Files

Bernardo Magri ba8963a385

Check / eval (push) Successful in 4m11s

Details

test(hibernate): #76 slice 3 — V2 hibernate→resume VM check (PASS)

Add checks.hibernate: a runNixOSTest modeled on nixpkgs' hibernate.nix,
wired to the actual risk #76 introduces — coexistence with the zram swap
that oom.nix now enables by default. It imports oom.nix, puts the disk
swap at a lower priority than zram with an explicit boot.resumeDevice,
and uses systemd-initrd. The test drives a full
hibernate -> power-off -> resume -> crash cycle: a volatile ramfs marker
survives resume but NOT a crash-boot (proving genuine resume), and it
asserts zram sits at priority 100 while the disk swap is lower, so the
hibernate image lands on disk rather than volatile zram.

What QEMU genuinely can't do — feed a boot-time resume_offset for a btrfs
swapfile (offset only exists after first-boot allocation) or drive a
LUKS-initrd unlock — stays a V3 laptop check (HARDWARE-QUEUE, dev box
already runs that exact layout), plus the no-swap notify in a real
session. With this, all V0/V1/V2 work on #76 is done; #76 collapses to a
[blocked:hw] stub and the design record moves to ROADMAP.

Verification: V2 PASS — nix build .#checks.x86_64-linux.hibernate built
the driver fresh and exited 0. nix flake check --no-build green.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-07-10 19:45:47 +01:00

134 KiB

Raw Blame History

Journal — append-only iteration log

One entry per iteration, newest first. This is the loop's short-term memory: the next session reads the last 3–5 entries to orient. Keep an entry under ~15 lines; durable lessons go to MEMORY.md, not here.

Template:

## YYYY-MM-DD — <task title> (<commit sha, or "no commit">)
- **Task:** which BACKLOG item (or QA sweep / bootstrap / escalation)
- **Did:** what changed, in 2–4 lines
- **Verified:** tier reached (V0/V1/V2) + the actual commands/checks run
- **Pending:** V3 queued? follow-ups filed? anything [stuck]?
- **Next suggestion:** what the following iteration should probably take

2026-07-10 — #76 slice 3: hibernate VM test (V2 PASS) → #76 agent-complete

Task: BACKLOG #76 slice 3 — the mandatory V2 hibernate test + V3 queue.
Did: Added checks.hibernate (flake.nix), modeled on nixpkgs' hibernate.nix but wired to Nomarchy's #76 risk: imports oom.nix (zram default-on), disk swap at lower priority + explicit boot.resumeDevice, systemd-initrd. Test drives hibernate→power-off→resume→crash: ramfs marker survives resume, a crash-boot loses it (real resume, not a survivor), and asserts zram@100 + disk-swap<100 so the image lands on disk not volatile zram. Queued the V3 laptop steps (real LUKS+@swap+offset resume + no-swap notify) in HARDWARE-QUEUE. Collapsed #76 to a [blocked:hw] stub; design record → ROADMAP § "Hibernation + zram by default".
Verified: V2 PASS — nix build .#checks.x86_64-linux.hibernate built the driver fresh and exited 0 (all assertions held). nix flake check --no-build green.
Pending: only V3 (dev box / laptop) — #76's last check; close on PASS.
Next suggestion: #76 is now blocked:hw; NEXT head falls to LATER/ PROPOSED triage or a QA sweep — pick per LOOP.md orientation next session.

2026-07-10 — #76 slice 2: Hibernate notify-on-failure

Task: BACKLOG #76 remaining slice 2 — per the settled "keep row, notify on failure" call.
Did: modules/home/rofi.nix power menu Hibernate arm now runs systemctl hibernate 2>/dev/null || notify-send "Hibernate" "…see docs/MIGRATION.md". On success the call blocks until resume, so the notify only fires on a genuine failure (e.g. swap=0 no-swap machine). Row stays unconditional (no hiding).
Verified: V1 — nix build …homeConfigurations.nomarchy.activationPackage green (rebuilds nomarchy-menu.drv; writeShellScriptBin runs bash -n at build). Control-flow proof on the arm extracted from the built script: failing systemctl → notify fires; succeeding → no notify. nix flake check --no-build green.
Pending: slice 3 = the V2 hibernate→resume VM test + V3 laptop (Hibernate→power off→resume on the LUKS+@swap machine; plus confirm the no-swap notify surfaces in a real session). Both remain open on #76.
Next suggestion: slice 3 — add a checks.* hibernate VM test (as far as QEMU allows) and queue the V3 laptop steps in HARDWARE-QUEUE.

2026-07-10 — #76 slice 1: MIGRATION.md hibernation runbook

Task: BACKLOG #76 remaining slice 1 — docs runbook for enabling hibernation on an existing machine (per the settled "docs not tool" call).
Did: Added § "Enabling hibernation on an existing machine (no reinstall)" to docs/MIGRATION.md: create @swap subvol (subvolid=5) + mkswapfile (NOCOW), read map-swapfile -r offset, wire fileSystems."/swap" + swapDevices + resumeDevice + resume_offset into system.nix. Flagged the mount as required on the hand-edit path (installer gets it from disko-gen hardware-config). Notes zram-priority reservation, no-LUKS variant, swap=0 opt-out. Worked example = this dev machine.
Verified: V0 — read-only commands run live against this machine's real LUKS(cryptroot)+btrfs(@)+/swap/swapfile layout (RAM 31G, fsuuid, findmnt SOURCE, mkswapfile flags, zram@100 vs swapfile@-1 confirms the design intent). nix flake check --no-build green.
Pending: slice 2 — Hibernate row notify-on-failure (V1/V2); then the V2 hibernate VM test + V3 laptop.
Next suggestion: slice 2 (menu notify-on-failure).

