GitHub Copilot Code Reviews Enter Public Preview for Azure Repos
Microsoft announces the public preview of GitHub Copilot Code Review for Azure Repos, featuring custom instructions, automated workflows, and enhanced cost tracking.
Microsoft has expanded GitHub Copilot Code Review to public preview for Azure Repos with new governance and automation features.
- GitHub Copilot Code Review for Azure Repos has transitioned from technical preview to public preview for all Azure DevOps customers.
- Key updates include support for Managed DevOps Pools, custom instructions for team-specific standards, and automated reviews via branch policies.
- New cost management features allow organizations to track Copilot spend using Azure DevOps project tags and budget alerts.
Overview of the Public Preview Announcement
Microsoft has officially transitioned GitHub Copilot Code Review for Azure Repos into public preview. This update, announced by Dan Hellem, Product Manager for Azure Boards, Repos, and Wiki, removes the requirement for users to sign up for early access. All Azure DevOps customers currently utilizing the service can now activate and configure the AI-assisted code review features within their environments.
The move to public preview follows a technical preview phase where Microsoft gathered feedback to refine the service. The current iteration introduces several administrative, functional, and financial management tools designed to integrate AI code reviews more deeply into enterprise development workflows.
Enhanced Administrative Controls and Onboarding
According to the official announcement, onboarding has been streamlined to offer granular control over feature deployment. Administrators can now manage the enablement of GitHub Copilot Code Review at three distinct levels: organization, project, and repository.
At the organization level, administrators have the option to enable the service globally across all projects. Alternatively, they can delegate this authority to project administrators, who can then decide whether to activate the feature for all repositories within their specific projects. For teams requiring the highest level of precision, individual repositories can be targeted, allowing organizations to pilot the AI tool in specific environments before a broader rollout.
Infrastructure and Managed DevOps Pools
Copilot Code Review relies on Azure Pipelines to orchestrate the review process. While the service defaults to Microsoft-hosted agents, Microsoft has introduced support for Managed DevOps Pools in this preview. Organizations that avoid default agent pools for security or compliance reasons can now configure the service to utilize their managed pools. This setting is applied at the organization level, ensuring all reviews across the entity adhere to the specified infrastructure requirements. It is important to note that self-hosted agents are not supported at this stage of the preview.
Customization and Workflow Automation
One of the primary additions in the public preview is the ability to provide custom instructions to the AI. Recognizing that different teams maintain different coding standards, Microsoft has enabled a system where instructions can be scoped to different levels. These instructions can be applied to an entire organization or project, or narrowed down to a specific repository or even specific file paths within a repository. This allows the AI to provide feedback that aligns with specific architectural patterns or internal practices unique to a codebase.
Automated Review Triggers
To reduce manual overhead, the service now supports automatic code reviews through branch policies. By configuring these policies, a Copilot review is triggered automatically whenever a pull request is created. This ensures that every piece of code is scrutinized by the AI without requiring the developer to initiate the request manually. Furthermore, this feature extends to draft pull requests, allowing developers to receive AI-driven suggestions and identify potential issues during the earlier stages of development.
Financial Governance and Cost Visibility
Usage for Copilot Code Review is billed via a linked Azure subscription and is monitored through the Azure Cost Management dashboard. In response to feedback regarding spend visibility, Microsoft has introduced Azure DevOps project tags to the billing meter. Previously, costs were only visible at the organization level, complicating efforts to attribute spending to specific departments or teams.
With the integration of project tags, administrators can now filter and group costs by project. This functionality supports standard Azure Cost Management features, including budget alerts. Administrators can set thresholds for specific project tags and receive email notifications if the actual or forecasted spend for Copilot reviews exceeds their defined limits.
Upcoming Features and Rollout Schedule
Microsoft has signaled that additional features are expected shortly. This includes the ability to view detailed logs for failed code reviews directly from the pull request interface. Additionally, the service will soon adopt Lite and Balanced review levels, similar to features recently announced for GitHub. These levels will allow teams to choose between a quick, focused review for simple changes and a comprehensive analysis for complex logic or security-sensitive updates.
The rollout of these public preview features is occurring gradually across different geographical regions. Microsoft estimates that it may take two to three weeks, or longer, for the update to reach all organizations. The timing of availability depends entirely on where an organization is hosted, meaning some administrators may see the features earlier than others.
Enjoyed this?
Get more posts like this delivered to your inbox.