Summary
GitHub shipped an improved Copilot code review experience. The overview comment now groups findings into Open, Resolved since last review, and Previously missed; Copilot auto-resolves its own comments with a Won't Fix or Incorrect reason based on later commits; and accepting an eligible batch of suggestions generates a relevant commit title and optional description.
What changed
Copilot code review now groups its overview findings into Open, Resolved since last review, and Previously missed with per-comment titles; resolves its own comments with an explicit Won't Fix or Incorrect reason inferred from subsequent commits; and generates a commit title and optional description when a user accepts an eligible batch of suggestions (including batches containing non-Copilot comments).
Why it matters
AI code review only saves time if reviewers can trust its state and not re-triage noise across iterations. Tracking how a review evolves, auto-closing addressed comments with reasons, and authoring batch commit messages reduce the bookkeeping overhead of AI review, pushing it from a one-shot commenter toward a maintainable part of the PR workflow.
Evidence excerpt
Copilot now resolves comments with a resolution reason, either Won't Fix or Incorrect, based on your subsequent commits.