Does Copilot send my code to the cloud?

Copilot sends context from your editor to GitHub's service to generate suggestions. Data handling and retention depend on your plan and organization settings, so review GitHub's documentation and your company policy.

Can Copilot write an entire application for me?

It can speed up large parts of the work, especially with its agent features, but you should break work into small steps, review each change, and keep tests as your safety net.

GitHub Copilot: Getting Started for Developers

Introduction

GitHub Copilot is an AI coding assistant that works inside your editor. It suggests code as you type, answers questions in a chat panel, and can help explain, fix, test, and refactor code. Plans, available models, and feature names change over time, so check GitHub's documentation for the current details.

Prerequisites

A GitHub account with Copilot access (individual, business, or organization plan)
An editor with Copilot support, such as Visual Studio Code

Step-by-Step Guide

Step 1: Install and sign in

Install the GitHub Copilot extension for your editor and sign in with the GitHub account that has Copilot access. Once active, you will see suggestions appear as you type.

Step 2: Use inline suggestions

Copilot proposes code as grayed-out text. Accept a suggestion with the Tab key, or keep typing to ignore it. A descriptive function name or a comment explaining the intent gives it better context to work from.

# Return the n most common words in a text, ignoring case and punctuation
def top_words(text: str, n: int = 10) -> list[tuple[str, int]]:
    ...

Step 3: Ask questions in Copilot Chat

The chat panel can explain code, suggest fixes, write tests, and add documentation. Slash commands such as explain, fix, tests, and doc offer shortcuts for common requests, and you can reference files or your workspace to give it more context.

/explain what does this function do and what are its edge cases?
/tests write unit tests for top_words, including empty input
/fix this code raises KeyError on empty input

Step 4: Give the project standing instructions

Repository custom instructions let you describe your conventions once, such as the frameworks, naming style, and testing approach, so suggestions follow them. This is typically a markdown file in the repository's .github folder; check the current documentation for the exact name and options.

Step 5: Review every suggestion

Treat suggestions like code from a new teammate. Read them, run them, and test them before committing. They can be subtly wrong, outdated, insecure, or inconsistent with your codebase, and you remain responsible for what you ship.

FAQs

Does Copilot send my code to the cloud?: Copilot sends context from your editor to GitHub's service to generate suggestions. Data handling and retention depend on your plan and organization settings, so review GitHub's documentation and your company policy.
Can Copilot write an entire application for me?: It can speed up large parts of the work, especially with its agent features, but you should break work into small steps, review each change, and keep tests as your safety net.

Summary

Install Copilot, write clear names and comments to guide suggestions, use chat for explanations and tests, set project instructions, and always review what it produces before you commit.