Skip to content
worth noting AI agents

GitHub Copilot now available in the JetBrains Air environment, including cloud tasks

clearly official source

GitHub Copilot can now be used directly in JetBrains Air for both local and cloud coding tasks via the Agent Client Protocol. Cloud tasks run in an isolated Docker container and use the existing Copilot subscription, not JetBrains AI credits.

GitHub has announced the integration of the GitHub Copilot product into the JetBrains Air development environment. The integration uses the Agent Client Protocol (ACP) and allows coding tasks to be assigned, tracked, and reviewed both locally in a JetBrains IDE and in the cloud. Alongside the existing GitHub Copilot plugin, users thus get an additional way to work with the agent directly in the Air interface.

For local tasks, Air needs to be installed in the JetBrains IDE, GitHub Copilot selected, and the account connected. Cloud tasks require a JetBrains account within an organization that has Air Teams activated, a GitHub Copilot subscription, and its enablement by an Air administrator; they can be run via the browser or directly from the IDE. According to GitHub, cloud tasks run in a Docker container on a clean virtual machine, all traffic goes through a proxy, and no state is retained between individual runs, so Copilot can build, test, and install packages without touching the user's local environment.

Cloud tasks draw on the existing GitHub Copilot subscription, not on JetBrains AI credits. The source also lists several limitations: the user cannot choose, for an individual task, whether a personal or organizational Copilot plan is used; analytics and management in Air do not include a Copilot usage overview; and Air administrators can disable GitHub Copilot via policy but cannot set spending limits for it within Air.

What changed

Why it matters

Developers using JetBrains IDEs gain the ability to run the GitHub Copilot agent directly within the Air interface and have more demanding tasks (build, tests, package installation) carried out in an isolated cloud environment without risk to the local machine. For businesses, it is important that cloud task consumption is charged against the existing Copilot subscription, but administrators do not have tools within Air for tracking usage or setting spending limits, which needs to be addressed outside this environment.

Two audiences, two different impacts

What this means

01

For individuals

Developers in JetBrains IDEs can now assign coding tasks to the GitHub Copilot agent directly from Air, either locally or in an isolated cloud environment, without risking changes to their own machine.

What to do Try out GitHub Copilot in JetBrains Air for a local task and compare it with the existing Copilot plugin.
More practical updates →
02

For a business

Companies using Air Teams must explicitly enable Copilot cloud tasks through an administrator and must account for the fact that consumption is billed from the existing Copilot subscription, while an overview of usage and the ability to set spending limits within Air are missing.

Development
What to decide Check whether the organization has Air Teams activated and whether it is necessary to set up a policy for an administrator to enable GitHub Copilot, including monitoring subscription consumption outside of Air.
More business impacts →
Agent Client Protocol cloud IDE code assistant Docker GitHub Copilot JetBrains Air

Check the original

Event sources

clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
GitHub Copilot Changelog primary source · first detected GitHub Copilot is now available in JetBrains Air