Set Up a Development Environment with VS Code, GitHub Copilot and AI Agents

Nowadays, setting up an efficient development environment goes far beyond installing a text editor. With the arrival of Artificial Intelligence agents integrated into code editors, the way we code has changed radically.

In this article we will walk step by step through how to set up a professional development environment using Visual Studio Code, GitHub Copilot with its AI agents, and a Git repository on GitHub. The goal is to have everything ready to work productively, exactly as we are using it right now to write this very article.

This workflow applies to any type of project: from web development, industrial applications, automation scripts, to PLC and SCADA projects where code or documentation is managed.

The header conceptual map shows how the tools work together to take an idea to a finished project:

• VS Code is the command centre — the workspace where your code, files and terminal live.
• GitHub Copilot integrates directly into VS Code and gives access to AI agents (Claude, GPT, Gemini).
• AI Agents execute complete tasks autonomously: they search, edit, run and fix.
• The Integrated Terminal is where agents execute, test and validate.
• Git + GitHub ensures every step is recorded. You can roll back and collaborate.

The flow is cyclical: idea → the agent executes → you review → adjust → repeat. Every iteration is backed up in Git.

Before we begin, we need to have the following ready:

• Operating system: Windows 10/11, Linux or macOS.
• Visual Studio Code: Latest stable version downloaded from code.visualstudio.com.
• Git: Installed and accessible from the command line.
• GitHub account: Required for Copilot and for hosting repositories.
• GitHub Copilot subscription: Individual, Business or Enterprise. A free tier with limitations is also available.
• Internet connection: AI agents require connectivity to function.

Visual Studio Code (VS Code) is a free, open-source code editor developed by Microsoft. It is lightweight, extremely extensible and has become the world's most popular editor for software development.

Installation steps:

1. Download the installer from code.visualstudio.com. Choose the User Installer or System Installer as preferred.
2. Run the installer. During installation, check the following important options:
   • ✔ Add to system PATH (allows running code from terminal).
   • ✔ Register VS Code as the default editor.
   • ✔ Add "Open with Code" to the context menu for files and folders.
3. Complete the installation and restart the computer if prompted.
4. Verify the installation by opening a terminal and running:

If the command returns the version number, the installation was successful and VS Code is available from the command line.

Git is the most widely used version control system. It will allow us to manage the project's change history and synchronise it with GitHub.

Installing Git:

Download and install from git-scm.com. During installation, accept the default options unless you have a specific preference.

Initial configuration:

Once Git is installed, open a terminal and run the following commands to set up your identity:

Set up SSH key for GitHub:

To avoid entering credentials every time you push/pull, it is recommended to set up an SSH key:

Then go to GitHub > Settings > SSH and GPG Keys > New SSH Key, paste the public key and save. To verify the connection:

With Git configured, the next step is to clone the repository where our project lives. This will download a complete copy of the repository including its full history.

Running code . will open VS Code with the project folder as workspace. This is fundamental because:

• The File Explorer (left panel) will show the entire project structure.
• Source Control (Ctrl+Shift+G) will automatically detect the Git repository.
• The Integrated Terminal (Ctrl+`) will open in the project directory.
• GitHub Copilot will have full project context for better suggestions.

Tip: You can also clone directly from VS Code with Ctrl+Shift+P > Git: Clone and paste the repository URL.

GitHub Copilot is the most tightly integrated AI assistant for VS Code. It's not just autocompletion: with Agent mode, it can execute complex tasks autonomously, edit multiple files, run terminal commands and much more.

Step-by-step installation:

1. Open VS Code and access the extensions panel: Ctrl+Shift+X.
2. Search for "GitHub Copilot" in the search box.
3. Install the official GitHub Copilot extension (by GitHub). This extension includes both inline autocompletion and Copilot Chat.
4. Upon installation, VS Code will ask you to sign in to GitHub. Click Sign In and authorise the application in your browser.
5. Once authenticated, verify that the Copilot icon (a double-sparkle symbol) appears in the VS Code status bar at the bottom.

