Quickstart

This is the smallest useful Bifröst: one machine from your config, mounted over SSHFS, kept alive by the daemon. You need Bifröst installed, plus sshfs and FUSE.

Check that plain SSH works without a prompt. Bifröst runs ssh with BatchMode=yes and never asks for anything, so you need a key or an agent, and the host key already in known_hosts:
Terminal window
```
ssh -o BatchMode=yes sami@agent-01 true && echo ok
```
If this asks for a password or a host key, fix that first (host keys).

Write ~/.config/bifrost/config.toml:

[mount]
root = "~/machines"

[[machines]]
name = "agent-01"
host = "agent-01"
user = "sami"
remote = "/home/sami"

Validate it. No daemon needed:

$ bifrost config check
ok: /home/sami/.config/bifrost/config.toml (1 machines, 0 providers, 1 mounts, root /home/sami/machines)

Start the daemon:
Terminal window
```
bifrostd &
```
For daily use, run it as a systemd user service instead.

Look at what it did:

$ bifrost status
bifrostd 0.1.0  pid 4242  up 12s  /run/user/1000/bifrost/bifrost.sock
config    /home/sami/.config/bifrost/config.toml (ok)
root      /home/sami/machines
machines  1 (1 eligible)   mounts 1/1 mounted, 0 degraded, 0 failed
providers static ok (1)
drivers   rclone ✓ · rclone-nfs ✗ macOS only · sshfs ✓ (default)

$ ls ~/machines/agent-01/

Take it down and bring it back:
Terminal window
```
$ bifrost unmount agent-01
agent-01  unmounted (held; 'bifrost mount agent-01' to resume)
$ bifrost mount agent-01
agent-01  mounted  /home/sami/machines/agent-01 (sshfs, pid 4311)
```
An unmount is a hold: the mount stays down, even across daemon restarts, until you mount it again. Both commands wait up to 60 seconds for the result unless you pass --no-wait.

Where next

Watch it live with bifrost-tui. The keys are in the TUI reference.
Find machines automatically with discovery providers: Tailscale, DNS or an HTTP inventory.
Decide which discovered machines may be mounted with policy.
Something off? Run bifrost doctor and read Troubleshooting.