Getting Started with GitHub Copilot Chat

Hey Y'all! VSLive @ Microsoft HQ is happening the first week of August. While I won't be there, my good friend Brian Randell will, and he has a couple of Copilot sessions you won't want to miss. One of the things I'm betting will come up in those sessions is GitHub Copilot Chat.

Now back in February, I wrote a post about "Getting Started with GitHub Copilot" where I introduced you to the completion feature of GitHub Copilot when you add comments to your code. I thought this would be a good time to now introduce you to GitHub Copilot Chat.

Why AI in Coding Is Important
In the rapidly evolving landscape of software development, AI has emerged as a transformative force. AI in coding is revolutionizing how developers approach problem-solving, significantly enhancing productivity and enabling the creation of more robust and efficient code. By automating repetitive tasks and providing intelligent code suggestions, AI reduces the cognitive load on developers, allowing them to focus on more complex and creative aspects of their projects. This integration of AI tools not only accelerates the development process but also fosters innovation, making it easier to experiment and iterate rapidly.

What Is GitHub Copilot Chat
GitHub Copilot Chat is a chat interface that lets you interact with GitHub Copilot, to ask and receive answers to coding-related questions within and supported IDEs (such as Visual Studio Code and Visual Studio 2022). While previously with GitHub Copilot, we were adding comments to our code and GitHub Copilot was then making code suggestions, GitHub Copilot Chat gives us a chat interface for working with GitHub Copilot.

This tool goes beyond simple autocomplete functionalities. GitHub Copilot Chat can generate entire code snippets, provide explanations for specific pieces of code, and even help debug errors. It acts as a conversational partner, making the coding process more interactive and intuitive. Whether you're trying to implement a new feature, understand an unfamiliar codebase, or fix a bug, Copilot Chat is designed to assist you at every step. Its ability to interpret natural language queries means you can ask for help in plain English, making it accessible to developers of all skill levels.

GitHub Copilot Chat takes the information you enter in the chat window, along with contextual information from the environment, such as the code selected or the tabs that are open in your IDE, and uses this information to help provide answers to your questions.

Getting Started with GitHub Copilot Chat
To allow you to compare and contrast using GitHub Copilot code completion vs. GitHub Copilot Chat, I thought we would take the same example from the February article, but use GitHub Copilot Chat to build it. So we are going to create a simple Python project, in VS Code, to create a function to calculate the factorial of a number.

You will need the following if you want to try this yourself:

  • VS Code installed
  • GitHub Copilot extension installed in VS Code
  • GitHub Copilot Chat extension installed in VS Code
  • An active GitHub Copilot subscription

Let's run through the following steps:

  1. Open VS Code and select the GitHub Copilot Chat extension. This will open a pane in VS Code with a text box for you to start chatting with GitHub Copilot. GitHub Copilot Chat has a variety of slash commands you can use, as shown in Figure 1.

    Figure 1: Slash Commands
    [Click on image for larger view.] Figure 1: Slash Commands

    With these commands you can ask GitHub Copilot to explain code to you, fix errors in your code, or even start a new project for you.

  2. We are going to use the /new command to create a new workspace for our python application. Figure 2 shows you the information I typed into the chat window, and the GitHub Copilot response.

    Figure 2: Create a Python Application
    [Click on image for larger view.] Figure 2: Create a Python Application

    You can see the workspace it is proposing to create for me. This looks good, so I click Create Workspace. It prompts me on where to create it, and then it creates the and files.

  3. Now here is where things get cool. First off, it creates the file, shown in Figure 3.

    Figure 3: Creating the File
    [Click on image for larger view.] Figure 3: Creating the File

    This looks very similar to what GitHub Copilot code completion did for us in February.

    But what is also cool is that it creates a file, explaining what this project and code do, partially shown in Figure 4.

    Figure 4: Creating File
    [Click on image for larger view.] Figure 4: Creating File
  4. Now, maybe I don't quite understand the code in Figure 3. I can highlight the code, then back over in GitHub Copilot Chat, I can ask it to explain the selected code (Figure 5).

    Figure 5: Explain the Selected Code
    [Click on image for larger view.] Figure 5: Explain the Selected Code

    And check it out, it not only explained the code to me, it also suggested I add input validation to handle invalid inputs, like negative numbers, and even provided me the updated code for that scenario.

This is just a simple introduction to GitHub Copilot Chat, using only a couple of basic commands. I've used this to create entire applications in programming languages that I have no experience with. And just wait until you start running into error messages that you don't understand. I've been able to feed the error message to GitHub Copilot Chat, and it has quickly come back with suggestions to resolve the error messages.

The more context you can provide GitHub Copilot Chat, the better it will do in answering your questions. For example, it will use all the information you have previously entered in the current chat, as well as the tabs you have open in VS Code.

GitHub Copilot Chat represents a significant leap forward in the realm of AI-assisted coding. By providing real-time, context-aware assistance directly within VS Code, it transforms the development experience, making it more efficient and enjoyable. This tool not only enhances productivity but also fosters a deeper understanding of coding practices, thereby improving the quality of the code produced.

As AI continues to evolve, tools like GitHub Copilot Chat will become increasingly integral to the software development process. They empower developers to tackle more complex problems with confidence, streamline the coding workflow, and ensure adherence to best practices. Embracing these advancements will undoubtedly pave the way for more innovative and effective software solutions in the future. Whether you're a seasoned developer or just starting, integrating GitHub Copilot Chat into your workflow can significantly enhance your coding capabilities and efficiency.

Happy Coding!

Posted by Mickey Gousset on 06/24/2024

Keep Up-to-Date with Visual Studio Live!

Email address*Country*