GitHub Copilot code review
GitHub Copilot code review adds a multi-agent Max level in public preview and GA approvals that count toward required reviews.
Overview
On October 28, 2026 GitHub added two capabilities to Copilot code review. Max, in public preview, is a third review effort level beside Lite and Balanced for higher-risk changes such as authentication logic or shared libraries; it runs several independent review agents across models from different providers and an aggregation agent reconciles their findings into one review, taking longer and using more AI credits. Approvals, now generally available, add an approval assessment to every review and, when an admin enables them, let Copilot approve a PR with that approval counting toward required approvals; new commits dismiss it. Both are available on Copilot Pro, Pro+, Max, Business and Enterprise.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- Max: multiple independent review agents across providers plus an aggregation agent (public preview)
- Review effort levels Lite, Balanced and Max, settable per PR, repository, organization or enterprise
- Approvals (GA): Copilot approval counts toward required approvals, off by default
- Repository admins can restrict which file paths Copilot may approve
- Admins can cap the maximum review effort level
Use cases
- Deeper automated review of auth, security-sensitive or shared-library changes
- Reducing human review load on low-risk pull requests
- Setting organization-wide review policies by effort level
Pricing
Included in Copilot Pro, Pro+, Max, Business and Enterprise; Max uses more AI credits than Lite or Balanced (credit amounts not published).
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- Not listed
- Record updated
- 2 Nov 2026
- Record last checked
- 2 Nov 2026