GitHub Copilot Code Review Gains AGENTS.md Support and UI Enhancements

GitHub has announced a series of enhancements to its Copilot code review capabilities, focusing on deeper integration with project-specific guidelines and improved developer experience. A key new feature is the support for repository-level `AGENTS.md` files. By placing an `AGENTS.md` file at the root of their repository, development teams can now provide explicit instructions and define coding conventions, architectural patterns, or specific feedback requirements for Copilot. This allows the AI to generate more relevant and customized review comments that align with the project's unique standards, moving beyond generic suggestions. This integration means that Copilot code review will automatically read and utilize the context provided in the `AGENTS.md` file, ensuring that its feedback better reflects the repository's established practices and expectations. This is particularly beneficial for maintaining consistency across large teams or complex projects where adherence to specific guidelines is crucial. In addition to the `AGENTS.md` support, the update includes several user interface improvements designed to streamline the code review workflow. One notable enhancement is the easier ability to request a review from Copilot on draft pull requests. A new 'Request' button now appears directly alongside Copilot in the reviewer picker, eliminating the need for developers to manually search for Copilot to initiate an AI review. This makes it more convenient to get an initial AI pass on work-in-progress, allowing for early detection of potential issues before a pull request is even formally published. Furthermore, GitHub has addressed feedback regarding the visual clutter that can occur on pull request timelines due to numerous Copilot review events. To alleviate this, certain Copilot code review timeline events are now collapsed. This decluttering effort helps developers and reviewers to more easily navigate the conversation tab, quickly identify human-generated comments, and focus on the most important discussions and action items within the pull request. These collective updates underscore GitHub's commitment to evolving Copilot into a more configurable and user-friendly tool for modern software development.