Quickstart

This quickstart takes you from nothing to a connected agent you can talk to, in a few minutes. You’ll install the CLI, authenticate, run setup, and send your first message. Your agent runs on its own persistent server and stays connected to Alfe once setup finishes.

Prerequisites

• Node.js (a current LTS release) and npm.
• An Alfe account. Sign up at app.alfe.ai if you don’t have one yet. (Agents can also self-onboard over MCP — see Agent-led bootstrap.)

1. Install the CLI

npm

npm i @alfe.ai/cli -- -g

pnpm

pnpm add @alfe.ai/cli -g

bun

bun add @alfe.ai/cli -g

Verify the install:

alfe --version

2. Authenticate

Log in to link the CLI to your account:

alfe login

By default this opens your browser to authorize the CLI. In a headless environment, you can pass a pre-issued API key instead:

alfe login --token <your-api-key>

Confirm who you’re authenticated as:

alfe whoami

3. Run setup

alfe setup walks you through connecting an agent on the current machine. It authenticates, installs and configures the agent runtime, starts the local gateway daemon, and registers the Alfe platform tools so the agent can reach memory, model routing, and integrations.

alfe setup

When it finishes, the agent’s daemon is running and connected to Alfe. You can check its health at any time:

alfe doctor

Running on a server?

For non-interactive, cloud-init-style provisioning (for example on a VM), use alfe setup --managed, which reads configuration from environment variables instead of prompting.

4. Say hello

Open your agent in the dashboard at app.alfe.ai and send it a message, or connect a channel like Slack or Discord from the Integrations surface so you can talk to it where you already work.

Next steps

• Learn the core concepts behind the platform.
• Explore the full CLI reference.
• Connect a channel or app from Integrations.
• Let agents drive their own setup with the Agent API and MCP.