Human-requested: a thorough visual review of the whole themed surface
(Waybar incl. whole-swaps, rofi, swaync, hyprlock, terminals, splash,
greeter) with "stunning" as the bar. Grounded in design theory rather
than taste: color-harmony models + perceptual (OKLCH) lightness
ordering of the palette roles, color-blind-safe status colors, an
extended contrast checker encoding the adopted rules, a consistent
type scale + spacing grid, and explicit visual hierarchy. Image-expert
models are available for designed artwork — slices that need graphics
stop and ask instead of shipping programmer-art. Sliced audit-first.
Placed at the head of NEXT for Bernardo to reorder.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
New modules/nixos/greeter.nix owns the greetd block (moved out of
default.nix) plus the theming, the Plymouth model — baked at SYSTEM
rebuild from nomarchy.system.stateFile. Two parts, because tuigreet
draws on the VT with the 16 ANSI slots: console.colors becomes the
theme's ansi[] (vt.default_* kernel params — which also themes raw
ttys and the LUKS passphrase prompt), and tuigreet gets a --theme of
NAMED slots (its parser is ratatui Color::from_str, checked in source;
the names map to the standard indexes the retinted palette now
serves). A sparse state without ansi skips theming cleanly.
Verified: V0; V1 — the template-system toplevel bakes the full --theme
spec into greetd.toml and all three vt.default_{red,grn,blu} palettes
into kernel-params. A checks.greeter VM test was attempted and dropped
after bisecting: tuigreet dies under runNixOSTest even without the
flag (harness limit; nixpkgs's own greetd test uses agreety) —
recorded in agent/MEMORY.md. Remains: V3 rendering +
theme-follows-rebuild on hardware (HARDWARE-QUEUE).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
nomarchy-display-profile gains `match <outputs>`: an exact set match
wins, else the largest profile whose named outputs are all connected;
any tie prints nothing — the watcher must never flap between ambiguous
layouts. nomarchy-display-profile-watch (exec-once when profiles are
declared, the keyboard-watcher poll pattern — session units race
Hyprland IPC on relogin) reacts to output-set CHANGES only, reads
settings.displayProfileAuto from the live state each time (the menu's
new Auto-switch row toggles it instantly, no rebuild), and applies via
the tool — so an auto-pick persists a concrete profile that the next
rebuild bakes. A separate boolean rather than an "auto" pseudo-profile
for exactly that bake semantics.
Verified: V0 (flake check); V1 — a scratch 4-profile downstream builds;
tool + watcher pass bash -n; the match matrix exercised end-to-end:
three exact picks, an unambiguous-superset pick, a tie -> none, a
no-fit -> none; the watcher is exec-once'd in the rendered config; the
menu renders the Auto-switch row. Remains: V3 real dock/undock
(HARDWARE-QUEUE, rides the slice-a entry). Slice c (workspace binding)
stays, tagged optional.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
nomarchy.displayProfiles: named layouts for the same outputs, each a
list of nomarchy.monitors-style entries. The active profile lives in
the in-flake state (settings.displayProfile) on the night-light
pattern: nomarchy-display-profile apply <name> fires the baked hyprctl
rules instantly and writes the state with --no-switch; the next rebuild
bakes the profile's entries over nomarchy.monitors whole-by-name.
Monitor-rule composition is extracted to pure monitor-rules.nix
(renderer + 3-layer resolve, incl. a disable-guard so a stale menu
resolution pick can't resurrect a profile-disabled panel, and junk
state degrading to base config instead of an eval error) — because a
writeText/toFile state fixture is unreadable at flake-check eval time,
the new checks.display-profiles unit-tests the pure function directly.
Menu: System › Display gains a self-gated "Profiles ›" row (●/○ active
mark, Base layout to clear); commented template example + README row.
Verified: V0 (flake check, forcing the new check's asserts); V1 — a
scratch downstream through lib.mkFlake builds: the generated tool
passes bash -n with correct rules (disable/vrr/float scale), the baked
hyprland.conf shows the exact expected overlay, the menu renders, and
list/usage smoke-test clean. Remains: V3 live switching on a dock
(HARDWARE-QUEUE). Slices b (hotplug auto-switch) + c (workspace
binding) stay in item 15.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The item required a written plan before coding. Option shape reuses the
existing monitorType submodule (attrsOf (listOf monitorType)); active
profile is in-flake state (settings.displayProfile) on the night-light
instant+bake pattern; whole-entry-by-name rebuild overlay with a
disable-guard so a stale menu resolution pick can't resurrect a
disabled panel; self-gated System > Display profile row; sliced a/b/c
with hotplug auto-switch and workspace binding deferred behind the
core picker. kanshi remains a non-goal.
