A Guth Labs publication ↗
Back to News

GitHub previews adaptive code reviews and secret detection

Gold-rimmed lens inspecting layered green glass and metal sheets.
AI-generated editorial illustration

GitHub added an Adaptive effort option to Copilot code review on November 17, letting the reviewer choose Lite, Balanced or Max based on a pull request's complexity and risk. The public preview shows which level was selected and respects the maximum allowed by repository, organization or enterprise settings; Balanced remains the default.

The same release previews cross-repository context for public repositories and secret detection in review comments. Secret checks combine detectors for more than 500 known formats with AI-based detection of credentials lacking recognizable patterns.

Copilot code review for Azure Repos also becomes generally available. Administrators can enable it by organization, project or repository, with usage billed in GitHub AI credits through the linked Azure subscription.

A separate Repository insights page tracks whether comments were addressed, later human change requests and merged work that skipped Copilot review. Its metrics require the Copilot usage metrics policy to be enabled.

Together, the changes address both review depth and follow-through: teams can direct more attention to risky changes and inspect whether AI feedback actually leads to action.

Sources and citations

Revision history

Published .