* Host the manual's images in the repo under manual/images Replaces all learn.omacom.io/manual.omakub.org hotlinks with local webp files capped at 1600px wide (~20MB total), so the manual is fully self-contained. Theme and unlock previews are converted from the canonical themes/*/preview.png files. A handful of illustration shots are interim conversions of the current images, pending retakes. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * Retake the illustration screenshots fresh at 3x scaling Navigation, clipboard history, notices, tmux layouts, prompt, and About reshot on a clean workspace at 3.13x (1920x1080 logical on 6K) across ten themes: Tokyo Night, Catppuccin, Gruvbox, Kanagawa, Everforest, Nord, Osaka Jade, Rose Pine, Matte Black, and Ristretto. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * Drop the extra themes chapter from the manual The community theme gallery will live elsewhere. Removes the chapter and its 114 gallery images, renumbers the following chapters, and repoints the two references (theme installs still work via Install > Style > Theme in the menu). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * Reference the shipped theme previews instead of copying them The themes chapter now points straight at themes/*/preview.png and preview-unlock.png, so previews can never drift from the source and manual/images drops from 6.3MB to 3.5MB. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
3.9 KiB
Unattended Installs
The Omarchy ISO can install itself with nobody at the keyboard. If the installer finds a second drive labeled cidata carrying its configuration files, it copies them off, skips the setup wizard entirely, and reboots into the finished system on its own. No special ISO build, no extra boot menu entry — with no such drive attached, nothing changes and you get the normal wizard.
This makes Omarchy great as a base image for disposable dev environments: create a VM in Proxmox or with Packer, boot it, walk away, SSH in. cidata is the cloud-init NoCloud label, so all the common virtualization tooling already knows how to attach such a drive.
The configuration files
The files are exactly what the installer's own wizard writes, so the easiest way to get a starting set is to run one interactive install (in a VM, say) and copy what it wrote into /root:
| File | Required | Purpose |
|---|---|---|
user_configuration.json |
Yes | Disk, hostname, timezone, keyboard |
user_credentials.json |
Yes | Username and password hash |
user_full_name.txt |
No | Git full name |
user_email_address.txt |
No | Git email |
user_encrypt_installation.txt |
No | true when the configuration carries a disk_encryption block |
authorized_keys |
No | SSH public keys, one per line |
tailscale_authkey |
No | Tailscale auth key for joining your tailnet on first boot |
Generate the password hash for user_credentials.json with openssl passwd -6 "yourpassword".
You can also drop in an empty file named defer-provisioning in place of user_credentials.json. That runs the same prepare-for-another-owner install you can trigger interactively: the machine installs with no personal details, and whoever boots it first picks their keyboard and creates their user. That's the mode for imaging rigs, where the drive shouldn't carry anyone's credentials at all.
SSH access
When authorized_keys is present, the install sets the keys up as the user's ~/.ssh/authorized_keys, enables sshd, and opens the firewall for it. (A stock Omarchy install ships openssh with the service disabled and the port closed, so an unattended machine would otherwise be unreachable.) The install only adds your keys — it doesn't loosen any of the SSH daemon's other authentication settings.
When tailscale_authkey is present, the machine joins your tailnet on first boot instead: Tailscale is installed from the ISO's bundled packages, the firewall allows the tailnet interface, and a background job runs the join as soon as the machine actually has network, retrying until it succeeds. Use a reusable, pre-authorized key so one drive image can serve many machines.
Building the cidata drive
Any filesystem with the right label works. A tiny ISO is the easy way:
mkdir cidata
cp user_configuration.json user_credentials.json authorized_keys cidata/
genisoimage -output cidata.iso -volid cidata -joliet -rock cidata/
Then attach it to the VM alongside the Omarchy ISO. Here's a full Proxmox example:
qm create 101 --name my-omarchy \
--bios ovmf --machine q35 --cpu host --cores 4 --memory 8192 \
--ostype l26 --scsihw virtio-scsi-single \
--efidisk0 local-lvm:0,efitype=4m,pre-enrolled-keys=0 \
--scsi0 local-lvm:40,discard=on,iothread=1 \
--net0 virtio,bridge=vmbr0 --vga virtio --serial0 socket \
--ide2 local:iso/omarchy.iso,media=cdrom \
--ide3 local:iso/cidata.iso,media=cdrom \
--boot order='scsi0;ide2'
qm start 101
The boot order lists the disk first on purpose: the empty disk falls through to the ISO on the first boot, and the installed system boots from disk ever after.
Two caveats
Encrypted unattended installs aren't fully unattended — someone still has to type the LUKS passphrase at the first boot. And the disk_encryption block in user_configuration.json carries that passphrase in plaintext, so treat a cidata drive built from an encrypted install as the secret it is.