How to Start Using Claude Code in Less Than 10 Minutes

You've heard people talk about having AI write code, browse the web, or manage their files. It sounds powerful, but also complicated. Where do you even begin if you're not a software developer?

The answer is simpler than you might think. You can get Claude Code, a tool that lets you work with an AI assistant directly on your computer, running and doing useful work in less time than it takes to make a cup of coffee. You don't need to write a single line of code yourself. This guide walks you through the very first steps.

What Exactly Is Claude Code?

Think of Claude Code as an AI assistant that you talk to in a special window on your computer, called a terminal. Unlike chatting with a chatbot in your web browser, Claude Code can directly interact with the files and programs on your machine [1].

It can read documents you have saved, create new files, edit text, run simple scripts, and even connect to other online services you use, like your calendar or email [1]. Because it works in this dedicated space, you can give it a series of tasks and it will carry them out step-by-step, learning about your preferences as it goes.

The goal isn't to turn you into a coder overnight. It's to give you a helper that can handle repetitive, text-based work so you can focus on the parts of your job that require a human touch.

Your First Two Minutes: Installation

The quickest way to start is by using the Claude desktop application. If you don't have it, you can download it from the official Anthropic website. Once installed, you'll find an option to open Claude Code.

Alternatively, if you're comfortable with a command line, you can install the Claude Code CLI (Command Line Interface). The official documentation provides a one-line command to set it up, which typically completes in about two minutes [1]. You'll then type claude in your terminal to start a session.

Whichever method you choose, you'll be prompted to log in with your Claude account. A paid Claude Pro or Max plan is required to use Claude Code.

Minute Three: Understanding Your First Conversation

When you first start Claude Code, you'll be in a chat interface. This is where you'll talk to the AI. You can ask it to do things like "list the files in this folder" or "create a new text file called ideas.txt".

At the beginning, Claude Code will ask for your permission before it takes certain actions, like running a script or installing something. This is a safety feature. For your first few sessions, it's fine to keep these prompts on so you can see what's happening. Later, you can switch to a more automatic mode where it proceeds with safe actions without stopping to ask each time [2].

Minutes Four to Seven: Creating Your CLAUDE.md File

Here is the most important step for a smooth experience. Instead of explaining your entire workflow from scratch every time you chat, you can write it down once in a special file.

Create a plain text file named CLAUDE.md in the folder where you want to work. This file is automatically read by Claude Code at the start of every session in that directory [3]. You can write instructions in simple English.

What should you put in it? Start with the basics of how you like to work. For example:

## My Preferences - Please write in clear, concise English. - Always create a backup before editing an important file. - When researching, prefer official documentation sources.

You don't need to write a novel. The point of the CLAUDE.md file is to be a reference guide, like notes you'd give a new assistant. It saves you from repeating the same instructions day after day.

You can have a global CLAUDE.md in your home directory for rules that apply everywhere, and specific ones in project folders for job-specific details.

Minute Eight: Trying a Simple Task

Now, put it to work with something concrete. Navigate to a folder where you keep some documents, maybe a "Projects" or "Downloads" folder. Ensure your CLAUDE.md file is there.

In the chat, type a simple request. For example:

"Please look at all the text files in this folder and create a new file called 'summary.txt' that lists their names and how many lines are in each."

Watch as Claude Code reads the files, counts the lines, and assembles the new document. You've just automated a small, tedious task.

If you want to see what it plans to do before it does it, you can use the /plan command. This is especially useful for more complex requests. It will outline its steps, you can give feedback, and then it will execute. This planning mode is read-only and helps avoid mistakes [4].

Minute Nine: Enabling Memory

One of the most useful features for beginners is memory. You can tell Claude Code to remember what it learns about your projects across different chat sessions. Just type /memory to enable it.

With memory on, Claude Code will write down key facts and decisions in the background. The next time you work on the same project, it can recall that context. For instance, if you tell it "I prefer all blog post titles in Title Case," it can remember that rule for future sessions [5]. This means you spend less time re-explaining and more time getting things done.

Minute Ten: Where to Go Next

You now have a working assistant. Where do you go from here?

1. Build a Habit: Use it for one small thing each day. Organize a folder, rename a batch of files, or draft an email. Repetition builds familiarity.
2. Create a Skill: If you find yourself typing the same multi-step request more than twice, you can save it as a skill. A skill is a packaged workflow that you can trigger with a short command, like /organize-files. You create one by simply asking Claude Code at the end of a successful task: "Please save this series of steps as a skill called 'organize-files'" [6].
3. Connect a Tool: Claude Code can use connectors to interact with apps like Gmail, Google Calendar, or GitHub. These are set up in your main Claude account settings and become available in your Code sessions [7]. Start with one you use daily.
4. Explore Safely: There is a community of users who share pre-made skills for different tasks. Be cautious and only install tools from trusted sources, as they can run commands on your computer [8]. Start with the official documentation and examples.

The biggest barrier for most people isn't the technology; it's knowing what to do in the first ten minutes. You don't need to understand everything. You just need to start the conversation. Install the tool, create your CLAUDE.md file with a few simple rules, and ask it to do one helpful thing. The rest you can learn gradually, one task at a time.

References

1. Anthropic. “Extend Claude Code, features overview.” Claude Code Documentation. https://code.claude.com/docs/en/features-overview
2. Anthropic. “Configure auto mode.” Claude Code Documentation. https://code.claude.com/docs/en/auto-mode-config
3. Anthropic. “Give Claude context: CLAUDE.md and better prompts.” Claude Help Center. https://support.claude.com/en/articles/14553240-give-claude-context-claude-md-and-better-prompts
4. Anthropic. “Claude Code power user tips.” Claude Help Center. https://support.claude.com/en/articles/14554000-claude-code-power-user-tips
5. Anthropic. “How Claude remembers your project.” Claude Code Documentation. https://code.claude.com/docs/en/memory
6. Anthropic. “Extend Claude Code, features overview.” Claude Code Documentation. https://code.claude.com/docs/en/features-overview
7. Anthropic. “Use connectors to extend Claude's capabilities.” Claude Help Center. https://support.claude.com/en/articles/11176164-use-connectors-to-extend-claude-s-capabilities
8. Anthropic. “Hooks reference.” Claude Code Documentation. https://code.claude.com/docs/en/hooks