Verified: V0 (docs-only change).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
.gitea/workflows/bump.yml: Mondays 05:17 UTC (+ workflow_dispatch for
security bumps), nix flake update — within the pinned release branches
by construction, never a release jump — gated by nix flake check
--no-build, then a pathspec-limited flake.lock commit pushed to main
with the update's change list as the body. That push triggers check.yml
as a second net. Red gate or push race = a visibly failed run, nothing
lands. v1 promotion stays human. py_compile/bash -n deliberately not in
the gate: they don't read the lock. Container recipe inherited from
check.yml (nixbld, sandbox=false, pinned Nix 2.31.5).
Verified: V0 (yq parse, bash -n per run block); the update/commit/push
logic simulated end-to-end in a scratch clone with a real nix flake
update and a bare remote (changed + unchanged branches, bystander file
stays uncommitted, push lands); today's bumped lock evals green.
Remains: the first real run (cron pickup + Actions-token push) is not
confirmable from a session — BACKLOG item 14 slimmed to that watch.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The last niceties slice: pick a pixel's color into the clipboard.
hyprpicker joins grim/slurp in systemPackages; a new nomarchy-menu
`colorpicker` mode captures its output, wl-copies it, and toasts the
hex (Esc = silent cancel). Tools › row self-gates on hyprpicker being
present; SUPER+CTRL+P added in keybinds.nix, so the Hyprland bind and
the SUPER+? cheatsheet row come from the one source. Item 13 complete
and deleted from BACKLOG.
Verified: V0 (flake check); V1 — template-home and template-system
build; rendered nomarchy-menu passes bash -n and contains the mode +
gated row; cheatsheet has the SUPER+CTRL+P row; hyprland.conf has the
bind; the system toplevel ships sw/bin/hyprpicker. Remains: V3 — real
loupe → click → wl-paste round-trip in a session (HARDWARE-QUEUE).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The bar colored the battery at 25/10% but nothing notified. New
shellcheck-gated watcher (pkgs/nomarchy-battery-notify, overlay) polls
the same sysfs the bar reads — type=Battery, scope!=Device — and fires
one toast per downward crossing (normal at 25%, critical at 10%; swaync
keeps critical up until dismissed), re-armed by charging. Self-gates on
battery presence (the powerprofile pattern), so desktops get a silent
no-op. HM user unit behind nomarchy.batteryNotify.enable (default on)
with libnotify on the unit PATH; notify-send resolves from PATH so the
VM check can shim it. poweralertd rejected: its UPower thresholds
(20/5) wouldn't match the bar, and it toasts every plug/unplug.
Verified: V0 (flake check, bash -n); V1 (template-home builds, unit in
the generation); V2 GREEN (new checks.battery-notify, test_power fake
battery: self-gate, silence at 80%, one toast per crossing, critical
urgency, charging re-arms). Remains: V3 — real swaync toast on a
draining laptop (HARDWARE-QUEUE).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Summer-night's whole-swap bar had an idle_inhibitor the generated bar
lacked (reverse parity gap from the 2026-07-04 sweep). The generated
bar now carries it — click → , warm tint while active, tooltips
saying what it holds — blocking hypridle's lock/display-off/suspend
for videos and presentations; state deliberately resets with the bar.
summer-day mirrors summer-night's module + CSS, and the keybinds
cheatsheet mentions the bar click (via keybinds.extra).
V1: rendered config has the module with correct glyphs, CSS rules
present, cheatsheet row renders. Inhibit behaviour needs a session →
V3 queued.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
nomarchy-record owns the lifecycle: start prefers wl-screenrec (VAAPI
hardware encode) and falls back to wf-recorder if it dies at startup
(the no-VAAPI case), stop SIGINTs so the file finalizes cleanly, and
status feeds a self-gating Waybar ⏺ REC (alert red) whose click is
the one stop surface — signal 8 pokes make it appear/vanish
instantly. Capture gains Record region/screen (+ audio variants) and
self-swaps to '■ Stop recording' while one runs. Parity: both summer
whole-swap bars carry the same module. Recordings land in
~/Videos/Recordings with the screenshot timestamp naming (deviation
from 'next to Screenshots': an .mp4 in Pictures felt wrong).
V1: rendered bar config + CSS asserted, summer jsonc parse, recorder
dry-run state machine correct (idle/active/usage exits), both
recorders in home-path. Real capture needs a session → V3 queued
(incl. the VAAPI-vs-fallback check on the AMD box).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
'Never has to master Nix' includes the error messages. The tool grows
validate_state (appearance schema = hard errors; settings.* and
unknown keys = warnings so menu writers and newer schemas keep
working), a read-only `validate` subcommand, and validation BEFORE
every write — an invalid set/apply is refused with the on-disk file
untouched. JSON syntax errors now say line, column, and fix. theme.nix
enforces the same schema at eval time (on the post-defaults state, so
sparse/older files still evaluate) and throws a message naming each
field, its got-value, and the fix paths — the hand-edit escape hatch
no longer ends in a Nix stack trace. Doctor's JSON fix-text points at
validate.
