Skip to main content
Native JetBrains plugin, built for remote development

Kilo turns your JetBrains IDE into a
multi-agent control room

Run parallel coding agents in isolated worktrees, test their changes, follow every diff and PR, and choose from 600+ models without leaving JetBrains.

Kilo Code for JetBrains - Fully native, open-source coding agent built for JetBrains | Product Hunt
Kilo Code Agent Manager coordinating multiple agents in isolated git worktrees inside JetBrains

Built for JetBrains

A native experience, not an editor migration

Install Kilo directly into the JetBrains IDE your team already uses. Its purpose-built interface brings models, pricing, usage, and agent controls into your existing development workflow.

  • Works across IntelliJ IDEA, PyCharm, WebStorm, GoLand, PhpStorm, Rider, and more.
  • Installs through the standard JetBrains Plugin Marketplace.
  • Supports local projects and JetBrains Remote Development through split mode.
  • Keeps your editor, project files, terminal, and development tools in one workflow.
Kilo Code dialog for creating a new worktree or importing one from a pull request or branch in JetBrains

Agent Manager

Run parallel coding agents without leaving JetBrains

Give every task an isolated git worktree, run and debug the results from one IDE frame, and keep the same open-source Kilo runtime across JetBrains and CLI.

  • Continue a conversation in a new worktree without losing task context.
  • Track active, completed, and blocked agents from one JetBrains tool window.
  • Inspect each worktree, run its code, and compare changes before merging.
  • Keep agent work isolated while your main checkout remains available.
Kilo Code Agent Manager showing multiple agents running in isolated git worktrees in JetBrains

Worktree-native workflow

From delegated task to reviewed change

Kilo connects agent execution to the JetBrains tools you already use to run, diagnose, and review code.

Run and debug every worktree

Create JetBrains run configurations for agent worktrees, see live run state, and open a worktree in a new IDE frame when you need the full debugger.

Kilo Code run configurations showing live run status for multiple git worktrees in JetBrains

Review changes before they land

See uncommitted changes, compare each worktree against its base branch, and keep GitHub pull request status beside the task that produced it.

Kilo Code showing GitHub pull request status, changes against the base branch, and uncommitted changes

Keep every agent within reach

Open subagents in separate editor tabs, switch between their conversations, and keep implementation context close to the code each agent is changing.

Multiple Kilo Code subagents open in separate JetBrains editor tabs

Local UI, remote IDE

Native from the start. Split-mode ready by design.

JetBrains split mode lets the IDE backend run on a remote machine while the interface stays responsive on your local machine. Kilo was architected for this model from the beginning, not retrofitted after the plugin was built.

Read the JetBrains split mode documentation
Kilo Code Agent Manager displaying worktrees and session history with a JetBrains split-mode run configuration
Agent Manager in JetBrains with worktree session history and a split-mode run configuration available from the same IDE frame.

Remote machine

IDE backend and project

Run the JetBrains backend, source code, terminal, and development tools close to your remote environment and compute.

Local machine

JetBrains and Kilo interface

Keep the visual interface local, including Kilo's purpose-built agent experience, while working against the remote IDE backend.

Why not ACP?

A custom interface, not a common shell

Many coding agents are available in JetBrains through Agent Communication Protocol (ACP). It provides a common interface that works natively and supports remote environments, but agents appear through the same standardized UI.

We chose not to take the ACP route because Kilo needs a custom interface for product-specific model selection, pricing, usage, features, and agent controls.

Quick comparison

Compare JetBrains AI agents

The right choice depends on whether you prioritize model freedom, first-party JetBrains integration, autonomous tasks, or GitHub-native workflows.

Kilo Code

JetBrains plugin, VS Code, and CLI

Model freedom
Best for
Open-source, model-agnostic agent workflows
Starting price
Free with BYOK
Model choice
600+ models

JetBrains AI Assistant

JetBrains-native assistant

Best for
Built-in chat, completion, and IDE actions
Starting price
JetBrains AI plan
Model choice
JetBrains-managed

Junie

JetBrains-native coding agent

Best for
First-party autonomous coding tasks
Starting price
JetBrains AI plan
Model choice
JetBrains-managed

GitHub Copilot

JetBrains plugin

Best for
GitHub-native teams and autocomplete
Starting price
$10/month base
Model choice
GitHub-managed

Pricing is approximate as of June 2026. Confirm current plans with each vendor.

Installation

Start coding with Kilo in minutes

Install from the built-in Plugin Marketplace. No custom repository or preview channel is required.

Install from your IDE

  1. 1Open your JetBrains IDE.
  2. 2Open Settings > Plugins.
  3. 3Search for "Kilo Code" in the Plugin Marketplace.
  4. 4Select Install and restart if prompted.

The guide covers installation, sign-in, and first-run setup.

Agent workflow

The right agent for each stage of the task

Move from understanding to planning, implementation, and diagnosis without forcing every request through the same workflow.

Ask

Understand unfamiliar code, trace behavior, and get answers grounded in your project without changing files.

Plan

Research the codebase, surface constraints, and turn an idea into an implementation plan before editing.

Code

Implement features and refactors across files with access to your editor, terminal, tests, and project context.

Debug

Investigate failures systematically, form hypotheses, inspect evidence, and fix the root cause instead of guessing.

Model freedom

Choose the model. Keep the workflow.

Match models to the work instead of tying your IDE to one provider. Switch between hosted, bring-your-own-key, and local options without relearning your coding agent.

600+ hosted models

Route across leading and open-weight model families through Kilo Gateway.

Bring your own keys

Connect supported providers directly and keep control of provider access.

Run local models

Use Ollama, LM Studio, and compatible local endpoints for private workflows.

Pay as you go

Use transparent provider pricing with no markup on AI token usage.

One Kilo platform

Standardize the agent layer, not the editor

JetBrains specialists, VS Code users, and terminal-first developers can share one model platform and one set of team controls.

Open source by default

The Kilo Code extensions are MIT-licensed. Inspect how context is handled, audit the implementation, contribute improvements, or adapt the project to your requirements.

Explore Kilo on GitHub

Governance for engineering teams

Add centralized billing, usage analytics, role-based access, model controls, SSO, SCIM, audit logs, support, and SLA commitments as your organization grows.

Model controlsUsage analyticsSSO / SCIMAudit logs

Supported IDEs

Kilo goes where JetBrains developers work

Use the same native plugin across language-specific JetBrains environments.

Also supported: GoLand, PhpStorm, Rider, RubyMine, CLion, and DataGrip. Check the Marketplace listing for current version compatibility.

Trusted by 5M+ developers, from individual builders to enterprise teams

Kilo Code for JetBrains FAQ

Installation, model choice, pricing, and team setup

Bring Kilo into your JetBrains workflow

Install the native plugin today, or talk with our team about rolling Kilo out across your engineering organization.

Native JetBrains plugin Open source Enterprise controls