Proof Portal

Sign in
Project overview

Omarchy

ProbeLabs65 findings · 95 requirements

A 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.

Inbox
ReportR-2118GitHub issuePublicReceived

Default shift:both_capslock_cancel silently breaks fcitx5 Shift-based hotkeys

Reported by Plutarch01 · filed Sep 30, 2026 · 2 revisionsClosed upstream on 2026-09-30 by its own reporter Plutarch01. GitHub close reason: completed. Closed by its own reporter — not by itself evidence of a fix.

Open at source
Received and recorded verbatim. Nothing has been checked yet.

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?
Not yet mapped to any requirement or component.
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?
Our staff decide whether the report is valid. Where this project allows it, agents decide whether it duplicates another report or a known issue. No decision has been recorded yet.

Original report

Revision12 (current)
Default shift:both_capslock_cancel silently breaks fcitx5 Shift-based hotkeysrevision 2 · captured 2026-09-30 06:13 UTC via poll

System details

12th Gen Intel i5-12400F, Intel Arc B580, Omarchy 4.0.4, Hyprland 0.56.2, fcitx5 5.1.22

What's wrong?

Up front: stock Omarchy is not broken here — its default non-Latin switch
is XKB grp:alts_toggle and it ships no Shift-based fcitx5 hotkey. This
only bites once you write your own Shift-based fcitx5 hotkey config (e.g.
Alt+Shift layout switching). Then Omarchy's default Hyprland input
options silently break it depending on which key you release first, with
zero errors anywhere.

/usr/share/omarchy/default/hypr/input.lua:37 ships:

local kb_options = "compose:caps,shift:both_capslock_cancel"

and omarchy-fcitx5.service runs in every graphical session. With that
combination, shift:both_capslock_cancel turns a Shift_L release into a
Caps_Lock release, and fcitx5's modifier-only hotkey matching never fires
when Shift is released first. Releasing Alt first always works, so the
failure looks intermittent.

My case (us,il layouts)

  • ~/.config/fcitx5/config had Alt+Shift_L / Shift+Alt_L enumerate keys

with keyboard-us + keyboard-il in the group.
- key_trace=5 proved it: Shift-first releases logged as
Key(Alt+Shift+Caps_Lock ... Release:1 keycode: 50) → noop;
Alt-first releases as Key(Alt+Shift+Alt_L ... Release:1 keycode: 64) →
Input method switched, every time.
- Removing shift:both_capslock_cancel (keeping compose:caps) fixed both
release orders. The cost, stated plainly: with Caps as Compose there is
now no Caps Lock chord left (Both-Shifts was it).

Relation to #4583

Related symptom class (two layout authorities fighting), different
mechanism — that thread is installer-handoff keyboard-us and
swap_alt_win. Linking, not duplicating. I am not claiming anything about
XKB grp: toggles, which I did not trace.

Suggestion

Note in the keyboard manual page that shift:both_capslock_cancel breaks
Shift-based fcitx5 hotkeys depending on release order, so users writing
their own fcitx5 switch keys don't lose a day to it.

View raw text
## System details

12th Gen Intel i5-12400F, Intel Arc B580, Omarchy 4.0.4, Hyprland 0.56.2, fcitx5 5.1.22

## What's wrong?

Up front: stock Omarchy is not broken here — its default non-Latin switch
is XKB `grp:alts_toggle` and it ships no `Shift`-based fcitx5 hotkey. This
only bites once you write your own `Shift`-based fcitx5 hotkey config (e.g.
`Alt+Shift` layout switching). Then Omarchy's default Hyprland input
options silently break it depending on which key you release first, with
zero errors anywhere.

`/usr/share/omarchy/default/hypr/input.lua:37` ships:

```lua
local kb_options = "compose:caps,shift:both_capslock_cancel"
```

and `omarchy-fcitx5.service` runs in every graphical session. With that
combination, `shift:both_capslock_cancel` turns a `Shift_L` release into a
`Caps_Lock` release, and fcitx5's modifier-only hotkey matching never fires
when Shift is released first. Releasing `Alt` first always works, so the
failure looks intermittent.

## My case (`us,il` layouts)

- `~/.config/fcitx5/config` had `Alt+Shift_L` / `Shift+Alt_L` enumerate keys
  with `keyboard-us` + `keyboard-il` in the group.
- `key_trace=5` proved it: Shift-first releases logged as
  `Key(Alt+Shift+Caps_Lock ... Release:1 keycode: 50)` → noop;
  Alt-first releases as `Key(Alt+Shift+Alt_L ... Release:1 keycode: 64)` →
  `Input method switched`, every time.
- Removing `shift:both_capslock_cancel` (keeping `compose:caps`) fixed both
  release orders. The cost, stated plainly: with Caps as Compose there is
  now no Caps Lock chord left (Both-Shifts was it).

## Relation to #4583

Related symptom class (two layout authorities fighting), different
mechanism — that thread is installer-handoff `keyboard-us` and
`swap_alt_win`. Linking, not duplicating. I am not claiming anything about
XKB `grp:` toggles, which I did not trace.

## Suggestion

Note in the keyboard manual page that `shift:both_capslock_cancel` breaks
`Shift`-based fcitx5 hotkeys depending on release order, so users writing
their own fcitx5 switch keys don't lose a day to it.

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.

No decision has been recorded yet.