Auto-resolution and analysis updates in Copilot code review
Copilot code review now auto-resolves fixed comments, writes commit messages for its suggestions, and uses shell tools plus multiple agents to check code.
GitHub has updated Copilot code review so it manages its own comment threads and checks its work with a wider set of tools, according to a changelog post.
- When a pushed commit addresses a Copilot code review comment, Copilot now resolves that comment during its rereview, so open threads only show feedback that still needs attention.
- Applying a suggestion from a Copilot code review comment now generates a commit message based on what is changing, instead of the standard autofill message.
- Copilot code review now uses the full set of shell tools from the Copilot SDK, running behind the Copilot agent firewall, to run build commands, run tests, execute targeted scripts and retrieve information from available tools and APIs while validating code under review. GitHub says this led developers to leave more positive feedback and Copilot to surface more high severity findings and fewer nits.
- The Lite effort level now uses an ensemble of agents that each contribute a perspective on the code, combined into a single review. GitHub's testing found this raised the average number of addressed comments per review by 47% for high severity findings, 31% for medium and 11% for low, while cutting review cost by about 8%.
GitHub
Where the world builds software — Git hosting, pull requests, issues, Actions CI/CD, and Copilot, free for public and private repos.
View GitHub →