Skip to main content
Amulet mounts run on Linux x86_64 and aarch64. The host must provide FUSE and allow the process to create a filesystem mount.
1

Check the environment

Confirm the operating system, architecture, and FUSE device:
The first command must print Linux. If the FUSE check prints nothing, enable FUSE through your Linux distribution or compute provider before continuing.
2

Install the CLI

For a non-root install, the binary is placed in ~/.local/bin. Run the PATH command printed by the installer if that directory is not already available.
3

Export an API key

Create a key in the Amulet dashboard, copy it when it appears, and export it in the shell that will run the mount:
4

Start the mount

The local path can be missing or empty. Amulet creates a missing path:
The command stays in the foreground. Keep it running and use another terminal or process to read and write ./demo.
5

Unmount cleanly

Wait for the mount process to exit before shutting down the host.

​
Read-only mounts

Pass --ro when a process should not change a live volume or branch:
Snapshot mounts are always read-only, even without --ro.

​
Containers and sandboxes

A container or sandbox must expose /dev/fuse and permit filesystem mounts. The exact setting depends on the compute provider. Use a normal host or VM if the environment does not allow FUSE. See Troubleshooting for common mount failures.