Question 1

What is the difference between Claude Code and ChatGPT?

Accepted Answer

Claude Code is a terminal-based AI coding assistant made by Anthropic. It can directly access your file system to read, modify, and execute code. ChatGPT is a web-based conversational AI that can suggest code but cannot directly modify files.

Question 2

Is Claude Code free?

Accepted Answer

Claude Code itself is free to install, but costs are incurred based on API usage. You need an Anthropic API key, and you are charged based on tokens used. Claude Pro subscribers may receive API credits.

Question 3

What programming languages are supported?

Accepted Answer

Claude Code supports most programming languages including Python, JavaScript, TypeScript, Go, Rust, Java, C++, Ruby, and more. It performs exceptionally well with Python and JavaScript/TypeScript.

Question 4

When should I use the /compact command?

Accepted Answer

Use it when your conversation gets long and the context window fills up. /compact summarizes previous conversation to save tokens and allows you to continue longer conversations. Use /compact when you get "context too long" errors.

Question 5

Can Claude Code execute dangerous commands?

Accepted Answer

By default, Claude Code asks for user confirmation before dangerous operations (file deletion, system commands, etc.). It is safe unless you use the --dangerouslySkipPermissions flag. Only use this flag carefully in CI/CD environments.

Command	Description
`claude`	Start interactive mode
`claude "..."`	Ask once and exit
`claude -p "..."`	For pipeline/scripts (non-interactive)
`claude --cwd /path`	Start in specific directory
`claude --version`	Check version

Command	Description
`/help`	Show help
`/clear`	Clear conversation history
`/compact`	Compress context (save tokens)
`/config`	View/change settings
`/cost`	Check current session cost
`/doctor`	Diagnose settings
`/init`	Create CLAUDE.md file
`/login`	Login to account
`/logout`	Logout
`/memory`	Edit memory
`/model`	Change model
`/permissions`	View permissions
`/review`	Request code review
`/terminal-setup`	Terminal setup (Shift+Enter, etc.)
`/vim`	Toggle Vim mode

Command	Description
`--model <model>`	Specify model to use
`--output-format text`	Output text only
`--output-format json`	Output in JSON format
`--output-format stream-json`	Streaming JSON output
`--system-prompt "..."`	Add system prompt
`--mcp-config <file>`	Specify MCP config file
`--allowedTools "..."`	Specify allowed tools
`--permission-mode plan`	Read-only mode
`--dangerouslySkipPermissions`	Skip all permission checks (for CI)
`--resume`	Resume last conversation
`--continue`	Continue most recent conversation

Command	Description
`cat file \| claude "Analyze this"`	Pass file content as input
`git diff \| claude "Review this"`	Analyze/review git diff
`git diff --staged \| claude "Write commit message"`	Generate commit message
`npm test 2>&1 \| claude "Analyze errors"`	Analyze test errors
`git log --oneline \| claude "Write PR description"`	Write PR description

Command	Description
`Shift+Enter`	Multi-line input (setup required)
`Tab`	Autocomplete (filenames, etc.)
`↑ / ↓`	Navigate previous/next input
`Ctrl+C`	Interrupt current response
`Ctrl+D`	End session
`Escape`	Cancel current input

Claude Code Cheatsheet

#Basic Usage

#Slash Commands

#CLI Flags

#Pipeline Usage

#Keyboard Shortcuts

#Environment Variables

#Frequently Asked Questions

Command	Description
`ANTHROPIC_API_KEY`	Anthropic API key
`CLAUDE_MODEL`	Default model setting
`HTTP_PROXY`	HTTP proxy setting
`HTTPS_PROXY`	HTTPS proxy setting