V1: new checks.theme-sync-validate fixture corpus green (good state,
trailing comma, bad hex, bad type, unknown key, refused set with
byte-identical file); negative mkFlake eval of a broken scratch
downstream shows the friendly throw; good-path template-home builds.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A git add -A in 2a23e82 swept in .claude/skills/use-gemini/SKILL.md —
a Claude Code harness-injected tool doc, not part of the distro. Same
treatment as settings.local.json (db6a0ba): untracked + gitignored.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The things that actually break user machines, one command (and menu ›
System › Doctor): failed system/user units, disk space (real
filesystems only — fstype allowlist makes it VM/live safe, device
dedupe collapses / and /nix), theme-state.json parses + is
git-tracked, flake dirty/behind as warnings, generation age (profile
symlink mtime — store paths are epoch-1), snapper timeline when
enabled. Read-only by contract; every ✖ prints the one fixing command;
exit 1 on any problem. Ships as pkgs/nomarchy-doctor so the VM check
runs on a minimal node.
Verified: V1 package build (shellcheck-gated) + a real-hardware smoke
that correctly flagged a genuinely failed user unit on the dev box;
V2 green — checks.doctor VM test: induced failed unit flips the sheet
to exit-1 naming the unit and its fix, reset-failed returns healthy.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Undo is one menu away: the new rollback flow lists the last ten Home
Manager generations (the theme/config history is already generations,
newest marked current) and activates the picked one in a terminal —
HM's supported rollback, reversible by re-picking a newer one, so no
typed-yes gate. System-level undo deliberately links out rather than
reimplementing destructive flows in rofi: the Snapshots row opens the
existing gated flow, and 'boot an older generation (how)' fires an
instruction notification pointing at docs/RECOVERY.md §3 (which now
cross-references the menu path).
V1: template-home built, generated menu bash -n clean, row rendering
and pick-parsing verified against fixture generations output. V3
exercise queued.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The informative half of 'informative + rock-stable': both rebuild
wrappers capture /run/current-system before the switch and print a
package-level nvd diff after it — store-path-pinned so no PATH
dependence, || true so an ornamental diff can never fail a succeeded
rebuild, and an identical-path fast-path says 'no changes'. README §3
notes it. V1: template-system toplevel rebuilt, generated scripts
bash -n clean, nvd smoke-run against the real closure.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The complete-workstation gap: no PDF/image viewer shipped and nothing
set mime defaults, so 'open a photo' fell to GIMP. Two new default-on
toggles: nomarchy.viewers (programs.zathura — Stylix themes it, target
added to the explicit list — plus imv) and nomarchy.mime (mkDefault
xdg.mimeApps: pdf→zathura, image→imv, av→mpv, text→code, dir→thunar;
html/http(s)→firefox as inert entries that only activate when a
browser is installed — the ship-a-browser Decision stays open, now
smaller). Associations for absent apps are skipped by GIO, so the
whole thing degrades as the user curates the suite. Deviation from
the backlog text: viewers are a module toggle, not template package
lines — zathura's theming needs its HM module (the real-config
boundary of the toggle discipline).
V1: template-home generation asserted — mimeapps.list rendered, themed
zathurarc, binaries in home-path. V3 open-a-file smoke queued.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Slice (c), the last of the docs review: the install/first-run story
read end to end against the tree. Facts held (mkFlake args match the
table, disko's @snapshots + hibernation swapfile, .#default/.#me
naming); the one seam — README §3 never pointed the manual path at the
template README's first-run steps — got a connective paragraph. Item 6
deleted from BACKLOG; ROADMAP § Full docs review flipped to ✓.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Slice (d) of the docs review: docs/RECOVERY.md tells the recovery
story that existed only in pieces, ordered by symptom — bad theme/HM
change (generations, apply-known-good, git-revert the flake), desktop
won't start (TTY, greetd + user journals, VM-GL and preactivate-log
pointers), broken system change (older systemd-boot generation, then
make it stick), lost files (btrfs-assistant / nomarchy-snapshots), and
the ISO + nixos-enter last resort. Commands checked against the tree;
linked from the README file tree and §5.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Slice (b) of the docs review: every factual claim in the three docs
spot-checked against the tree (mkDefault sites, bind priority, CLI
keys, awww naming, preset/wallpaper counts, summer-night's light bar,
live-ISO monitor rule, CI workflow, QEMU GL flags, template steps) —
all held. Two gaps fixed: OVERRIDES §2 predated nomarchy.monitors and
never routed readers to it (added, plus a quick-reference row);
TESTING §1 didn't list the shell-syntax check that §1b says CI runs
(added bash -n).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Slice (a) of the docs review: 9 undocumented nomarchy.* options got
README rows or inline .leaf mentions (autoTimezone, camera IR-hide,
intel.guc, amd.pstate/vaapi, package, system.stateFile, restic.paths).
