Install
Get Goalrail running in a few minutes. Install the CLI, set up your credentials, and launch your first agent. Once it's running you can add the native macOS app for a full UI — it's the last step of this same flow.
Requirements
- Python 3.12+
- Node.js 22 LTS and npm
- tmux
Install Goalrail
curl -fsSL https://goalrail.dev/install.sh | shRun the same command to upgrade an existing installation.
Set up credentials
goalrail setup
The wizard detects credentials in your environment and prompts for any that are missing. See Models & Credentials for the full reference.
Launch Debby
Try your install with Debby, a built-in multi-AI agent that sends every question to both Claude and GPT and lets them debate.
goalrail debby
This launches a web UI and prints the server URL and port. Ask something
like What came first, the chicken or the egg? and watch the models
challenge each other. See the Debby reference
for more.
Add the macOS app
Prefer a native experience? The macOS app gives you a full UI, session history, and the ability to connect to a local or remote server — on top of the install you just set up. It's optional; everything above already works from the terminal.
Open the downloaded app — it will ask for a server URL. The goalrail debby
command above prints that URL when it launches the web UI. By default it is
http://localhost:6767. If port 6767 is already taken, the server uses the
next available port, so check the web UI for the actual URL (e.g.
http://localhost:6768). Paste that into the desktop app to connect.
Now your desktop app is ready to use. Pick Debby from the agent picker in the UI (along with the other agents we ship) and ask it anything.
This also works with a remote server. See Shared Server to learn how to deploy your own server.