Agent Setup

Your agent is ONLINE. Here's how to get set up quickly.

1. Save your credentials

Open your agent and go to Access & Credentials. Save:

Public URL — the https:// address for your agent
Username and Password — used to log in to the runtime UI
Token — used for API access (the exact label varies by runtime)

If the token field still shows "pending", click Refresh from container — it usually populates within a few seconds of the agent coming online.

2. Add secrets your runtime needs

Go to the Secrets tab. Add any API keys or environment variables your runtime requires — for example, your OpenAI key, Anthropic key, or any other credentials.

Secrets are encrypted at rest and injected into your agent container. They're never shown again after saving, but you can update or delete them at any time.

Changes to secrets take effect on the next restart or redeploy.

3. Check the logs

The Logs tab shows startup output and runtime activity. If your agent is online but behaving unexpectedly, logs are usually the fastest way to spot a configuration issue.

4. Try the terminal

The Terminal tab gives you a live shell inside your agent container. It's useful for:

Checking that environment variables are set correctly
Running a quick command to validate your setup
Debugging without leaving the dashboard

5. Optional: connect Tailscale

If you want private SSH access to your agent from your own machine, enable Tailscale from the Tailscale Network card:

Click Connect
If you have a Tailscale auth key saved in your account settings, connection is automatic
Otherwise, open the generated auth link in your browser and approve the device
Once connected, a tailscaleIp (usually 100.x.x.x) appears in the card

From your machine:

ssh <username>@<tailscale-ip>

Use the same username and password from Access & Credentials.

Quick checklist

Credentials saved (URL, username, password, token)
Required secrets added
Logs look clean at startup
Terminal accessible
Tailscale connected (if you need private SSH)