The diff is now a flake check: option names are eval-walked from the
four option-declaring modules (config halves stay lazy) and compared
to the tables by tools/check-option-docs.py — undocumented options and
stale table rows both fail. Negative-tested against the pre-fix README
(reports exactly the 9 gaps). Item 6 split into slices a–d in BACKLOG.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Audited every rofi -dmenu call site in the generated nomarchy-menu +
nomarchy-vpn. The earlier "Back everywhere" pass held up: power,
power-profile, theme, clipboard, files, audio, display (both stages),
capture, tools, system and all four VPN lists already end in ↩ Back.
Fixed: the keybinds cheatsheet — Enter on any row silently closed it;
it now ends in ↩ Back returning to the root picker (its launch point),
and Enter elsewhere stays a deliberate no-op.
Documented (rofi.nix, at the convention comment + call sites): external
modi (rofi-calc, rofi-emoji, networkmanager_dmenu, rofi-pulse-select)
cannot take an injected row — Esc is their back path; free-text
prompts (web, ask) have no list to append to.
Verified: V0; V1 (built nomarchy-menu, bash -n OK; awk audit over the
generated script shows the only back-less lists are the root picker —
by design — and the free-text prompt). V3 spot-check queued.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Source-read of the pinned networkmanager_dmenu 2.6.2: with
compact = False (what our config.ini set) the script inserts literal
empty rows as separators between the ethernet/wifi/VPN sections —
the nameless entries Bernardo saw on the Latitude. compact = True
drops the separators and uses the unpadded row format, which also
renders better in rofi's proportional font than space-aligned columns.
Known residual, documented in config comment + HARDWARE-QUEUE: a
hidden-SSID AP still renders one nameless row (sec+bars only; the
name dedup keeps exactly one, and no config option filters it). If the
V3 retest still shows it, the follow-up is a small source patch
skipping empty-name APs in create_ap_list.
Verified: V0; V1 (rendered config.ini shows compact = True). Behavior
needs Wi-Fi hardware — V3 queued on the Latitude.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Follow-on to the swaync fix (a643391): the generated Waybar CSS colored
the module row and #window @subtext and dim states @muted, and the
generated rofi theme put @fg text on @surface chips — invisible where
palettes use those roles as "on-surface" (flexoki-light: subtext==base,
surface==text; gruvbox: muted/base = 1.27). flexoki-light has no
whole-swap, so it ran this CSS. Generated surfaces only — the summer
whole-swaps are hand-authored and untouched.
- waybar: secondary text → alpha(@text, 0.85); dim states (inactive
workspaces, DND bell, muted volume) → alpha(@text, 0.5).
- rofi: @dim → text+"80", @surface → text+"1a" (#RRGGBBAA fg tints).
- check-theme-contrast.py pairings/docs now cover all three generated
surfaces; status glyphs (good/warn/bad on base, worst 2.05) are
deliberately not held to a text ratio — raising them is palette
design, a human decision.
Verified: V0; V1 (checks.theme-contrast green over 21 themes;
downstream-template-home builds; rendered waybar style has zero
@subtext/@muted/@surface; rendered custom.rasi shows fg-derived
tints). Honesty: NOT visually verified — V3 queued (flexoki-light
eyeball + dark-theme regression spot-check).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Palette roles aren't uniform: summer-day and flexoki-light use subtext
as text-ON-surface (== base) and surface as a dark chip (== text), so
swaync's @subtext-on-@base body text and @surface/@text buttons were
self-colored (caught on the Latitude).
- body/summary: @text on @base — the one text pairing that passes in
every palette (survey: worst WCAG ratio 5.18 across all 21 themes;
subtext/base bottomed at 1.00 twice, muted/base at 1.27).
- hover rows + widget-title buttons: alpha(@text, 0.1) tints —
contrast by construction; subtext/surface defines dropped.
- NEW checks.theme-contrast (tools/check-theme-contrast.py): asserts
the hex-on-hex text pairings across themes/*.json, cheap (no VM).
The audit also found the generated waybar/rofi CSS shares the bug
class, live on flexoki-light (no whole-swap) → BACKLOG item 27.
Verified: V0; V1 (checks.theme-contrast builds green, 21 themes;
rendered services.swaync.style has zero @subtext/@surface). Honesty:
NOT visually verified — V3 queued (notify-send on summer-day).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
? is typed as Shift+/, so Hyprland saw mods SUPER+SHIFT with keysym
`question` and the exact-modmask match skipped the `$mod, question`
bind — the keypress fell through to the focused window (caught on the
Latitude, BACKLOG item 26; the menu itself was fine from a terminal).
- keybinds.nix: bind is now `$mod SHIFT, question`.
- rofi.nix prettyKeys collapses "SHIFT + ?" back to "?" so the
cheatsheet row keeps the documented SUPER + ? label.
- MEMORY.md gotcha: shifted keysyms need SHIFT in mods — invisible to
eval-tier tests.
