How JetBrains AIR Redefines AI‑Powered Development: A Hands‑On Guide

JetBrains AIR (AI IDE from JetBrains) is a new independent IDE built from the ground up with AI as the default workflow rather than an optional add‑on.

Key principles: AI‑first workflow, deterministic and explainable behavior, and developers retain full control over code.

1. Installation and first launch

Currently available only on macOS (Windows/Linux planned for 2026). After installing, sign in with a valid subscription and the welcome page appears.

2. Opening a project

Two ways: open a local folder via Open , or clone a Git repository using Clone from Git by providing the repository URL and destination path.

3. Defining a task

Use the chat‑based AI agent to describe the desired task, optionally providing context such as file paths (e.g., src/orders/service.py). AIR supports a “plan mode” where the AI first generates an execution plan before proceeding.

4. Permission modes

Ask permission – prompts on first use of a tool.

Auto‑edit – automatically accepts file edits.

Plan mode – only analyses code without editing.

Full access – skips all permission prompts.

5. Adding context

Files and folders

Git branches / commits / local changes

MCP servers

Terminal tabs

Upload local files

Code snippets can also be selected and added directly to a task.

6. Running and task management

Task states include Running, Waiting for user action, Finished, and Canceled/Archived. Multiple tasks can run in parallel without interfering with each other.

7. Review and commit changes

In the Review tab, view differences either unified or side‑by‑side, and add comments next to line numbers to provide feedback.

8. Advanced: MCP server integration

Model Context Protocol (MCP) lets AIR connect to external tools such as databases or APIs. Configuration is done by pasting a JSON block into Settings → AI → MCP Servers:

{
  "mcpServers": {
    "alphavantage": {
      "command": "uvx",
      "args": ["av-mcp", "YOUR_API_KEY"]
    }
  }
}

Example shows connecting to the Alpha Vantage API.

Overall, JetBrains AIR combines an intelligent agent with visual interaction, offering a clear three‑step workflow—define task, configure execution environment and permissions, review and commit—while supporting parallel tasks, MCP extensions, and local snapshots for efficient, controllable AI‑assisted development.