Skip to main content
After Augment Code Review has reviewed your pull request, it will leave a comment summarizing the work to that point. This description saves reviewers time by handling the ‘what’ and ‘how’ of the changes, allowing them to focus on explaining the ‘why’ and the broader context. The result: faster reviews, better collaboration, and more time on building.

​
How It Works

  1. Create Your PR: Open a pull request in GitHub with your code changes
  2. Automatic Analysis: Augment’s Context Engine analyzes your code diff and understands the changes in the context of your entire codebase
  3. Description Generation: A comprehensive PR description is automatically generated, covering what changed, how it works, and potential impacts
  4. Review and Edit: Review the generated description and add any additional context about why the changes were made
  5. Faster Reviews: Reviewers can quickly understand your changes and provide meaningful feedback

​
Best Practices

Review Before Accepting: Always review the auto-generated description to ensure accuracy and completeness. The AI understands your code, but you understand the business context. Add the ‘Why’: Supplement the generated description with the reasoning behind your changes. This helps reviewers understand not just what changed, but why it matters. Keep PRs Focused: Smaller, focused PRs result in more accurate and useful descriptions. Avoid mixing unrelated changes in a single PR. Update for Significant Changes: If you make substantial changes to your PR after the description is generated, consider regenerating or manually updating the description. Use as a Starting Point: Treat the auto-generated description as a foundation. Enhance it with project-specific context, design decisions, and any nuances the AI might miss. Give Reviewers Context: Give reviewers comprehensive context upfront, leading to faster and more thorough reviews. At the same time, help new team members understand changes more quickly with detailed, consistent PR descriptions.