Skip to content

Latest commit

 

History

History
73 lines (50 loc) · 3.97 KB

copilot-cheat-sheet.md

File metadata and controls

73 lines (50 loc) · 3.97 KB

GitHub Copilot Cheat Sheet

Boost your productivity with these essential keyboard shortcuts for GitHub Copilot in Visual Studio Code:

General Shortcuts

  • Trigger Copilot Suggestions: Ctrl + Space (Windows/Linux) or Cmd + Space (macOS)
  • Accept Suggestion: Tab
  • Reject Suggestion: Esc
  • Open Chat View: Ctrl + Shift + I (Windows/Linux) or Cmd + Shift + I (macOS)
  • Open Inline Chat: Cmd + I (macOS) or Ctrl + Alt + I (Windows/Linux)

Navigation

  • Cycle Through Suggestions: Alt + ] (Windows/Linux) or Option + ] (macOS)
  • Cycle Backward Through Suggestions: Alt + [ (Windows/Linux) or Option + [ (macOS)

Agent Feature

  • Open Command Palette: Ctrl + Shift + P (Windows/Linux) or Cmd + Shift + P (macOS)
  • Activate Copilot Agent: Search for "Copilot Agent" in the Command Palette.

Editing

  • Inline Suggestions: Start typing and Copilot will suggest inline completions.
  • Block Suggestions: Write a comment describing the code you need, and Copilot will generate a block of code.

Using Slash Commands in Copilot Chat

  • Accessing Commands: Type / in the Copilot chat to see a list of available commands.
  • Code Generation: Use commands like /generate to request specific code snippets or solutions.
  • Explain Code: Use /explain to get a detailed explanation of a selected piece of code.
  • Refactor Code: Use /refactor to request improvements or optimizations for your code.
  • Debugging Assistance: Use /debug to ask for help identifying and fixing issues in your code.

Using # Commands in Copilot Chat

  • Inline Instructions: Use # commands to provide inline guidance or context for Copilot. For example, # TODO: Add error handling can help Copilot generate relevant suggestions.
  • Annotations: Use # to annotate code and clarify its purpose, improving Copilot's understanding of your intent.
  • Command Menu: When you type # in Copilot Chat, a menu may appear with options for inline commands or annotations. These options allow you to:
    • Add TODOs or refactoring notes.
    • Provide context for specific code blocks.
    • Guide Copilot's behavior with targeted instructions.

Writing Good Prompts

  • Be specific: Provide detailed instructions to guide Copilot effectively. For example, instead of "Write a function," say "Write a function to calculate the sum of an array."
  • Use context: Reference existing code or describe the desired outcome clearly.
  • Iterate: Refine your prompt if the initial suggestion isn't perfect.

Best Practices for Using GitHub Copilot

  • Choose the Right Tool: Use Copilot code completions for repetitive tasks and Copilot Chat for answering questions or generating large sections of code.
  • Create Thoughtful Prompts: Be specific and break down complex tasks into smaller, manageable prompts.
  • Validate Suggestions: Always review Copilot's suggestions for accuracy, security, and maintainability.
  • Provide Context: Open relevant files and close irrelevant ones to help Copilot focus on the task at hand.
  • Stay Updated: Regularly check for new features and improvements in Copilot.

Writing Tests with GitHub Copilot

  • Unit Tests: Use Copilot to generate unit tests for individual functions or methods. Provide clear prompts and review the generated tests for edge cases and completeness.
  • Integration Tests: Leverage Copilot to write tests that verify interactions between components. Use mocks to simulate external systems.
  • Slash Commands: Use the /tests command in Copilot Chat to quickly generate test suites.
  • Provide Examples: Open existing test files to give Copilot context about the testing framework and style.

Tips

  • Use comments to guide Copilot's suggestions.
  • Combine Copilot with other extensions for enhanced workflows.

Further Resources