alfe setup

alfe setup is the interactive wizard that connects an agent on the machine you run it on. It’s the fastest path from a fresh install to a running, connected agent.

alfe setup

What it does

The wizard runs through four steps:

1. Authenticate. Reuses your existing credentials if you’re already logged in; otherwise it prompts for an API key and validates it.
2. Runtime. Detects a locally installed agent runtime or installs the one your agent is configured to use, initializes the agent’s workspace, and applies the agent’s configuration. A diagnostics helper is installed on a best-effort basis (it won’t fail setup if unavailable).
3. Gateway daemon. Installs the gateway daemon as a system service and starts it, so the agent stays connected across reboots.
4. Platform tools. Registers the Alfe platform MCP server so the agent can reach memory, model routing, and integrations.

When it finishes, the agent is connected. Verify with:

alfe doctor

Options

• --managed — run non-interactively for server/cloud-init provisioning, reading configuration from environment variables instead of prompting. See alfe setup --managed.

Where configuration is written

Setup writes to ~/.alfe/config.toml, including your API key, whether the runtime should auto-start, the active runtime, and the runtime’s workspace path. You don’t normally edit this file by hand.

Re-running setup

alfe setup is safe to re-run. It reuses valid credentials and existing configuration, so you can use it to repair a connection or pick up new agent configuration.