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.
4.0.x: user ~/.config/hypr/envs.lua is never loaded — user env overrides (and the updater's own iHD fix) are dead code
Reported by Makedayz · filed Sep 2, 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
Intel UHD Graphics (Raptor Lake-P, iGPU-primary) + NVIDIA RTX 4050 hybrid laptop, Omarchy 4.0.1, Hyprland / Wayland
What's wrong?
~/.config/hypr/envs.lua — the personal counterpart of default/hypr/envs.lua — is never loaded. Nothing ever requires it, so env vars a user sets there are silently ignored.
Evidence
- Neither the migrated user config, the shipped template (
config/hypr/hyprland.lua), nor currentmainrequireshypr.envs. The personal requires are onlyhypr.monitors,hypr.input,hypr.bindings,hypr.looknfeel,hypr.autostart— and the comment block above them ("Put your personal overrides in these files") does not mention envs either. grep -rn 'hypr\.envs' /usr/share/omarchy ~/.config/hyprfinds onlydefault.hypr.envs(omarchy.lua:16). No reference to the user module.default/hypr/bootstrap.luaonly adds~/.config/?.luatopackage.path(making the module requireable); it does not load it.require_allis used only for apps, bindings, toggles, and workspace-layouts — never for~/.config/hypr.- Meanwhile the defaults do set hardware envs:
omarchy.lua→default.hypr.envs→default.hypr.nvidia→ on a hybrid laptop with GSP,LIBVA_DRIVER_NAME=nvidiais what actually lands in the session.
Concrete failure on this hybrid laptop
An Omarchy update (Aug 29, 4.0.x) itself rewrote the user's envs.lua from LIBVA_DRIVER_NAME=nvidia to iHD, leaving a timestamped .bak and this comment:
-- Video decode (VAAPI) forced to Intel iGPU: the panel is wired to Intel,
-- so decoding on NVIDIA required an extra cross-GPU frame copy that caused
-- browser video glitches. NVIDIA still handles OpenGL/Vulkan (games) above.But since nothing loads envs.lua, the updater's own fix never takes effect. Browsers inherit no LIBVA_DRIVER_NAME, libva auto-detects a driver per render node, selects the NVDEC VAAPI shim on the dGPU node, and hardware-decoded video in Chromium/Brave paints a checkerboard (or never paints at all).
Steps to reproduce
- Omarchy 4.0.1 on an iGPU-primary Intel+NVIDIA hybrid laptop.
- Set
hl.env("LIBVA_DRIVER_NAME", "iHD")in~/.config/hypr/envs.lua, relogin. systemctl --user show-environment | grep LIBVA→ variable not applied; browser video still broken.- Launch a browser directly with
LIBVA_DRIVER_NAME=iHD brave→ video plays fine (confirming the env is the whole fix). - Add
require("hypr.envs")to the personal block ofhyprland.lua, relogin →LIBVA_DRIVER_NAME=iHDis exported and video works without the manual env.
Regression note: in 3.x the personal env file was loaded — hyprland.conf sourced ~/.config/hypr/envs.conf in the personal overrides section (see the load order in #6068). The 3.x → 4.x Lua migration dropped it from the personal require list, while the updater still patches the file as if it were live.
Suggested fix: add require_optional.module("hypr.envs") (or require("hypr.envs")) alongside the other personal requires in config/hypr/hyprland.lua, and have the updater insert that require into migrated hyprland.lua files that predate 4.0.
Related but distinct: #8989, #9890, #9096, #8328, #8215, #9708 report the video corruption caused by nvidia.lua forcing NVIDIA VA-API on hybrid laptops. This report is about the user env override channel itself being dead code — which also silently disables the updater's own envs.lua fix.
View raw text
### System details
Intel UHD Graphics (Raptor Lake-P, iGPU-primary) + NVIDIA RTX 4050 hybrid laptop, Omarchy 4.0.1, Hyprland / Wayland
### What's wrong?
`~/.config/hypr/envs.lua` — the personal counterpart of `default/hypr/envs.lua` — is never loaded. Nothing ever `require`s it, so env vars a user sets there are silently ignored.
**Evidence**
1. Neither the migrated user config, the shipped template (`config/hypr/hyprland.lua`), nor current `main` requires `hypr.envs`. The personal requires are only `hypr.monitors`, `hypr.input`, `hypr.bindings`, `hypr.looknfeel`, `hypr.autostart` — and the comment block above them ("Put your personal overrides in these files") does not mention envs either.
2. `grep -rn 'hypr\.envs' /usr/share/omarchy ~/.config/hypr` finds only `default.hypr.envs` (`omarchy.lua:16`). No reference to the user module.
3. `default/hypr/bootstrap.lua` only adds `~/.config/?.lua` to `package.path` (making the module *requireable*); it does not load it. `require_all` is used only for apps, bindings, toggles, and workspace-layouts — never for `~/.config/hypr`.
4. Meanwhile the defaults do set hardware envs: `omarchy.lua` → `default.hypr.envs` → `default.hypr.nvidia` → on a hybrid laptop with GSP, `LIBVA_DRIVER_NAME=nvidia` is what actually lands in the session.
**Concrete failure on this hybrid laptop**
An Omarchy update (Aug 29, 4.0.x) itself rewrote the user's `envs.lua` from `LIBVA_DRIVER_NAME=nvidia` to `iHD`, leaving a timestamped `.bak` and this comment:
```lua
-- Video decode (VAAPI) forced to Intel iGPU: the panel is wired to Intel,
-- so decoding on NVIDIA required an extra cross-GPU frame copy that caused
-- browser video glitches. NVIDIA still handles OpenGL/Vulkan (games) above.
```
But since nothing loads `envs.lua`, the updater's own fix never takes effect. Browsers inherit no `LIBVA_DRIVER_NAME`, libva auto-detects a driver per render node, selects the NVDEC VAAPI shim on the dGPU node, and hardware-decoded video in Chromium/Brave paints a checkerboard (or never paints at all).
**Steps to reproduce**
1. Omarchy 4.0.1 on an iGPU-primary Intel+NVIDIA hybrid laptop.
2. Set `hl.env("LIBVA_DRIVER_NAME", "iHD")` in `~/.config/hypr/envs.lua`, relogin.
3. `systemctl --user show-environment | grep LIBVA` → variable not applied; browser video still broken.
4. Launch a browser directly with `LIBVA_DRIVER_NAME=iHD brave` → video plays fine (confirming the env is the whole fix).
5. Add `require("hypr.envs")` to the personal block of `hyprland.lua`, relogin → `LIBVA_DRIVER_NAME=iHD` is exported and video works without the manual env.
**Regression note:** in 3.x the personal env file *was* loaded — `hyprland.conf` sourced `~/.config/hypr/envs.conf` in the personal overrides section (see the load order in #6068). The 3.x → 4.x Lua migration dropped it from the personal require list, while the updater still patches the file as if it were live.
**Suggested fix:** add `require_optional.module("hypr.envs")` (or `require("hypr.envs")`) alongside the other personal requires in `config/hypr/hyprland.lua`, and have the updater insert that require into migrated `hyprland.lua` files that predate 4.0.
Related but distinct: #8989, #9890, #9096, #8328, #8215, #9708 report the video corruption caused by `nvidia.lua` forcing NVIDIA VA-API on hybrid laptops. This report is about the user env override channel itself being dead code — which also silently disables the updater's own `envs.lua` fix.
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 10:38 UTC