Claude Code — Getting Started
Install and configure Claude Code, Anthropic's terminal-based AI coding agent. Covers installation, API key setup, project configuration, and first session walkthrough.
Claude Code — Getting Started
Claude Code is Anthropic's terminal-based AI coding agent. It reads your codebase, writes and edits files, runs terminal commands, and can autonomously complete multi-step engineering tasks — all through natural language prompts in your terminal.
Installation
# Via npm (recommended)
npm install -g @anthropic-ai/claude-code
# Via Homebrew
brew install claude-code
# Verify installation
claude --version
Authentication
You need an Anthropic API key. Get one at console.anthropic.com (API Keys section).
# Set as environment variable
export ANTHROPIC_API_KEY="sk-ant-..."
# Or Claude Code will prompt on first run
claude
First Session
Navigate to your project directory and launch Claude Code:
cd my-project
claude
Claude Code scans your project structure automatically. On first run, you'll see:
⚡ Welcome to Claude Code!
Directory: /Users/you/my-project
Model: claude-sonnet-4-20250514 (default)
Context: 200K tokens
Type /help for available commands.
>
Type a prompt to start:
> Explain the architecture of this project and suggest improvements
Claude Code will read relevant files, analyze the structure, and respond inline in your terminal.
Essential Slash Commands
| Command | What it does |
|---|---|
/help | Show all available commands |
/file <path> | Add a file to the conversation context |
/clear | Clear conversation history |
/compact | Compress context by summarizing history |
/config | Open configuration settings |
/cost | Show current session token costs |
/doctor | Diagnose installation issues |
/init | Initialize CLAUDE.md for this project |
Project Configuration: CLAUDE.md
Create a CLAUDE.md file in your project root (or .claude/CLAUDE.md). Claude Code reads this every session — it's your project's system prompt.
# CLAUDE.md
## Commands
- `npm run dev` — Start development server
- `npm run build` — Production build
- `npm run lint` — ESLint + TypeScript check
- `npm run test` — Run test suite
## Architecture
- Next.js 14 App Router
- TypeScript strict mode
- Tailwind CSS for styling
- PostgreSQL via Prisma ORM
## Conventions
- Use server components by default
- `'use client'` only for interactivity
- Path alias `@/` maps to project root
- No any types — use explicit types or unknown
See Prompt Engineering in Claude Code for advanced CLAUDE.md patterns.
Environment Variables
| Variable | Purpose |
|---|---|
ANTHROPIC_API_KEY | API authentication (required) |
ANTHROPIC_MODEL | Override default model |
ANTHROPIC_MAX_TOKENS | Max output tokens per request |
ANTHROPIC_BASE_URL | Custom API endpoint (proxies) |
Related Pages
- Prompt Engineering in Claude Code — CLAUDE.md patterns, extended thinking
- Claude Code Configuration Reference — claude.json, MCP, hooks, .claude/ conventions
- Tool Comparison — Claude Code vs OpenCode vs Gemini CLI vs Cursor vs Copilot
- Claude Code vs Gemini CLI vs OpenCode (Blog) — Head-to-head comparison
- Prompting Claude Code (Prompts Section) — Detailed CLAUDE.md command patterns
- MCP Setup for Claude Code — Adding MCP servers
Related Articles
AI Coding Tools — Prompt Engineering & Configuration
Complete guides for Claude Code, OpenCode, Gemini CLI, Cursor, GitHub Copilot, and more. Tool-specific prompt patterns, configuration files, and setup guides for every major AI coding assistant.
Prompt Engineering in Cursor
Master prompt engineering in Cursor IDE. Agent mode vs inline editing, composer prompts, codebase-aware prompting, and .cursorrules integration strategies.
Tabnine — AI Coding Tool Guide
Enterprise-focused AI code assistant with code privacy, on-premise deployment, and team learning. Configuration and prompt patterns for Tabnine Chat.