AI & Automation

How to Set Up Claude Code: Complete Installation Guide (2026)

April 23, 2026 · 7 min read
Featured image for How to Set Up Claude Code: Complete Installation Guide (2026)

Setting up Claude Code takes less than five minutes regardless of your operating system or preferred development environment. But getting the most out of it — configuring project instructions, connecting external tools, and optimizing your workflow — takes a bit more intention.

We’ve set up Claude Code across dozens of projects at Markana Media, from marketing automation systems to full client websites. Here’s the complete setup guide we wish we’d had from day one.

Prerequisites Before You Install

Before installing Claude Code, you need one of the following:

  • A Claude subscription — Claude Pro ($20/month), Team ($30/user/month), or Enterprise. This gives you direct access through Anthropic’s infrastructure.
  • An Anthropic Console account — If you’re using the API directly, you’ll need an API key from console.anthropic.com. You pay per token used.
  • A third-party provider — Claude Code also supports Amazon Bedrock, Google Vertex AI, and Microsoft Foundry for organizations that need to run on their own cloud infrastructure.

You’ll also need Node.js 18+ already installed on your system (the installer handles the rest), and on Windows, you’ll need Git for Windows.

Installing the Terminal CLI

The terminal CLI is the most powerful way to use Claude Code. It gives you full access to every feature: file editing, command execution, git operations, MCP integrations, and sub-agents.

macOS and Linux:

curl -fsSL https://claude.ai/install.sh | bash

Windows (PowerShell):

irm https://claude.ai/install.ps1 | iex

Windows (CMD):

curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

Homebrew (macOS):

brew install --cask claude-code

WinGet (Windows):

winget install Anthropic.ClaudeCode

The native installer (curl/irm methods) auto-updates in the background. Homebrew and WinGet installations require manual updates — run brew upgrade claude-code or winget upgrade Anthropic.ClaudeCode periodically.

Once installed, open a terminal, navigate to any project directory, and type:

cd your-project
claude

On first run, you’ll be prompted to authenticate. Follow the browser link, log in, and you’re ready to go.

Setting Up the VS Code Extension

If you prefer working inside your editor, the VS Code extension brings Claude Code directly into your workflow with inline diffs, @-mentions, and plan review.

Install it through one of these methods:

  • Open VS Code, go to Extensions (Cmd+Shift+X on Mac, Ctrl+Shift+X on Windows), and search for “Claude Code”
  • Or run from the terminal: code --install-extension anthropic.claude-code
  • For Cursor users: cursor --install-extension anthropic.claude-code

After installing, open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P), type “Claude Code,” and select Open in New Tab. The extension connects to the same Claude Code engine as the CLI, so your CLAUDE.md files and settings carry over automatically.

Installing the Desktop App

The desktop app is a standalone interface that’s perfect for reviewing diffs visually, running multiple sessions side by side, and scheduling recurring tasks.

Download the app for your platform:

  • macOS: Download the DMG from claude.com/download (Intel and Apple Silicon supported)
  • Windows: Download the setup executable from the same page (x64 and ARM64 available)

Launch the app, sign in with your Claude account, and click the Code tab to start a session. You can point it at any local project directory.

Configuring Your First Project With CLAUDE.md

This is the step most people skip, and it’s the one that makes the biggest difference. A CLAUDE.md file in your project root gives Claude Code persistent instructions it follows every session.

Here’s a minimal example:

# Project: My Marketing Site

## Tech Stack
- Next.js 14 with App Router
- Tailwind CSS
- PostgreSQL via Prisma

## Conventions
- Use TypeScript strict mode
- Components go in src/components/
- API routes go in src/app/api/
- Always write tests for new features

## Deployment
- Push to main triggers Vercel deploy
- Never commit .env files

At our agency, our CLAUDE.md files are much more detailed — covering brand voice rules, SEO checklists, deployment procedures, and API integration patterns. The more context you give Claude Code upfront, the less you have to repeat yourself in every conversation.

Claude Code also builds auto memory as it works. It saves learnings like build commands, debugging insights, and project-specific patterns across sessions without you writing anything. These memories persist in a .claude/ directory.

Setting Up for Teams

If you’re deploying Claude Code across a team, there are a few additional configuration options:

Claude Team or Enterprise plan: These plans include SSO, centralized billing, admin controls, and usage dashboards. You can self-serve at claude.ai/upgrade or contact Anthropic’s sales team for enterprise needs.

Cloud providers: For organizations that need data residency or compliance controls, Claude Code can run through Amazon Bedrock, Google Vertex AI, or Microsoft Foundry. Each requires configuring your cloud account first — enabling the Claude models, setting up IAM credentials, and pointing Claude Code at your cloud endpoint.

Shared CLAUDE.md: Commit your project’s CLAUDE.md to version control so every team member gets the same instructions. You can also create user-level settings in ~/.claude/ for personal preferences that don’t need to be shared.

Custom skills: Create team-wide slash commands that encode your workflows. A /deploy command that runs your exact deployment steps, a /review command that follows your code review checklist. These get committed to the repo and shared with everyone.

Connecting External Tools With MCP

The Model Context Protocol (MCP) is what turns Claude Code from a coding tool into a full workflow platform. MCP servers let Claude Code connect to external systems — your CRM, project management tools, design systems, databases, and anything else with an API.

Some popular MCP servers we use at Markana Media:

  • Google Drive — Claude Code reads design docs, briefs, and spreadsheets directly
  • Slack — Pull context from channels and send notifications
  • GitHub — Enhanced PR and issue management beyond the built-in git support
  • Databases — Query production data for debugging or analysis

MCP servers are configured in your Claude Code settings and persist across sessions. Once connected, you can reference external data naturally in your prompts: “Check the client brief in Google Drive and build the landing page to match.”

If you need help setting up Claude Code for your team or integrating it into your development workflow, we specialize in exactly this kind of AI automation setup. Whether it’s a single project or a company-wide rollout, we can get you running in a day.