Installing Selu

This guide walks you through everything — from choosing where to run Selu to sending your first message to an AI agent. No prior experience required. You’ll be chatting in about 10 minutes.

What you’ll end up with

Selu running on your own machine
A language model connected
An AI agent installed and ready
Your first conversation going

Pick your device

Selu runs on macOS, Linux, and Windows (via WSL2). You need a machine with:

At least 2 GB of free RAM
At least 5 GB of free disk space
An internet connection

Any modern laptop or desktop will work. If you plan to run local language models later, you’ll want more power — but that’s not needed for this guide.

Supported platforms

Install Docker

Selu runs inside Docker containers. If you already have Docker installed and running, skip to Open your terminal.

Install Docker Desktop from docker.com/get-docker. Follow the installer, then start Docker Desktop and wait until it shows it’s running.

On Windows, you’ll also need WSL2 — Docker Desktop will prompt you to enable it during setup.

To verify Docker is ready, open a terminal and run:

docker --version

You should see a version number. If you get an error, make sure Docker is running.

Open your terminal

Press Cmd + Space, type Terminal, and hit Enter.

Empty terminal

Copy the install command

Select and copy this command:

bash <(curl -fsSL https://selu.bot/install.sh)

Paste and run it

Paste the command into your terminal and press Enter.

Install command in terminal

Wait for the installer to finish

The installer will now:

Check that Docker is running
Download the Selu container images (this takes a minute or two the first time)
Generate your local configuration
Start all services

Installer progress

When it’s done, you’ll see a message telling you Selu is ready.

Installer done

Open Selu in your browser

Open your browser and go to:

http://localhost:3000

You’ll see the Selu welcome screen.

Selu welcome screen

Create your account

Pick a username and password, then click Create account.

This is stored locally on your machine — nothing is sent anywhere.

Create account form

Log in

After creating your account, you’ll be logged in automatically. You’ll land on the Chat screen — it’ll be empty for now. That’s normal.

Empty chat screen

Go to Providers

Click Providers in the left sidebar.

Sidebar — Providers

Click “Add Provider”

You’ll see a list of language model providers that Selu supports.

Pick a provider and enter your API key

Choose one of the providers below. If you don’t have an API key yet, follow the link to create one — it only takes a minute.

Select OpenAI from the list and paste your API key.

Don’t have one? Sign up at platform.openai.com and create an API key under API keys in your account settings.

Selu will run a connection check automatically. You should see a green checkmark or success message.

Provider connected

Go to Agents

Click Agents in the left sidebar.

Sidebar — Agents

Switch to the Marketplace tab

At the top of the Agents page, click the Marketplace tab. You’ll see a collection of ready-made agents.

Marketplace tab

Install the General Assistant

Find the General Assistant agent and click Install. It’s a great all-rounder for getting started.

Install General Assistant

Selu will download and configure the agent. After a few seconds, it appears on the Installed tab.

Agent installed

Go to Pipes

Click Pipes in the left sidebar.

A pipe connects your chat to an agent — think of it as the line between you and your AI.

Sidebar — Pipes

Click “Add Pipe”

Select “Web UI”

This lets you chat directly in the Selu dashboard — the quickest option.

Select Web UI

Name your pipe and create it

Type a name (like “My Chat” or “General”) and click Create Web Pipe.

Create Web Pipe

Go to Chat

Click Chat in the left sidebar. You’ll see your new pipe listed.

Chat with pipe listed

Start a new conversation

Click New conversation.

New conversation button

Send your first message

Type something like “Hello! What can you do?” and hit Send.

Type first message

You’re chatting with your agent!

Your agent should respond within a few seconds. Congratulations — you have an AI agent running on your own machine.

First conversation

What happens behind the scenes

When you send a message, here’s what Selu does:

Your message goes to the Selu server running on your machine
The server passes it to your installed agent
The agent sends the conversation to your language model provider (e.g., OpenAI)
The response comes back through the agent to your chat window

No third party sees your conversation — except the language model provider if you’re using a cloud-based model. Everything else stays on your machine.

Troubleshooting

Problem	What to do
Installer says Docker isn’t running	Open Docker Desktop (Mac/Windows) or run `sudo systemctl start docker` (Linux)
Port 3000 is already in use	Run the installer with `--port 3001`
Can’t reach localhost:3000	Check your firewall. On macOS, check Docker Desktop’s resource settings
Provider connection check fails	Double-check your API key — no extra spaces, full key copied
Agent doesn’t respond	Make sure a provider is connected and the agent is installed. Check logs with `docker compose logs -f`

What’s next?

Explore your conversations

Learn how multi-turn conversations and capabilities work in Your First Conversation.

Chat from your phone

Set up Telegram, iMessage, or WhatsApp so you can message your agent on the go.

Try more agents

Browse the Marketplace for specialized agents — writing assistants, research helpers, and more.

Build your own agent

Ready to go deeper? The Developer Guide shows you how to create custom agents.