Selecting the AI model:

One of Copilot's most powerful features is the ability to select from multiple AI models. To change the model:

1. Open Copilot Chat: Ctrl+Shift+I (or click the Copilot icon).
2. At the top of the chat panel, click the current model name.
3. A dropdown will show the available models. Choose based on your needs.

As of today (February 2026), GitHub Copilot provides access to multiple AI models that you can select depending on the task at hand:

What is Agent mode?

Agent mode is the most significant evolution of copilots. Unlike simple chat where the user asks and the AI replies, in Agent mode the AI can:

• 🔍 Search the workspace: explore files, find patterns, understand the project structure.
• ✏️ Edit files: create new files, modify existing code, refactor.
• 💻 Run commands in the terminal: install dependencies, run tests, compile.
• 🔄 Iterate: if it finds an error, it analyses and fixes it autonomously.
• 📋 Plan: breaks down complex tasks into steps and executes them sequentially.

In essence, the agent becomes an autonomous programming companion that can carry out tasks from start to finish.

Once we have everything configured, the typical workflow with AI agents is as follows:

Step 1: Open the project

We open VS Code in the project directory. The explorer shows the file structure and Git detects the repository.

Step 2: Start Copilot Agent

We open the Copilot Chat panel with Ctrl+Shift+I. We select the desired model (for example, Claude Opus 4.6 for complex tasks) and enable Agent mode at the top of the chat.

Step 3: Describe the task

We tell the agent what we want to do. The more detailed the description, the better the result. For example:

Step 4: The agent works

The agent will analyse the project, look for patterns in existing articles, create the necessary files, generate images and register everything where it belongs. During the process you can see:

• The searches it performs in the workspace.
• The files it reads to understand the structure.
• The files it creates or modifies.
• The commands it runs in the terminal.
• The questions it may ask if it needs clarification.

Step 5: Review and commit

Once the agent finishes, we review the changes in the Source Control tab (Ctrl+Shift+G). VS Code shows a diff of all modified files. If everything looks good:

Based on the experience of working with AI agents in VS Code, here are some practical tips:

• Be specific in your instructions: The clearer you are with the agent, the better the results. Specify format, style and project conventions.
• Choose the right model: For quick tasks use Sonnet or Haiku. For complex, multi-file tasks, Claude Opus 4.6 is the best choice.
• Always review changes: Agents are powerful but not infallible. Reviewing the diff before committing is essential.
• Use .gitignore: Make sure sensitive files (keys, local configurations) are excluded from the repository.
• Commit frequently: Making small, frequent commits makes it easier to revert changes if something goes wrong.
• Useful additional extensions:
  • GitLens — Advanced Git history visualisation.
  • Prettier — Automatic code formatting.
  • Error Lens — Shows errors inline in the editor.
  • Remote - SSH — Development on remote machines.
• Integrated Terminal: Take advantage of VS Code's terminal maintaining the workspace context. It's more efficient than using an external terminal.
• Essential shortcuts:
  • Ctrl+Shift+I — Open Copilot Chat / Agent.
  • Ctrl+Shift+G — Source Control (Git) panel.
  • Ctrl+Shift+X — Extensions panel.
  • Ctrl+` — Integrated Terminal.
  • Ctrl+Shift+P — Command Palette (search any action).

Setting up a development environment with VS Code, Git and AI agents is a minimal time investment that enormously multiplies productivity. The ability to have an autonomous agent that can search, edit, run and iterate on your code completely changes the way you work.

The workflow described in this article is exactly what was used to create it: an AI agent (Claude Opus 4.6 via GitHub Copilot) analysed the project structure, created the images, wrote the content and registered the article in the corresponding categories. All from within VS Code, with Git backing up every step.

The combination of VS Code + GitHub Copilot + AI Agents is, as of today, one of the most powerful tools for any developer or technical professional.

28 February 2026

All trademarks, software and logos are the property of their respective owners.