Verified: V0 (flake check --no-build); V1 both surfaces (eval'd
settings.bind renders `$mod SHIFT, question, exec, nomarchy-menu
keybinds`; built nomarchy-menu's cheatsheet renders `SUPER + ?`).
V3 queued: SUPER+? on the Latitude after home-update + relogin.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
timedatectl shows Europe/London → auto-timezone checked off (manual
set-timezone re-trigger not exercised). `nomarchy-menu keybinds` opens
fine from a terminal, so item 26 is confirmed to be only the
`$mod, question` bind, not the menu.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
PASS: waybar supervisor resilience, snapshots polkit+GUI (restore/
rollback residual queued), sys-rebuild, night-light on/off,
auto-commit, hibernate single-unlock (1b0eeea confirmed).
Partial: auto-timezone (enabled; zone check pending).
Item 26: SUPER+? never fires — `$mod, question` can't match because a
real ? carries SHIFT in the modmask; fix is `$mod SHIFT, question`.
SUPER+SHIFT+K no-op was the comma-layout gate working (none
configured) — retest annotated in HARDWARE-QUEUE.
PROPOSED: menu parity for downstream-flake options (Bernardo) — every
hand-edit-the-flake option gets a menu writer, per the in-flake-state
philosophy.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Plan/reason/review on the strong model (Fable 5); delegate mechanical,
fully-specified subtasks (grep sweeps, table reconciliation, repeated
edits, log summarization) to cheaper models (haiku/sonnet) via subagent
model overrides — the strong model writes the spec and owns the commit.
Plus narrow-reading and terse-writing rules, and a cheaper --model
allowance for low-stakes headless iterations.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- HARDWARE-QUEUE: media keys + touchpad gestures (dccceb4) checked off
— verified working on the Latitude 5410 by Bernardo.
- BACKLOG item 25 (NOW): swaync text unreadable on summer-day. Traced:
swaync.nix pairs @subtext on @base, but summer-day (and flexoki-light)
use subtext as text-ON-surface (== base), and summer-day also has
surface == text — self-colored body text and buttons.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Recovers the crashed iteration #7 (items 21+23, Latitude QA findings):
- nomarchy-waybar supervisor (waybar.nix, exec-once'd from
hyprland.nix): any waybar exit respawns the bar; crash-loop guard
(5 fast exits -> critical notify + stop); TERM trapped for a real
stop. Fixes the bar-less session after a theme-switch crash.
- nomarchy-theme-sync prefers a clean `pkill -x waybar` (supervisor
restart with fresh config+style) over the crash-prone in-place
SIGUSR2 reload when the supervisor is running; SIGUSR2 stays as the
unsupervised fallback.
- sys-rebuild: snapshot-first system rebuild against the CURRENT lock
(no `nix flake update`) — the no-update twin of sys-update;
README §3/§5 + motd list it.
Verified: V0 re-run green after crash recovery (flake check --no-build
+ py_compile); V1/V2 per the crashed session's journal entry (HM
renders exec-once=nomarchy-waybar; headless software-GL VM:
SIGKILL -> new pid, clean kill -> respawn, SIGUSR2 -> alive).
V3 pending on the Latitude (queued in HARDWARE-QUEUE.md).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
SUPER+SHIFT+K → hyprctl switchxkblayout current next, rendered only
when the session layout has a comma — the same gate as the Waybar
language indicator. Bind data is a separate multiLayoutBinds list in
keybinds.nix (still one source); hyprland.nix and the rofi cheatsheet
gate on the identical condition, so bind and cheatsheet cannot drift.
`current` targets the focused keyboard, so per-device-overridden
boards (single layout) are a no-op.
Parity: hyprland/language (with the keyboard glyph) added to both
summer waybar.jsonc whole-swaps + #language in their CSS. Static JSON
cannot eval-gate, so on summer themes the module shows even with one
layout — parity-over-minimalism, deliberate. Found en route:
summer-night has an idle_inhibitor the generated bar lacks (reverse
gap) — folded into backlog item 13.
Also — agent/BACKLOG.md gains items 21-24 from Bernardo's Latitude
5410 hardware QA (waybar crash on theme switch + no respawn; nameless
network-menu rows; a no-update sys-rebuild; Back-everywhere audit),
queued ahead of everything; HARDWARE-QUEUE's snapshot re-test now
carries the "update to >= a47aa3a + relogin" precondition, since
findings 5/6 match the pre-polkit-agent behavior exactly.
Verified: V0 (flake check, both summer JSONs parse) + V1 both
directions (default single-layout build renders no bind; us,de via
extendModules renders bind AND cheatsheet row). V3 cycle check queued.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- templates/downstream/home.nix: commented "Webcam tuning" entries
(cameractrls, v4l-utils) — bare packages, so template not toggle,
per the option-surface rule; rarely needed since the IR-hide covers
the common dual-sensor case.
