### System details

Intel Core i5-1235U / Intel Iris Xe, Omarchy 4.0.1-1, Hyprland 0.56.2-1, Quickshell 0.3.1-1, kernel 7.1.9-arch1-2

### What's wrong?

When a laptop is used in clamshell mode with its lid closed and one active external monitor, powering off or disconnecting that final monitor leaves the session unlocked and able to accept keyboard input while no display is visible.

This was reproduced twice with a scratch editor. It is an input-safety and secure-state bug: the user cannot see where input is going, but the previously focused client remains active and receives keys until a usable display returns.

#### Steps to reproduce

1. Connect exactly one external monitor to a laptop.
2. Open a scratch editor containing no sensitive information and leave it focused.
3. Close the laptop lid. Confirm the external monitor remains active and the internal panel is disabled for clamshell operation.
4. Physically power off or disconnect the final external monitor.
5. While no display is visible, briefly hold a harmless letter key.
6. Power on or reconnect the external monitor.

#### Actual behavior

- Hyprland removes the external output.
- Omarchy's clamshell reconciliation enables/modesets the internal panel even though the lid remains physically closed.
- No visible output is available to the user.
- The session does not lock; the previously focused application continues receiving keyboard input.
- When the external display returns, the session is still unlocked and the unseen input is present in the scratch editor.

#### Expected behavior

When a closed-lid session loses its final active external monitor, Omarchy should immediately enter a secure state before further keyboard input reaches the focused client. When a usable display returns, the user should see the lock screen and authenticate before returning to the application.

Open-lid unplugging, removing one of several external monitors, and already-locked sessions should retain their current behavior.

#### Sanitized display evidence

The relevant sequence was the same on both reproductions:

```text
drm: Connector DP-1 disconnected
drm: Disabling output DP-1
drm: DP-1 is not connected, clearing stale crtc
drm: Modesetting eDP-1
drm: Connector eDP-1 enabledState changed false -> true
```

There was no accompanying USB/HID device removal or re-enumeration during either reproduction.

#### Source-level path and possible fix direction

`bin/omarchy-hyprland-monitor-watch` handles monitor-removal events and invokes clamshell reconciliation. `bin/omarchy-hyprland-monitor-clamshell` disables the internal panel only while clamshell mode is true and an external output remains active. After the final external output disappears, that condition becomes false and the helper enables the physically hidden internal panel. Neither path checks for the combination of closed lid plus no remaining external display before returning input to the live session.

One fix direction is for the existing monitor watcher to serialize the removal event, verify that the lid is closed and no external display remains, check that no lock is already active or pending, and request `omarchy-system-lock`. Existing display recovery should still run so the lock screen becomes visible when a usable output returns.

I generated `omarchy-debug --no-sudo --print` locally, but the full output contains unrelated host, network, storage, and application identifiers, so I am not attaching it. The relevant versions and sanitized event sequence are included above.