wanctl Docs

Make a machine remotely controllable (Mac / Linux)

Run three lines on the machine you want to control — install the tool, set the instance addresses (once), sign in and join:

curl -fsSL https://github.com/Daily-AC/wanctl/releases/latest/download/install.sh | sh
wanctl config set relay=https://wanctl-relay.z10.dev portal=https://wanctl.z10.dev
wanctl start

A browser opens the portal login (GitHub account) and shows a one-time code. Paste it back into the terminal and press enter. The service then moves into the background: wanctl stop stops it, wanctl status shows how it is doing. Installing needs no token at all.

Signing in to the portal at all assumes you are already part of this deployment: the first user to log in becomes the administrator, and everyone after that needs an invite code from an administrator, redeemed after login (see "Invites, friends and sharing").

You can skip the second line and run wanctl start on its own — the first run asks in the terminal which relay to use, this deployment's or the project's hosted instance, and saves your answer. wanctl config shows and changes what you set, and WANCTL_RELAY/WANCTL_PORTAL still override it for one run. Nothing scripted is ever blocked by the question: with no terminal, or with WANCTL_NO_PROMPT=1, the command prints the wanctl config set line and exits.

Both the tool and the install script come from the project's GitHub Releases: the installer verifies the signed release manifest, checks each binary's size and hash, and only then writes anything to disk (with the system's own openssl — macOS's LibreSSL works too). The release source and your relay are independent of each other, so a relay having a bad day does not stop an install.

To start at boot, run wanctl service install — it writes the relay address and transport you configured into the service unit, so a reboot does not depend on environment variables (--relay / --transport set them explicitly).

Windows machines: see the next page.