- agent/BACKLOG.md PROPOSED: the portal/Flatpak camera picker still
lists the internal IR sensor (libcamera path, untouched by the v4l2
rule) — three options laid out, recommending document-only until a
T14s-style machine is available to test a libcamera-side rule.
Closes BACKLOG NOW#7. Verified: V0 (flake check green; template still
evaluates through mkFlake).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Re-diagnosis of the "btrfs-assistant 2.2 segfaults on launch" bug, and
the fix is not the planned nixpkgs override:
1. gdb: the crash is btrfs_util_subvolume_iterator_next() —
libbtrfsutil's UNPRIVILEGED subvolume-iteration path in btrfs-progs
6.17.1 (upstream-fixed after 6.17.1, kdave/btrfs-progs 886571653).
Symbol versions were checked and match: not ABI/link drift.
2. VM A/B on real btrfs: as root it works (exit 0); as a user it
segfaults (139). The pkexec launcher runs it as root — so the GUI
path was fine all along, except…
3. …the distro shipped NO polkit authentication agent: every pkexec
prompt in the session failed silently. That is the actual root
cause of "the GUI doesn't open".
Shipped: hyprpolkitagent (Hyprland's Qt agent — the prompt is
Stylix-themed) via exec-once in hyprland.nix; `nomarchy-menu snapshot`
prefers btrfs-assistant-launcher again, with the nomarchy-snapshots
fzf flow kept as fallback (and for SSH); checks.snapshot-gui guards
the root path + offscreen GUI event loop on a real btrfs volume so a
lock bump can't silently regress it. No btrfs-progs patch: all our
flows are root-side; the unprivileged fix arrives with a lock bump.
Verified: V0 (flake check) + V2 (snapshot-gui check, executed green) +
V1 (HM generation builds; hyprland.conf carries the agent exec-once;
generated menu passes bash -n with the launcher wiring). Remaining V3
queued: first on-hardware themed polkit prompt.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
First successful run of .gitea/workflows/check.yml on the live
act_runner: the eval tier is confirmed end to end. What remains of
item 20 is optional infra ([human]): a KVM/nix host-mode runner to
unlock the commented vm-checks job.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
modules/nixos/oom.nix — running out of memory now kills the offending
process (with a desktop notification via systembus-notify) instead of
freezing the desktop for minutes until the kernel OOM killer fires.
earlyoom over systemd-oomd, deliberately: oomd kills whole cgroups,
and a Hyprland session runs as ONE scope (nothing here spawns per-app
systemd scopes, unlike GNOME) — under pressure oomd would take out the
entire desktop to save it. earlyoom kills the single largest process
before the thrash point. nixpkgs default-enables oomd in an inert
state (no slices monitored); it is disabled outright so there is one
owner. Session plumbing is --avoid-listed (unanchored — NixOS wrappers
rename comm to .foo-wrapped); no --prefer tuning, largest-RSS
selection already finds the hog. All mkDefault; opt out with
services.earlyoom.enable = false (README note added).
Verified: V0 (flake check) + V2 — new checks.oom-protection
runNixOSTest, executed: a chunked allocator (686 MB peak, 1 GB VM) is
SIGTERM'd by earlyoom in 0.1 s, a bystander unit survives (the
process-level granularity the module exists for), and systemd-oomd is
asserted inactive. 25 s runtime.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The runner was alive all along (gitea/act_runner, ubuntu-latest label);
no run ever appeared because Gitea only reads .gitea/workflows/ (or
.github/workflows/) and ignores the .forgejo/ path the workflow shipped
under. The legacy repo's .gitea/workflows/check.yml was the clue.
References corrected (workflow header, TESTING.md, agent files); this
push doubles as the first live trigger of the workflow.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
.forgejo/workflows/check.yml runs on every push to main/v1 (+ manual
dispatch): nix flake check --no-build (full module-system eval incl.
the downstream template through mkFlake), py_compile of
nomarchy-theme-sync, and bash -n over tracked .sh files. The always-on
net under direct-to-main pushes — first slice of the ROADMAP lock-bump
CI item.
Scoped to the eval tier deliberately: the instance's runner is an
act_runner docker container (no systemd, no /dev/kvm — established
from the legacy repo's .gitea/workflows/check.yml, which ran 57 times
on it), so the checks.* VM suite and real builds can't run there. A
commented vm-checks job documents the KVM-runner upgrade path; the
legacy workflow's container gotchas (nixbld setup for the single-user
installer, sandbox=false for Stylix IFD, Nix pinned 2.31.5 vs
lazy-trees, no JS actions past node20) are carried over verbatim in
the header. docs/TESTING.md §1b documents what a green run does and
does not mean.
