GitHub Copilot Agent Now Native in JetBrains AI Assistant

GitHub CopilotView original changelog

GitHub and JetBrains deepened their partnership on June 30, 2026, making GitHub Copilot a first-class, natively integrated agent inside JetBrains AI Assistant's agent picker, rather than requiring the separate ACP Registry setup used previously. Developers can now select GitHub Copilot directly from the agent picker in JetBrains IDEs, choose their preferred model with adjustable reasoning depth, and run multistep coding tasks that iterate autonomously. GitHub says the integration is more stable and requires no manual configuration, with Next Edit Suggestions and a Skills feature planned as upcoming enhancements.

Key Takeaways

  • GitHub Copilot is now built natively into the JetBrains AI Assistant agent picker, replacing the older ACP Registry setup with a one-time OAuth login.
  • The integration supports full agentic workflows: analyzing codebases, proposing edits, running terminal commands, and self-correcting errors across multistep tasks.
  • Developers can adjust reasoning depth per task, trading speed for thoroughness directly from the model picker.
  • Copilot CLI slash commands like /remote and /chronicle now carry over into JetBrains AI chat.
  • Upcoming enhancements include Next Edit Suggestions (NES) for guiding multistep changes and a Skills feature for invoking specialized capabilities.
  • Using the integration still requires an active, separately billed GitHub Copilot subscription, it is not included with a JetBrains AI subscription.

GitHub Copilot Becomes a Native Agent in JetBrains AI Assistant

JetBrains and GitHub announced a deeper integration on June 30, 2026 that makes GitHub Copilot a first-class option directly inside JetBrains AI Assistant's agent picker. The change lets developers choose GitHub Copilot as their entry point for agentic coding without leaving their JetBrains IDE workflow.

From ACP Registry to Native Integration

Previously, connecting GitHub Copilot to JetBrains AI Assistant required going through the ACP Registry, an extra setup step that added friction for developers who just wanted an agent that worked out of the box. With this update, Copilot is built directly into the agent picker menu. Developers open AI chat in their JetBrains IDE, select GitHub Copilot from the agent picker, and complete a one-time OAuth login to connect their GitHub account, no ACP configuration required. GitHub and JetBrains describe the result as a more reliable, thoroughly tested agent experience available by default inside the IDE.

What Developers Get Today

Once connected, GitHub Copilot functions as a full agentic pair programmer inside JetBrains IDEs. It can interpret a request in chat, analyze the surrounding codebase to understand intent, break a task into smaller steps, and execute multistep coding work including proposing edits, running terminal commands, and iterating on its own output when it detects errors. The integration also includes model selection with adjustable reasoning depth, letting developers dial effort up for hard problems or down for quick edits, and it carries over Copilot CLI slash commands like /remote and /chronicle directly into AI chat.

What's Coming Next

GitHub and JetBrains flagged several enhancements still on the roadmap. Next Edit Suggestions (NES) will help guide developers through multistep code changes by proposing the next logical edit as a task progresses. A Skills feature is planned to let the agent invoke specialized capabilities on demand, and further tool-orchestration improvements are planned for handling complex, multi-tool development tasks.

Requirements and Feedback

The integration requires an active GitHub Copilot subscription, it is not bundled with a JetBrains AI subscription. It works alongside the existing GitHub Copilot plugin already available for JetBrains IDEs. GitHub is collecting feedback through in-product channels, its public feedback repository, and a short survey.