Kepler Documentation

Getting Started with Kepler

Last updated: June 2026

This page walks you through installing Kepler, completing setup, and creating your first Task.

What is Kepler?

Kepler is GitKraken’s Agentic Development Environment (ADE), built for developers directing multiple AI coding agents across multiple repos in parallel.

Kepler showing the task list on the left, a diff view in the center, and an agent chat panel on the right — Kepler with a Task open. The task list is on the left, the agent chat is in the center, and agent changes on the right.

Install Kepler

Kepler is available for Windows, Mac, and Linux.

Platform	Installer	Minimum OS
Windows	64-bit, ARM64	Windows 10+
macOS	Apple Silicon, Intel	macOS 12+
Linux (x64)	.deb, .rpm, .AppImage	Ubuntu LTS 18.04+ / Debian 10+ / RHEL 8+ / Fedora 39+
Linux (ARM)	.deb, .rpm, .AppImage	Ubuntu LTS 20.04+ / RHEL 8+ / Fedora 39+

Go to gitkraken.com/kepler/download and download the installer for your platform and architecture.
Run the installer and follow the on-screen prompts.
Launch Kepler.

On first open, Kepler displays the Home view. No repos or agents are configured yet. The next two sections cover the minimum setup required before creating your first Task.

Kepler List view on first launch showing an empty task list in the sidebar and a 'No task selected' state in the center panel with a Launch a task button — Kepler on first launch. No tasks exist yet — click **Launch a task** or **+ New task** to create your first one.

Complete the setup checklist

Complete all five items in the Setup checklist before creating your first Task. Access the checklist from the Setup button in the top navigation bar. The checklist shows Setup · N/5 and turns fully green when all five items are complete.

The Finish setting up Kepler checklist showing 5 of 5 items marked DONE and More setup options expanded to reveal Connect a remote environment and Control Kepler remotely — The Setup checklist with all five required items complete and the optional items expanded.

1. Sign in to GitKraken

2. Connect an AI agent

Connect at least one coding agent in Settings → Agents. For supported agents and connection steps, see Agent Integrations.

The Agents section in Kepler Settings showing GitHub Copilot, Cursor CLI, and OpenCode entries each with a Not installed status, an Install button, a Binary path row, and a Re-scan and Custom path option — Settings → Agents. Each agent shows its binary status. Click **Install** or **Custom path…** to point Kepler to an existing installation.

3. Connect issue & PR trackers

Connect your issue tracker and Git hosting provider in Settings → Provider Integrations. For setup steps, see Issue Tracker Integrations and Pull Request Integrations.

The Provider Integrations section in Kepler Settings listing Azure DevOps, Bitbucket, GitHub, GitHub Enterprise, GitLab, GitLab Self-Hosted, Jira, Linear, and Trello, each with a Connect or Reconnect button — Settings → Provider Integrations. Connected providers show a **Reconnect** button; others show **Connect**.

4. Set default folder locations

Open Settings → General and set both paths before creating your first Task.

Default Repositories Folder — where Kepler clones repos. Every Task that references a repo not already on disk will clone it here.
Default Worktrees Folder — where Kepler creates Git worktrees. Kepler creates one worktree per Task per repo.

Click the folder icon next to each field to browse, or type a path directly. The worktree path supports placeholders:

Placeholder	Value
`<REPOSITORY_PATH>`	Main repo folder
`<REPOSITORY_NAME>`	Repo name
`<SOURCE_PATH>`	Files source (main repo, or the source worktree when forking)
`<WORKTREE_PATH>`	The new worktree folder

Commands run inside the new worktree folder using your default login shell, so there’s no need to cd into it and tools configured in your shell profile (e.g. nvm) are available.

Settings → General showing Default Repositories Folder set to /Users/jonathansilva/kepler/repositories and Default Worktrees Folder set to /Users/jonathansilva/worktrees, both highlighted with a teal dashed border, and a Placeholders table below listing REPOSITORY_PATH, REPOSITORY_NAME, SOURCE_PATH, and WORKTREE_PATH — Settings → General. Set both folder paths before launching your first Task.

Expand More setup options in the checklist to see two optional items:

Connect a remote environment — run Tasks on a remote machine or inside WSL. See Remote Environments.
Control Kepler remotely — start Kepler’s local server to access the UI from another device. See Remote Environments.

Create your first Task

A Task is the core unit of work in Kepler. It holds work across one or more repos and contains the worktrees, agent sessions, and changes that belong to a single unit of work.

This section covers creating a Task from scratch. For all three Task creation methods and full option details, see Create a Task.

The + New task button in the Kepler top navigation bar, highlighted with a teal border — Click **+ New task** in the top-right corner to open the Task Launcher.

The Start a task dialog in Kepler showing the Repositories section with an Add repo button, Task Name field, Prompt field, and Agent, Model, Mode, and Effort dropdowns at the bottom — The Task Launcher. Add a repo, name the task, add a prompt, and click **Launch task**.

Click + New task in the top-right corner to open the Task Launcher.
Under Repositories, click + Add repo and select a repository. Kepler clones it into your default repos directory if it is not already present locally.
Enter a Task Name.
(Optional) Enter a Prompt with starting instructions for the agent.
Select an Agent and configure Model, Mode, and Effort as needed.
Click Launch task.

Kepler creates an isolated Git worktree for this Task, starts the agent session, and the Task appears in List or Kanban view.

Have feedback about this article? Did we miss something? Let us know!