Verified: V0 locally (the same check commands, minus the container
Nix install) + YAML parse. A real green run depends on the runner
still being registered — not API-visible unauthenticated, so that is
queued as [human] BACKLOG item 20.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
settings.autoCommit — toggled from System › Auto-commit (self-gated on
the flake being a git repo) — makes every nomarchy-theme-sync apply/set
also git-commit theme-state.json in the downstream flake, so settings
history is `git log`. Off by default.
Design: the flag is live-read by the tool on each write (nothing in Nix
consumes it → instant toggle, no rebuild, no new option); the commit is
pathspec-limited to theme-state.json so unrelated dirty files are never
swept up; it fires when the flag is on before OR after the write, so
the disable-toggle itself lands in history; same-value writes no-op
(diff against HEAD); missing git identity falls back to
Nomarchy <nomarchy@localhost>; `bg next` is deliberately excluded.
Rider fix: `get` now prints booleans JSON-style (true, not Python's
True) — un-sticking the System menu's "Auto timezone (on/off)" label,
whose `= true` comparison could never match. All shell consumers
already normalise true|True, so no other behavior change.
Verified: V0 (py_compile, nix flake check) + V1 (HM generation builds,
generated nomarchy-menu passes bash -n, wiring present in the rendered
script) + a 7-assertion sandbox-repo round-trip covering enable/set/
same-value/disable/post-disable/apply paths and the identity fallback.
Remaining: on-machine check queued in agent/HARDWARE-QUEUE.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Checked guesses against the tree before adding (GC/auto-optimise/boot
limit and all 21 theme previews already ship — not added). New items,
all confirmed absent:
- NOW#2 CI checks-on-push, promoted because agents push main directly
- NOW#3 memory-pressure protection (systemd-oomd/earlyoom; nothing
guards the session from OOM today and the release bump died to one)
- NEXT#8 PDF/image viewers + xdg-mime defaults (template ships only mpv)
- NEXT#9 update diff (nvd) + menu-driven rollback
- NEXT#10 nomarchy-doctor health check
- NEXT#11 state-file validation + friendly eval errors
- NEXT#12 screen recording in Capture (+ stop via Waybar indicator)
- NEXT#13 niceties: idle-inhibit toggle, low-battery notify, hyprpicker
- LATER: OCR screenshot-to-text
- Decisions: zram-vs-hibernation, default browser for mime defaults
- docs item now includes a recovery runbook
Verified: V0 — docs-only.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A git-tracked agent/ directory so AI agents can iterate on the distro
unattended (runner-agnostic: /loop, headless claude -p, or a fresh
manual session — all state lives in the checkout, per the distro's own
philosophy):
- LOOP.md — the iteration protocol: orient → pick one BACKLOG task →
verify up the V0–V3 ladder → commit+push main → record. Safety rails
(v1 untouchable, no force-push, no surprise lock bumps) and
stop-and-escalate conditions.
- BACKLOG.md — the forward half of docs/ROADMAP.md reworked into a
prioritized queue (5 NOW / 6 NEXT / LATER / PROPOSED / Decisions);
ROADMAP.md stays the design/decision record + shipped log.
- GOALS.md — the four pillars (stable > reproducible/zero-hidden-state >
effortless config > beautiful), quality bars, non-goals.
- CONVENTIONS.md — coding/design rules (in-flake state, menu placement,
Waybar whole-swap parity, toggle-vs-package, no formatter).
- MEMORY.md — curated hard-won lessons (VM recipes, btrfs-assistant
segfault watch, rofi/WirePlumber/hyprlock gotchas).
- HARDWARE-QUEUE.md — every pending V3 on-hardware check collected from
the ROADMAP, with exact steps, split by machine.
- JOURNAL.md — append-only iteration log, seeded with this bootstrap.
Plus a root CLAUDE.md entry point and README/ROADMAP pointers.
Verified: V0 — docs-only; nix flake check --no-build green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Phase 3 of the in-flake-state philosophy (mirroring night-light + the
keyboard-layout watcher): a System ▸ Display menu item picks an output's
resolution and persists it into the flake instead of a runtime-only tweak.
- rofi.nix: the `display` case reads `hyprctl monitors -j` (auto-skips the
chooser with one output), lists the output's advertised modes (rounded
refresh, deduped, highest-first) plus the preferred/highres/highrr
tokens, applies the pick INSTANTLY via `hyprctl keyword monitor` —
keeping the output's current position + scale, so only the resolution
changes — then persists it with `nomarchy-theme-sync set
settings.monitors.<name> <mode> --no-switch` (no rebuild).
- theme.nix: seed `settings.monitors = {}` (output-name -> "WxH@R") so a
sparse/older state file still evaluates; exposed via nomarchy.settings.
- hyprland.nix: `resolvedMonitors` overlays settings.monitors onto
nomarchy.monitors by name — a declared output keeps its other fields
with only the resolution swapped, an undeclared output (wildcard-only,
e.g. the laptop panel) becomes a fresh rule. So a later rebuild bakes
the menu pick into the generated monitor rule — the monitor twin of the
keyboard graduation. The menu pick wins over a hand-set resolution (it's
the explicit live action).
