> ## Documentation Index
> Fetch the complete documentation index at: https://docs.luumen.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Setting up your first SSH connection

> Add your first host in Luumen, connect over SSH or WinRM, and ask LuumenAI your first question — all in under five minutes.

This guide walks through the first-run flow in Luumen: completing onboarding, adding a host, opening a session, and trying LuumenAI. By the end you'll have an active terminal and know how to use LuumenAI with live server context.

## Before you start

Make sure you have:

* **Luumen installed and signed in.** See [Installing Luumen](/get-started/installing-luumen).
* **A host you can reach** over SSH or WinRM, including its hostname or IP.
* **Credentials** for that host — either a username/password or a username/SSH key.

## Get connected

<Steps>
  <Step title="Complete onboarding">
    When you sign in for the first time, a short onboarding flow walks through your account preferences. You'll be taken to the **Add your first host** screen.
  </Step>

  <Step title="Add your first host">
    <Frame>
      <img src="https://mintcdn.com/luumen/ql0wFCz4fcCU2RtM/images/connect-hosts-quick.png?fit=max&auto=format&n=ql0wFCz4fcCU2RtM&q=85&s=144d9f34d0d54bc88b10c79e6fd6ef1c" alt="The Add your first host screen in Luumen with a field for hostname or IP address" width="1292" height="796" data-path="images/connect-hosts-quick.png" />
    </Frame>

    On the **Add your first host** screen, enter the following:

    <Frame>
      <img src="https://mintcdn.com/luumen/ql0wFCz4fcCU2RtM/images/connect-hosts-quick2.png?fit=max&auto=format&n=ql0wFCz4fcCU2RtM&q=85&s=7e9cc102c11fa7f11c6a99dc204b51f1" alt="The Connection details screen in Luumen showing display name, connection type, port, and credential fields" width="1292" height="796" data-path="images/connect-hosts-quick2.png" />
    </Frame>

    1. **Hostname or IP address** — for example, `prod-web-01.example.com` or `10.0.1.42`.
    2. **Connection type** — choose **SSH** (Linux) or **WinRM** (Windows).
    3. **Credentials** — either:
       * **Password** — enter username and password.
       * **SSH key** — enter username, then upload or paste your private key. Luumen stores the key securely in the cloud so you can use it from any device.

    <Tip>
      If you use SSH keys, set them up once in Luumen and they'll be available across every device you sign in on.
    </Tip>
  </Step>

  <Step title="Connect to the host">
    Click **Connect**. Luumen opens your terminal as a widget in the current tab. The **AI Chat** widget opens alongside it automatically, ready to help with the host you're connected to.

    <Frame>
      <img src="https://mintcdn.com/luumen/ql0wFCz4fcCU2RtM/images/connect-session1.png?fit=max&auto=format&n=ql0wFCz4fcCU2RtM&q=85&s=0c385157a09c3d323ff9440820543f5e" alt="An open Luumen session with a terminal widget and the LuumenAI chat panel alongside it" width="1292" height="796" data-path="images/connect-session1.png" />
    </Frame>
  </Step>

  <Step title="Try LuumenAI">
    With the session open, ask LuumenAI a question about the host like:

    ```text theme={null}
    Why is memory usage so high on this server?
    ```

    <Frame>
      <img src="https://mintcdn.com/luumen/ql0wFCz4fcCU2RtM/images/ai-propose-command.png?fit=max&auto=format&n=ql0wFCz4fcCU2RtM&q=85&s=518ed14c971417a88d93f5f3bf425fb2" alt="LuumenAI answering a question about memory usage and proposing a command to run on the host" width="1292" height="796" data-path="images/ai-propose-command.png" />
    </Frame>

    LuumenAI reads the live state of the host and answers based on what it actually finds — not generic advice. Try a few questions to get a feel for how it handles real environment context.

    <Note>
      LuumenAI's context is scoped to the host you're currently connected to. Switch to a different host and it answers about that one instead.
    </Note>
  </Step>
</Steps>

## What to try next

Now that you've got a connection working, a few useful follow-ups:

* **Ask harder questions.** "What's the current disk usage on this server?" "Are there any failing services right now?" "Has this server been patched recently?"
* **Approve a remediation command.** When LuumenAI proposes a fix, read its explanation and approve it to run on the host.
* **Add more hosts.** From the Hosts area, add the rest of the servers you work with. SSH keys carry over automatically.
* **Try a Skill.** [Skills](/guides/luumenai-skills) turn runbooks into reusable AI procedures. If your team already has SOPs, those are natural candidates.

## Need help?

* Browse the [Luumen help center](https://support.luumen.ai/en/) for articles on connecting, billing, and team management.
* Running a Luumen-managed agent across a fleet for scheduled compliance scans, SAP monitoring, or integrations is part of [Luumen Enterprise](/enterprise/get-started/introduction).