Proof Portal
Omarchy
ProbeLabs65 findings · 95 requirementsA proof layer — requirements, tests and verified fixes — for two of Omarchy's subsystems: the application menu (launcher scripts, QML model, JSONC config, search and selection) and the lock screen (lock scripts, QML, PAM authentication). Scope is deliberately limited to those components of omacom/omarchy; the rest of the distribution is not covered.
Screenshot picker offers rectangles for windows hidden behind a maximized/fullscreen window
Reported by Ethanxu1 · filed Aug 25, 2026 · open at source
At a glance
- 1 · What is claimed?
- Only the report text below — no validator has restated it as checkable claims yet.
- 2 · What does it affect?
- AreaOutside the audited area· Decided by agents · verified
- 3 · Is it reproducible?
- No validation has been run yet.
- 4 · What remains unclear?
- Not validated yet: no claims, no reproducer.
- 5 · What happens next?
- Decided by agents and verified by a second, independent agent — see the decision.
Original report
System details
Omarchy: 4.0.0-1 Hyprland: 0.56.2 (v0.56.2, commit efb50993780079460b0cbed1363e2166a2de1d9f), Kernel: 7.1.8-arch1-3
What's wrong?
What happened
When a workspace has a maximized or fullscreened window on top of other windows, omarchy screenshot (and omarchy capture screenshot windows) still lets you select rectangles for the windows underneath, even though they are not visible on screen. The picker "snaps" to a window that is completely covered.
What I expected
Only the maximized/fullscreen window's rectangle should be selectable while it's covering the workspace, since anything underneath isn't actually visible.
Steps to reproduce
- Open two or more windows on the same workspace.
- Maximize one (
SUPER + ALT + F, i.e.hl.dsp.window.fullscreen({ mode = "maximized" })). - Run
omarchy screenshot(or the "windows" capture mode) and hover/click where a covered window used to be. - The picker highlights and lets you select the hidden window's old rectangle, not just the maximized window on top.
Root cause
bin/omarchy-capture-region's window_rects() builds selectable rectangles from hyprctl clients -j, filtering only on hidden != true:
window_rects() {
hyprctl clients -j | jq -r --arg ws "$(active_workspace)" \
'[.[] | select(.workspace.id == ($ws | tonumber) and .hidden != true) | "\(.at[0]),\(.at[1]) \(.size[0])x\(.size[1])"] | unique[]'
}Hyprland's hidden field does not track occlusion by a fullscreen/maximized sibling — a window covered by a maximized window on the same workspace still reports hidden: false in hyprctl clients -j. So window_rects() (used by get_rectangles, geo_at_cursor, and the --select-window Tab-cycling) keeps offering rectangles for windows that aren't actually visible.
Confirmed against the current bin/omarchy-capture-region on the quattro branch — the bug is present in the latest source, not just an older release.
Suggested fix
In window_rects(), when any window on the workspace is fullscreen/maximized, only emit that window's rectangle:
window_rects() {
hyprctl clients -j | jq -r --arg ws "$(active_workspace)" '
[ .[] | select(.workspace.id == ($ws | tonumber) and .hidden != true) ] as $ws_windows |
if ($ws_windows | any(.fullscreen == 1 or .fullscreen == true)) then
$ws_windows | map(select(.fullscreen == 1 or .fullscreen == true))
else
$ws_windows
end |
map("\(.at[0]),\(.at[1]) \(.size[0])x\(.size[1])") | unique[]
'
}(fullscreen is an int in some Hyprland versions and a bool in others, hence checking both.)
View raw text
### System details
Omarchy: 4.0.0-1 Hyprland: 0.56.2 (v0.56.2, commit efb50993780079460b0cbed1363e2166a2de1d9f), Kernel: 7.1.8-arch1-3
### What's wrong?
## What happened
When a workspace has a maximized or fullscreened window on top of other windows, `omarchy screenshot` (and `omarchy capture screenshot windows`) still lets you select rectangles for the windows underneath, even though they are not visible on screen. The picker "snaps" to a window that is completely covered.
## What I expected
Only the maximized/fullscreen window's rectangle should be selectable while it's covering the workspace, since anything underneath isn't actually visible.
## Steps to reproduce
1. Open two or more windows on the same workspace.
2. Maximize one (`SUPER + ALT + F`, i.e. `hl.dsp.window.fullscreen({ mode = "maximized" })`).
3. Run `omarchy screenshot` (or the "windows" capture mode) and hover/click where a covered window used to be.
4. The picker highlights and lets you select the hidden window's old rectangle, not just the maximized window on top.
## Root cause
`bin/omarchy-capture-region`'s `window_rects()` builds selectable rectangles from `hyprctl clients -j`, filtering only on `hidden != true`:
```bash
window_rects() {
hyprctl clients -j | jq -r --arg ws "$(active_workspace)" \
'[.[] | select(.workspace.id == ($ws | tonumber) and .hidden != true) | "\(.at[0]),\(.at[1]) \(.size[0])x\(.size[1])"] | unique[]'
}
```
Hyprland's `hidden` field does not track occlusion by a fullscreen/maximized sibling — a window covered by a maximized window on the same workspace still reports `hidden: false` in `hyprctl clients -j`. So `window_rects()` (used by `get_rectangles`, `geo_at_cursor`, and the `--select-window` Tab-cycling) keeps offering rectangles for windows that aren't actually visible.
Confirmed against the current `bin/omarchy-capture-region` on the `quattro` branch — the bug is present in the latest source, not just an older release.
## Suggested fix
In `window_rects()`, when any window on the workspace is fullscreen/maximized, only emit that window's rectangle:
```bash
window_rects() {
hyprctl clients -j | jq -r --arg ws "$(active_workspace)" '
[ .[] | select(.workspace.id == ($ws | tonumber) and .hidden != true) ] as $ws_windows |
if ($ws_windows | any(.fullscreen == 1 or .fullscreen == true)) then
$ws_windows | map(select(.fullscreen == 1 or .fullscreen == true))
else
$ws_windows
end |
map("\(.at[0]),\(.at[1]) \(.size[0])x\(.size[1])") | unique[]
'
}
```
(`fullscreen` is an int in some Hyprland versions and a bool in others, hence checking both.)
Validation
Not checked yet. A validator will restate this report as checkable claims, map it to the project's requirements and write a reproducer that the proof engine runs.
Linked
Not linked to any known issue, other report or change yet.
Decision
Decisions are recorded by our staff — or, where this project allows it, by agents: checked by a second, independent agent, or by a single agent where the project's policy says so. Any decision reached so far is shown below.
- Out of scopeDecided by agents · verified · 2026-10-09 09:09 UTC