Monitor connector names are dot-free, so a direct dotted set-path is safe
(no whole-map read-modify-write like the keyboard watcher needs).
VM-verified via a headless-Hyprland nixosTest (run + removed; too heavy
for a CI gate): the graduation baked the rule into the generated
hyprland.conf, the set/get round-trip persisted, and `hyprctl keyword
monitor` actually changed the reported resolution. Only the interactive
rofi picker isn't headlessly driveable (same caveat as the keyboard watcher).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Completes the remaining "Distro branding, round 2" work.
GRUB theme: hosts/live.nix builds nomarchyGrubTheme whose background is
the same composed splash as the isolinux/BIOS splash (accent logo on
base), with a palette-coloured boot menu and accent timeout bar. Derived
from nixos-grub2-theme only to reuse its bundled DejaVu .pf2; the stock
NixOS logo is dropped (ours is in the background). Wired as
isoImage.grubTheme so UEFI boot matches BIOS.
distroId: set system.nixos.distroId = "nomarchy" so os-release is honest
(ID=nomarchy, ID_LIKE=nixos). Safe — switch-to-configuration builds its
"is this NixOS?" guard from the configured distroId (and /etc/NIXOS is
the fallback), so rebuilds keep working. isNixos→false blanks the
nixos.org URLs, so extraOSReleaseArgs restores them to the project.
Adds checks.distro-id (runNixOSTest): boots distroId=nomarchy and runs
switch-to-configuration dry-activate green; os-release eval-verified from
the real downstream config; grub theme built + structure-verified.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add checks.battery-charge-limit (runNixOSTest): the test_power module
fakes a Mains adapter, toggling ac_online emits a real power_supply
uevent, and the test asserts the udev rule restarts the oneshot (changed
InvocationID). Verifies the re-apply *trigger* at runtime, beyond the
earlier eval check; the sysfs write still needs real hardware. Also
records the SSH_AUTH_SOCK rendered-artifact verification in the roadmap.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The agent's SSH socket was only exported via the zsh integration, so a
GUI client launched outside a shell (rofi launcher, autostarted apps)
couldn't use gpg-agent for SSH. Add a home.sessionVariables export
resolving the socket with gpgconf --list-dirs agent-ssh-socket — the
same lookup the shell integration uses, so the two can't drift — which
reaches GUI apps via the login shell that starts Hyprland (the path
NIXOS_OZONE_WL already takes).
Eval-verified: the var resolves to the gpgconf substitution; the home
config still builds.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The threshold was written boot-only, so a firmware that clears
charge_control_end_threshold on unplug would silently lose the cap
until the next reboot. Add a services.udev rule on the mains adapter
(ATTR{type}=="Mains", vendor-neutral) that restarts the oneshot via
systemctl --no-block on every AC state change. Uses restart (not
try-restart) so it re-applies even if the boot run was inactive.
Eval-verified: rule present when the charge limit is on, absent when
off; downstream-template-system still evaluates.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A built-in RGB+IR webcam (common on recent ThinkPads, e.g. the T14s AMD
Gen 4) exposes its IR face-unlock sensor as a second, identically-named
"Integrated Camera"; apps that pick it show a dark, 8-bit-greyscale
image. nomarchy.hardware.camera.hideIrSensor disables that node on the
V4L2 PipeWire path (WirePlumber monitor.v4l2.rules) so only the colour
camera is offered, matched by card name (overridable irMatch).
- libcamera is left untouched: an external camera you plug in is never
affected, and surgical internal-only libcamera scoping isn't possible
(its distinguishing props bind after the monitor rule runs).
- Only the PipeWire node is hidden; the kernel /dev/video* stays open,
so Howdy face-unlock still reads the IR sensor directly.
- The installer's hardware-db.sh auto-detects a paired RGB+IR webcam
from /sys/class/video4linux/*/name and bakes the toggle into
system.nix; commented example added to the downstream template.
Validated live on a T14s: collapses the four enumerated "Integrated
Camera" entries to one colour source; the exact shipped regex disables
the IR node (confirmed in the WirePlumber log) and leaves libcamera
intact.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
nomarchy.services.tailscale now makes the login user the Tailscale
operator (services.tailscale.extraSetFlags = ["--operator=<username>"],
mkDefault), so `tailscale up/down/set` — and the System → VPN menu's
Tailscale controls — run without sudo. The user is already in wheel, so
the operator grant is no real new privilege, just skips the prompt.
nomarchy-vpn's Tailscale actions are now operator-aware: a ts_priv
helper runs the subcommand inline and only falls back to a sudo terminal
if the operator grant is absent (downstream dropped it). Connect brings
an authed-but-down node up inline; a node that still needs interactive
login goes to a terminal so the auth URL is visible.
Option description + template comment updated; ROADMAP follow-up closed.
System + home builds green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
