GitHub’s changelog of 13 November says GitHub Enterprise Server (GHES) support is now generally available in the GitHub Copilot app. It was previously an opt-in preview. Users can add a GHES account to work with remote repositories, issues and pull requests from the app.
The facts
- Version: GitHub Copilot app 1.2.0 or later.
- Setup: sign in first with a GitHub.com or GHE.com account, then go to Settings > Accounts > Add account and choose GitHub Enterprise Server.
- Limitation: cloud sessions are not available for GHES.
What it means
Until now, the most capable AI coding workflows have tended to arrive for cloud-hosted GitHub first. Companies that keep code on a self-hosted server, often for regulatory or contractual reasons, have been last in the queue. This release closes part of that gap: an engineer can use the Copilot app against a self-hosted GitHub instance.
The missing piece matters, though. Cloud sessions are the feature that lets an agent keep working away from a developer’s machine. Without them, GHES teams get the local, supervised version of AI-assisted development, not the hands-off one.
So what
If your organisation is regulated, in healthcare, finance or the public sector, “can we use AI coding tools on our own infrastructure?” is now a question with a partial yes. Before you rely on it, ask your development partner three things: where the code and prompts are processed, which features are unavailable on your setup, and who reviews what the AI produces. The tooling is only half the answer; the process around it is the other half.
BuildApps builds this way for clients with strict data requirements. See our approach to AI-assisted development or healthcare software development, or start a project.