How to Use Claude Code: A Beginner’s Guide

If you have been copying code into ChatGPT, pasting it back into your editor, and losing all your project context every single time  there is a better way to work with AI.

Claude Code is Anthropic’s official AI coding assistant that runs directly in your terminal. It reads your entire codebase, edits files across your project, runs commands, handles Git operations, and executes multi-step tasks all from a single natural language prompt.

This beginner guide walks you through everything from installing Claude Code for the first time to making your first real code change. No steps skipped. No jargon left unexplained. By the end, you will have Claude Code running and a clear workflow you can use every day. Let us get started!

Quick TL;DR Summary

1.  This guide explains what Claude Code is how it differs from Claude in the browser and other AI coding tools, and why developers are adopting it as their primary coding workflow.

2.  It walks through the complete installation process, including system requirements, how to install via the terminal on Mac, Linux, and Windows, and how to log in with your Claude subscription.

3.     The guide covers CLAUDE.md, Claude Code’s project memory system what it is, how to create it with /init, and why it makes Claude significantly more consistent across sessions.

4.  It explains key commands and Plan Mode, showing you how to use /plan for complex tasks so Claude reasons through the problem and shows you its full approach before changing a single file.

5.    A real-world use case shows how a beginner used Claude Code to build and debug a working web app from scratch in a single afternoon without prior experience.

6.  The FAQ section answers the top Google searches around Claude Code pricing, whether it is safe, how it compares to Cursor and GitHub Copilot, and what to do when things go wrong.

What is Claude Code?

Claude Code is a terminal-native AI coding agent built by Anthropic. Unlike AI tools that work through a browser chat window, Claude Code runs directly inside your computer’s terminal and has full access to your local files, folders, and project structure.

The key difference between Claude Code and something like Claude.ai in the browser is action. Claude in the browser discusses your code. Claude Code actually works on it. It can read files, write code, run terminal commands, commit to Git, and execute multi-step tasks from start to finish.

It is available as a terminal CLI, a VS Code extension, a JetBrains plugin, a desktop app, and on the web. For beginners, the terminal CLI is the most powerful and most recommended starting point; it gives Claude direct access to everything in your project.

Prerequisites: What You Need Before You Start

Before you install Claude Code, make sure you have the following in place:

•  Claude subscription: Claude Pro ($20/month) is the right starting point for beginners. It gives you enough usage for everyday coding tasks. Claude Max ($100/month) adds more usage and full access to Opus 4.6.

•  Node.js installed: Claude Code requires Node.js version 18 or higher. Download the LTS version from nodejs.org if you do not have it already.

•   A terminal application: Terminal on macOS, any Linux terminal, or PowerShell / Command Prompt on Windows 10 or 11.

•   Supported operating system: macOS 13.0 or later, Ubuntu 20.04 or later, or Windows 10 (version 1809) or later.

•   4 GB RAM minimum: Claude Code is lightweight locally but communicates with Anthropic’s servers, so a stable internet connection is also required.

Step 1: Install Claude Code

The recommended installation method is the native installer, which auto-updates Claude Code in the background so you always have the latest version without doing anything manually.

On macOS or Linux

Open your terminal and run the following command:

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

This downloads and installs Claude Code automatically. The whole process takes under a minute on a normal internet connection.

On Windows

Open PowerShell (not Command Prompt) and run:

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

If you see an error saying the token ‘&&’ is not a valid statement separator, you are in Command Prompt instead of PowerShell. Switch to PowerShell and try again. Windows users also need Git for Windows installed first download it from git-scm.com if you do not have it.

Verify the Installation

Once the install finishes, confirm it worked by typing:

claude –version 

You should see the Claude Code version number printed in your terminal. If you see a ‘command not found’ error, close your terminal, reopen it, and try again — the PATH update sometimes requires a fresh terminal session.

Step 2: Log In and Start Your First Session

With Claude Code installed, navigate to your project folder in the terminal. Then type:

claude 

On first launch, Claude Code will ask you to choose a text style and then log in. You can authenticate using your Claude Pro or Max subscription, an Anthropic Console API key, or a supported cloud provider like Amazon Bedrock.

Once logged in, you will see the Claude Code prompt. This is your working environment. Type any natural language request and Claude will read your project files, plan its approach, and execute the task. You approve each change before it is applied.

Step 3: Set Up CLAUDE.md: Your Project Memory

By default, Claude Code has no memory between sessions. Every time you close a session and start a new one, Claude starts fresh; it does not remember your tech stack, your coding style, or your project preferences.

CLAUDE.md fixes this. It is a simple text file you create in the root of your project that Claude reads automatically at the start of every session. Think of it as giving Claude a permanent briefing about your project so you never have to repeat yourself.

How to Create CLAUDE.md

The fastest way to create a starter CLAUDE.md is to run the /init command inside Claude Code:

/init

Claude will examine your codebase and generate a tailored CLAUDE.md file automatically. You can then edit it to add anything Claude missed. Here is what a good CLAUDE.md typically includes:

• Tech stack: Your framework, language version, database, and any key libraries.
• Coding conventions: Naming rules, file structure preferences, and style guidelines.
•  Run commands: How to start the dev server, run tests, and build the project.
• Testing requirements: Whether every change needs a test, and which test framework to use.
• Important notes: Anything Claude should always know — authentication patterns, environment variables, or deployment targets.

Treat CLAUDE.md as a living document. Whenever you want Claude to remember something permanently, tell it: ‘Update CLAUDE.md with this.’ Over time it becomes the brain of your entire project.

