ██║ ██╔══██║██║ ██║██║ ██║██╔══╝ ██║ ██║ ██║██║ ██║██╔══╝ ╚██████╗███████╗██║ ██║╚██████╔╝██████╔╝███████╗ ╚██████╗╚██████╔╝██████╔╝███████╗ ╚═════╝╚══════╝╚═╝ ╚═╝ ╚═════╝ ╚═════╝ ╚══════╝ ╚═════╝ ╚═════╝ ╚═════╝ ╚══════╝

Claude Code Cheat Sheet

Quick reference for Anthropic's CLI coding assistant
Last updated: January 2026

Quick Links: Installation | Slash Commands | Config Files | Context | GitHub | CLI | Hooks | MCP | Shortcuts | Best Practices | Resources

Installation

Platform	Command
macOS/Linux	curl -fsSL https://claude.ai/install.sh | bash
Homebrew	brew install --cask claude-code
Windows	irm https://claude.ai/install.ps1 | iex
WinGet	winget install Anthropic.ClaudeCode

Slash Commands

Command	Description
/agents	Create and manage subagents
/allowed-tools	Manage tool permissions
/bug	Report issues
/config	Interactive settings configuration
/context	View context usage breakdown
/help	Display all available commands
/hooks	Set up automation triggers
/install-github-app	Set up GitHub Actions integration
/mcp	Configure MCP servers
/terminal-setup	Install keyboard shortcuts
/vim	Enable vim-style editing

Read more: Comprehensive slash commands guide

Tip: Use /context regularly to monitor token usage and avoid context exhaustion.

Configuration Files

File	Scope	Location
User settings	All projects	~/.claude/settings.json
Project settings	Team (git tracked)	.claude/settings.json
Local settings	Personal (git ignored)	.claude/settings.local.json
Memory file	Project context	CLAUDE.md (hierarchical)
Custom agents	Subagent definitions	.claude/agents/
Custom commands	Slash commands	.claude/commands/

Read more: CLAUDE.md memory system explained

Context Categories

Category	Description
System prompt	Claude's base instructions
System tools	Built-in tools (Read, Write, Bash, etc.)
Custom agents	Your .claude/agents/ definitions
Memory files	Your CLAUDE.md files
MCP tools	Model Context Protocol servers
Reserved	Buffer (~40-45k) for auto-compact
Messages	Conversation history
Free space	Available for new content

Read more: Understanding the /context command

GitHub Integration

Quick Setup:

/install-github-app

@claude Mentions:

@claude review this PR
@claude fix the TypeError in dashboard
@claude implement user authentication
@claude /review

Local Git Operations:

commit these changes
create a pr for this branch
what changes made it into v1.2.3?

Read more: GitHub workflow integration

CLI Arguments

Argument	Description
-p "prompt"	Print mode (single query, then exit)
--model <model>	Specify Claude model
--max-turns <n>	Limit conversation turns
--allowed-tools <list>	Restrict available tools
--mcp-config <path>	MCP configuration file
--debug	Enable debug output

Piping:

tail -f app.log | claude -p "analyze these logs"
cat error.txt | claude -p "explain this error"

Hooks

Hook	Trigger
PreToolUse	Before a tool is executed
PostToolUse	After a tool completes
UserPromptSubmit	When user sends a message
SessionStart	When a new session begins

Example use cases:

• Auto-format code after edits
• Run linters before commits
• Send notifications on completion

MCP (Model Context Protocol)

Connect external tools and data sources:

• Google Drive (design docs)
• Figma (designs)
• Slack (team communication)
• Jira (issue tracking)
• Custom tooling

Setup:

/mcp

Warning: MCP tools can consume significant context (40%+). Disable unused servers.

Keyboard Shortcuts

Shortcut	Action
Ctrl+C	Cancel current operation
Ctrl+D	Exit Claude Code
Tab	Autocomplete
Up/Down	Navigate history
Ctrl+T	Toggle syntax highlighting (in /theme)

Install terminal shortcuts:

/terminal-setup

Best Practices

1. Use CLAUDE.md - Encode project conventions, build commands, standards (guide)
2. Start focused sessions - New conversations for distinct tasks
3. Be specific - Target file reads, specific grep patterns
4. Use Task tool - Let Claude spawn subagents for exploration
5. Monitor /context - Avoid context exhaustion
6. Review changes - Always review before accepting

Resources

• Official Docs
• GitHub Repository
• GitHub Action
• Best Practices

[ More articles coming soon - check back for updates ]