Getting Started
Maestrly is a desktop app for macOS, Windows, and Linux that acts as a command center for your AI coding agents. It orchestrates the AI coding CLIs you already use — in one place — with Kanban boards, built-in terminals, code review, and more. Maestrly never includes or resells access to any AI model: you bring your own subscription with each provider, and your code and credentials stay on your machine.
This guide takes you from install to your first conversation.
1. Install the app
Download the Maestrly installer for your operating system (macOS, Windows, or Linux) and follow the standard installation steps. The first time you open it, the app starts a short welcome tour.
2. Sign in
Maestrly signs you in with Google or GitHub. On the welcome screen, choose "Sign in with Google" or "Sign in with GitHub", authorize access, and you'll be returned to the app. No extra password is created — you use the social login of your existing account.
3. Set up your AI coding CLIs
Maestrly drives the AI coding command-line tools that run on your machine — bring your own. Open Settings (the gear icon in the sidebar): Maestrly automatically detects which supported CLIs are installed on your system. Make sure the tools you want to use are installed and signed in with their own credentials before you start an agent. Each tool keeps its own credentials in your OS keychain; Maestrly only orchestrates them.
4. Create a workspace
A workspace is a Git repository — the project the agents work on. Open the sidebar, click Add workspace, and point it at your project's folder. Each conversation you start is born on its own isolated branch/worktree, so different agents don't step on each other's work.
5. Start your first conversation
Inside a workspace, create a new conversation. Pick which CLI will handle it and the branch to work on; Maestrly launches that tool's terminal UI and you start chatting with the agent. The agent works directly on your project's files, on your computer. You can run several conversations in parallel, each isolated on its own branch.
6. Organize work with boards
Every workspace has a Kanban board. Create cards for tasks, move them between columns, and configure "agent" columns that automatically start a new conversation when a card arrives. It's a practical way to build a pipeline of AI-assisted development — produce, review, and ship.
Next steps
- Concepts — understand workspaces, conversations, handoff, and how notes, memory and delivery search differ.
- Features overview — tour the conversation drawer and the advanced toolkit.
Stuck? Visit Support for contact options, including reporting problems straight from the app.