Junie CLI Now Integrates with JetBrains IDEs—No Setup Required

TL;DR

• Junie CLI can now connect directly to running JetBrains IDEs and tap into their full code intelligence
• Auto-detects your IDE, uses your actual build configs and test runners—no guessing, no setup
• Available now in beta for all JetBrains IDEs; Android Studio support coming soon

What Dropped

Junie CLI just got IDE-aware. Instead of operating as a standalone agent that reads files and guesses your project structure, it now connects to your running JetBrains IDE and uses the same indexing, semantic analysis, and code intelligence your IDE already has. Auto-detection means zero manual configuration.

The Dev Angle

This is a meaningful shift. Most AI coding agents work blind—they scan your repo, make assumptions about your build system, and hope the test runner they guess at actually works. That breaks fast on real codebases: monorepos with custom build configs, projects with dozens of modules, test environments your team spent weeks tuning.

Junie now skips the guessing. It sees what you see. It understands the files you have open, the code you've selected, the builds and tests you just ran. When it needs to refactor a symbol, it uses the IDE's semantic index to find every usage across your codebase—not a grep-based approximation. When it runs tests, it uses your pre-configured test runners, not a command it invented.

For complex projects—monorepos, cross-compilation targets, unusual test setups—this is the difference between an agent that works and one that fails silently. Understanding how AI coding agents work helps explain why IDE integration matters this much.

Should You Care?

If you use JetBrains IDEs (IntelliJ, PyCharm, GoLand, etc.) and have a JetBrains AI subscription, this is a straight upgrade. Junie CLI will auto-authenticate and start working immediately. If you bring your own keys (Anthropic, OpenAI), that's fully supported too.

If your project has a straightforward structure and simple build setup, the improvement is incremental. If you work in a monorepo, use custom build tools, or have a complex test matrix, this removes a major friction point. The agent stops failing on assumptions and starts working with your actual setup.

Android Studio support is coming soon. The integration is currently in beta, so expect the feature set to expand based on feedback.

Source: Junie