News

Air Launches as Public Preview – A New Wave of Dev Tooling Built on 26 Years of Experience

Download Air – free for macOS. Windows and Linux versions coming soon.

We hold a principled optimism for agentic software development – and a pragmatic one. After 26 years of building developer tools, we have a clear view of what needs to be built and a strong conviction that agents will fundamentally change how software gets made. But new concepts are emerging faster than anyone can validate them, so we’d rather ship what works than hype what might.

The current state of working with coding agents is fragmented: Each agent runs in a separate tool, with a different setup, different context, and no structural understanding of your code. Air is an important piece in solving that puzzle, and today marks the launch of its Public Preview. It’s available to developers with a JetBrains AI subscription or those with existing subscriptions to agent providers (except Anthropic) and API keys.

A real agentic development environment, not a chat window

JetBrains Air is an agentic development environment for delegating coding tasks to multiple AI agents and running them concurrently. Like IntelliJ IDEA, an IDE, Air is built on the idea of integrating the essential tools into a single coherent experience. But there’s a key difference: IDEs add tools to the code editor, while Air builds tools around the agent. The new development experience is optimized for you to guide the agent and fine-tune its output.

Air helps you navigate your codebase. You can mention a specific line, commit, class, method, or other symbol when defining a task. As a result, the agent gets precise context instead of a blob of pasted text. And when the task is done, your review doesn’t stop at the code diff – Air lets you see the changes in the context of your entire codebase, and you’ll have essential tools like a terminal, Git client, and built-in preview right in front of you. 

Let’s be honest: Complex codebases aren’t yet ready for pure agentic coding. This is where our 26 years of experience building IDEs come into play. Air focuses on agent orchestration without replacing existing development workflows. Air handles the agent-powered development; your IDE handles the rest.

Switch agents freely, run tasks concurrently

Air supports Codex, Claude Agent, Gemini CLI, and Junie out of the box. AI vendors are leapfrogging each other – Air makes switching agents across projects a natural part of the workflow, not a migration. Air supports the Agent Client Protocol (ACP) and will soon add support for other agents available via ACP through the ACP Agent Registry.

Run agents locally by default, or isolate them in Docker containers and Git worktrees for sandboxing and concurrent work.

Air helps you to avoid the mess of having multiple windows and terminal tabs open for each task. You see one task (meaning one agent session) at a time. You’ll get a notification when another task needs your attention, so you can quickly switch to it while the agent keeps working. Air then helps bring your changes from a container or worktree to your main copy.

Getting started

If you have a subscription to JetBrains AI Pro (which is included in the All Products Pack and dotUltimate) or AI Ultimate, all agents are included – just sign in with your JetBrains Account. Prefer to use your own API keys from Anthropic, OpenAI, or Google? You can bring them along! You can also use personal-use subscriptions from Google and OpenAI. If you take the BYOK (Bring Your Own Key) approach, your own keys will always be used first, and any usage not covered by those keys will default back to your JetBrains subscription. A dedicated offering for enterprises is coming soon.

Cloud execution (i.e. remote agent runs in isolated sandboxes) is in tech preview and will be available soon for Air users. 

Next step: Team collaboration

This release focuses on individual developer productivity. At the same time, we see this as a step toward a future where humans and agents collaborate more closely.

One insight we’ve gained from working with agents is that collaboration doesn’t start when reviewing agent output. It starts earlier, when defining the task itself. Teams benefit from refining and aligning on the task together before any agents get involved at all. We’ll share more on this soon.

Download Air, sign in, and start your first task. We listen to your feedback and are constantly using it to improve – join us on X or get in touch directly.