This repository has been designed to work seamlessly with GitHub Copilot, an AI-powered code companion tool that can help you learn and write code faster.
GitHub Copilot can be used for FREE in VS Code, Visual Studio and Codespaces, follow the instructions below to get started.
Tip
In addition to Copilot Free tier, students, educators, and open source maintainers also can have a free access to unlimited Copilot Pro.
Go to this GitHub Copilot page and click on the "Get started for free" button to enable GitHub Copilot for your account.
Once you have enabled GitHub Copilot, you need to login to GitHub Copilot in your IDE:
- VS Code: Install the GitHub Copilot extension from the VS Code marketplace. After installing the extension, you will be prompted to sign in to GitHub and authorize the extension to access your account.
- Visual Studio: Follow the instructions in this GitHub Copilot for Visual Studio documentation.
- Codespaces: GitHub Copilot is pre-installed in Codespaces. You just need to sign in to your GitHub account when prompted.
Once you have set up GitHub Copilot, you can start using it to assist you with learning and coding tasks in this repository.
First, open the repository in your IDE, and open a Copilot chat window. In VS Code, you use ⌃⌘I on Mac or Ctrl+Alt+I on Windows/Linux to open the chat.
From there, you attach files or folders from the repository to the chat context, and you can start asking questions about the codebase, such as:
- "Explain me like I'm five what Agent API does, under 100 words."
Copilot chat supports different agents roles that can help you get more specific answers or actions based on your needs. You can switch between agents by clicking on the mode selector in the chat window.
Copilot comes with some built-in agents, such as:
- Agent (default): Default agent is optimized for making autonomous edits across multiple files in your project.
- Ask: Ask agent is optimized for answering questions about your codebase, coding, and general technology concepts.
- Plan: Plan agent researches and drafts plans for a given task without making any changes to your codebase.
In addition to the built-in agents, you can also create custom agents to tailor the experience to your specific needs. For example, we have created a custom agent called "Codebase Explorer" that is optimized for exploring and understanding the codebase of this repository. This mode may also first ask you some questions to better understand your skill level and adapt the answers accordingly.
Try selecting the "Codebase Explorer" agent from the agent selector, and then ask questions like:
- "How does the agent-webapp service communicate with the agent-api service?"
Once you're comfortable with the codebase, you can explore how to extend or customize the sample to fit your own use cases. You can modify existing code, add new features, or integrate with other services or APIs. Agent is the recommended mode for making such changes.
After you're more familiar with Copilot, you can explore more advanced features, such as customizing prompts, instructions or creating your own custom agents to suit your specific needs.
You can also explore the Awesome Copilot repository for community-created reusable prompts, custom agents, and other resources to enhance your Copilot experience.