A Git desktop window shows parallel branches joining through an assisted merge panel
A Git desktop window shows parallel branches joining through an assisted merge panel
+ Microsoft News

GitHub Desktop 3.6 brings Copilot into worktrees and merge conflicts

GitHub Desktop 3.6 adds Git worktrees, Copilot-powered commit authoring, and AI-assisted conflict resolution, moving agent-era Git workflows into the GUI.

GitHub Desktop 3.6 brings worktrees and deeper Copilot integration into GitHub’s desktop client. The June 26 changelog says Desktop now supports Git worktrees, Copilot-powered commit authoring, model selection, bring-your-own-key options, and AI-assisted merge conflict resolution.

The release matters because it moves several agent-era Git workflows into a GUI that many developers already use. Worktrees have become more important as coding agents run isolated branches in parallel. Merge conflicts are still one of the moments where developers drop out of flow. Commit messages are now another place where teams want AI help without losing repository conventions.

Desktop 3.6 puts those pieces in one client.

Worktrees are now a first-class desktop workflow

Git worktrees let a developer keep multiple working directories for the same repository. That makes it possible to work across branches without repeatedly stashing changes, switching branches, or cloning the repository again.

That is useful on its own. It becomes more useful when coding agents are part of the workflow. Agents often work best in isolated branches or worktrees, where a task can run without touching the developer’s current working state. A GUI that can show and manage those worktrees reduces the gap between agent work and normal local development.

GitHub’s changelog frames this as working across branches in parallel. That is exactly the new pressure point. The local machine is no longer only one developer editing one branch. It may be a developer plus one or more agent sessions, each producing changes that need review, merge, or cleanup.

Copilot commit messages get repository context

Desktop’s commit message generation now reads custom instructions from .github/copilot-instructions.md and AGENTS.md, and honors commit metadata rules defined for the repository.

That is a small feature with a practical effect. Generic generated commit messages are easy to ignore. Repository-aware messages are more likely to match the team’s actual standards, especially in repos with strict prefixes, issue references, or body formats.

The Copilot SDK is the shared foundation behind the enhanced commit message experience and the new conflict workflow. GitHub also says every Copilot feature in Desktop now has a model picker, and BYOK can connect a third-party provider or a local model when available through the user’s setup.

That brings Desktop closer to the broader Copilot surface area across CLI and IDE tools.

Merge conflicts get an AI review layer

The most sensitive feature is merge conflict resolution. GitHub says Desktop can explain conflicting changes and suggest a resolution that the user can review, accept, or edit before completing the merge.

That review step is the difference between useful assistance and dangerous automation. Merge conflicts are semantic problems disguised as text problems. The right resolution depends on intent, tests, product behavior, and sometimes release timing. A model can make the conflict easier to understand, but the developer still owns the final decision.

The best use case is not “Copilot fixes my merge.” It is “Copilot explains the conflict and drafts a resolution I can inspect.” That can reduce the intimidation factor without hiding the risk.

Sources

The AI Feed Desk

The AI Feed Desk

Editorial desk

The AI Feed Desk tracks AI provider updates, model releases, agent tooling, and enterprise adoption, turning fast-moving announcements into source-linked context for builders and operators.

Noticed a typo, incorrect information, or translation error?

Tell us so we can fix it.

Help Improve This Article

Related Articles

A usage dashboard shows AI credit consumption bars for individual developers

GitHub adds per-user AI credit metrics for Copilot admins

GitHub's Copilot usage metrics API now reports per-user AI credit consumption, giving enterprise and organization admins a clearer signal for adoption and budget planning.

The AI Feed Desk

By The AI Feed Desk

A routing switchboard sends coding tasks into different model lanes from an admin control plate

Copilot auto model selection turns routing into an admin setting

GitHub Enterprise admins can now default Copilot conversations to auto model selection through managed-settings.json, tying model routing to enterprise AI standards.

The AI Feed Desk

By The AI Feed Desk

A browser-shaped glass panel, image tile, and testing checklist converge on a developer workstation

Copilot browser and vision tools make agents multimodal testers

GitHub made Copilot browser tools in VS Code generally available and opened Copilot vision across plans, giving coding agents browser, screenshot, image, and PDF context.

The AI Feed Desk

By The AI Feed Desk

A pull request review panel connects to Jira cards and a guarded plugin marketplace

GitHub turns Copilot governance into a code-review and Jira story

GitHub's June 25 Copilot updates pair cheaper code-review analysis with Jira agent visibility and stricter controls over plugin marketplaces.

The AI Feed Desk

By The AI Feed Desk

Shared AI credit tokens flow through team lanes while one cost center lane is stopped by a cap gauge

GitHub cost centers can cap shared AI credit pools

GitHub added AI credit pools for cost centers, letting enterprise admins cap how much monthly included Copilot credit one group can draw before metered budgets apply.

The AI Feed Desk

By The AI Feed Desk