Step 4: Understanding Key Commands

You do not need to memorise slash commands to be productive with Claude Code. Most of the time, you type plain English and Claude figures out the rest. But a few commands are genuinely useful to know from day one.

• /init Generates a starter CLAUDE.md file by examining your project structure.
• /plan  Activates Plan Mode. Claude analyses the problem, outlines every step, shows its reasoning, and waits for your approval before executing anything.
• /clear  Clears the conversation context. Use this when switching to a different task to keep Claude focused.
• /help Shows all available commands and a quick reference for getting started.
• /model Lets you switch between Sonnet 4.6 (fast, everyday tasks) and Opus 4.6 (complex refactoring and architecture decisions).

A practical rule: use Sonnet 4.6 for about 80% of your work — it is fast and cost-effective. Switch to Opus 4.6 for the remaining 20% that involves complex multi-file refactoring or architectural decisions where depth matters more than speed.

Step 5: Use Plan Mode for Complex Tasks

Plan Mode is one of the most important features for beginners. When you type /plan before your request, Claude shifts into a reasoning mode before touching a single file.

In Plan Mode, Claude reads the relevant files across your project, traces dependencies, identifies everything that will be affected, outlines a step-by-step plan, and waits for your approval. Nothing is executed until you say go.

When to Use Plan Mode

• Complex refactoring renaming a module that is used across many files
• Adding a new feature anything that touches routes, controllers, models, and tests simultaneously
• Exploring an unfamiliar codebase  when you want to understand the impact before making changes
• Debugging a difficult bug let Claude trace the problem and explain its thinking before fixing anything

For simple, single-file tasks like ‘fix this typo’ or ‘add a comment here,’ you do not need Plan Mode. Just type the request directly. Reserve /plan for anything that will touch more than two or three files.

Did You Know?

Using /plan before a complex task is not just safer it actually produces better code. When Claude is forced to reason through a problem before acting, it catches edge cases, avoids breaking changes, and produces more consistent results than jumping straight into execution.

Tips for Getting the Best Results From Claude Code

A few habits make a significant difference in your results, especially as a beginner just getting started.

• Be specific in your prompts. Instead of ‘fix the bug,’ say ‘The login form is submitting but the user is not being redirected. Find the issue in auth.js and fix it.’ Specific prompts produce specific, useful results.
• Use /plan for anything complex. If you are not sure whether a task is ‘complex enough,’ use Plan Mode anyway. The cost is a few extra seconds. The benefit is avoiding a broken codebase.
• Build up your CLAUDE.md over time. Every time you correct Claude or explain something about your project, add it to CLAUDE.md. Within a week, Claude will feel like it has known your project for months.
• Iterate in small steps. Do not ask Claude to build an entire feature in one prompt. Build the core first, review it, then add one piece at a time. This keeps every change reviewable.
• Always review the generated code. Claude Code is powerful, but you are the developer. Read through what was generated, run the tests, and treat every output as a first draft that benefits from your review.
• Switch models intentionally. Use Sonnet 4.6 as your default. Switch to Opus 4.6 with /model when you are tackling something architectural or multi-file and complex. This saves usage and keeps fast tasks fast.

If you’re serious about mastering AI-powered coding tools and want to apply them in real-world scenarios, don’t miss the chance to enroll in HCL GUVI’s Intel & IITM Pravartak Certified Artificial Intelligence & Machine Learning Course, co-designed by Intel. It covers Python, Machine Learning, Deep Learning, Generative AI, Agentic AI, and MLOps through live online classes, 20+ industry-grade projects, and 1:1 doubt sessions, with placement support from 1000+ hiring partners, helping you build intelligent systems and work efficiently with modern AI development tools

Conclusion

In conclusion, Claude Code is not just a smarter autocomplete tool. It is a genuine shift in how you can approach building software from writing a prompt to reviewing a working, multi-file solution in minutes, all without leaving your terminal.

The key is building the right habits from the start. Set up your CLAUDE.md. Use Plan Mode before complex tasks. Be specific in your prompts. Review what Claude generates. These four things alone will make you productive with Claude Code faster than any other approach.

AI-assisted development is already standard practice in modern engineering teams. The sooner you build fluency with a tool like Claude Code, the more leverage you have as a developer — regardless of what you are building next

FAQs

1. What is Claude Code and how does it work?

Claude Code is Anthropic’s terminal-native AI coding agent. It runs inside your computer’s terminal and has direct access to your local project files. You give it instructions in plain English, and it reads your codebase, plans changes across multiple files, and executes tasks autonomously while you review and approve each step.

2. Is Claude Code free to use?

Claude Code is not free. It requires a paid Anthropic subscription. Claude Pro at $20 per month is the recommended starting point for beginners and covers everyday coding tasks. Claude Max at $100 per month provides five times more usage and full access to Opus 4.6 for heavier workloads.

3. Do I need to know how to code to use Claude Code?

Basic programming knowledge helps, but Claude Code is accessible to beginners. It explains what it is doing as it works, handles complex setup tasks automatically, and guides you through each step. Many users with minimal coding experience have successfully built and deployed working applications using Claude Code.

4. How is Claude Code different from Cursor?

Cursor is an IDE-first tool optimised for inline completions and edits inside a visual editor. Claude Code is terminal-first and optimised for autonomous, multi-step tasks that touch many files at once.

5. What is CLAUDE.md and do I need it?

CLAUDE.md is a file you create in your project root that Claude reads at the start of every session. It stores your tech stack, coding conventions, run commands, and testing preferences. Without it, Claude starts every session fresh with no knowledge of your project.