2026-07-10 — #76 hibernation design calls (no commit yet → this bookkeeping)

Task: BACKLOG #76 design slice — Bernardo made the open calls.
Did: Recon showed the hibernation half is already built for new installs (installer defaults swap=RAM hibernation-ready @swap swapfile; patch-template.py writes swapDevices/resumeDevice/resume_offset; Power menu Hibernate row + LUKS/hyprlock handling exist). Settled 3 calls: (1) swap sizing = exactly RAM (no change); (2) migration = docs runbook in MIGRATION.md, not a tool; (3) no-swap Hibernate = keep row, notify on failure. Folded into BACKLOG #76 (design ✓, re-sliced remaining: MIGRATION.md runbook + notify-on-failure menu) and Decisions.
Verified: V0 — nix flake check --no-build green (docs-only).
Pending: two remaining slices below; V2 hibernate VM test + V3 laptop.
Next suggestion: slice 1 — docs/MIGRATION.md hibernation-enable runbook (V0), then slice 2 — Hibernate notify-on-failure (V1/V2).

2026-07-10 — #76 zram slice: compressed-RAM swap default

Task: BACKLOG #76 ([big]) — split; took the low-risk zram-default slice.
Did: zramSwap on by default in modules/nixos/oom.nix (zstd, 50% RAM, priority 100) as the memory-pressure layer before earlyoom, following the earlyoom mkDefault precedent (no toggle). High priority keeps day-to-day paging in zram so a disk swapfile stays reserved for #76's hibernate image. Added checks.zram-swap runNixOSTest (imports oom.nix). Split #76 in BACKLOG → remainder is the hibernation half (swapfile/resume/installer).
Verified: V2 — nix build .#checks…zram-swap PASS: /dev/zram0 active in /proc/swaps, [zstd] comp_algorithm, priority 100. V1 full distro toplevel builds. V0 flake check --no-build green.
Pending: hibernation half of #76 still open (needs hibernate→resume VM test + V3 laptop). No V3 for zram (under-load tuning is secondary).
Next suggestion: #76 hibernation half — design default swapSize + the resume/installer wiring; it's the boot-critical, V2-hibernate + V3 part.

2026-07-10 — #41 hardware-confirmed: OpenFile portal floats (V3 close)

Task: BACKLOG #41 — final on-hardware confirmation (Bernardo reported PASS).
Did: #41 already out of NEXT + portal rule shipped (hyprland.nix xdg-desktop-portal-gtk float/center). Bernardo verified the GTK OpenFile file-picker floats on real hardware; dropped the "OpenFile not re-probed / optional residual" caveats in HARDWARE-QUEUE — #41 fully closed.
Verified: V3 (Bernardo, Latitude). No code change. Bookkeeping only; QA-sweep guards run green this session: nix flake check --no-build, option-docs/template-sot/theme-sync-validate builds, contrast 24×7.
Pending: none for #41. NEXT head is #76 (hibernation + zram, [big]).
Next suggestion: split #76 — take the low-risk zram-default slice (V2-testable headlessly) before the boot-critical hibernate/resume work.

2026-07-10 — Close #41 GTK portal float rule

Task: BACKLOG #41 residual (GTK file-chooser portal class).
Did: Added float+center for xdg-desktop-portal-gtk / .xdg-desktop-portal-gtk-wrapped (desktop + libexec name from package). Polkit already PASS on Latitude. Removed #41 from NEXT; only #20 remains.
Verified: V0 windowrule-syntax check path unchanged; class from package xdg-desktop-portal-gtk.desktop / community configs. SoftGL still cannot materialize portal dialogs.
Pending: optional hardware spot-check OpenFile floats (HARDWARE-QUEUE).
Next suggestion: #20 is [human] only — or LATER/PROPOSED triage.

2026-07-10 — Product calls: charge live, identity, btop, NVIDIA skip

Task: Bernardo decisions on remaining PROPOSED.
Did: (1) Instant charge-limit: udev GROUP=users on threshold, menu/powermgmt live sysfs + state, oneshot reads live theme-state + AC re-apply. (2) Look & Feel already had Theme/wallpaper/night light; added Reset wallpaper (auto). (3) NVIDIA first-class deferred past v1 (HARDWARE.md). (4) Identity retunes: white L-steps, lumon subtext, hackerman surface+UI warn/bad, matte-black subtext, miasma bad. (5) Hand btop.theme for 9 identity/whole-swap slugs from palette map.
Verified: V0 contrast 24×7; V1 battery-charge-limit VM (udev restart). V3 live charge write on real laptop (HARDWARE-QUEUE).
Next suggestion: only #20/#41 blocked, or Decisions.

2026-07-10 — Summer pair polish + status CSS + preview recapture

Task: finish agent-able PROPOSED theme leftovers.
Did: summer-day/night — JSON hierarchy (night subtext/muted/overlay ladder; day overlay≠muted); waybar CSS tokens track JSON roles; font 13px both; day battery 25/10; VPN pill + full status states (.recording/.available/.on/.activated/battery warn). Recaptured executive-slate + neon-glass preview.png via theme-shot desktop (480×270). Closed those PROPOSED lines.
Verified: V0 contrast 24×7 + wholeswap; V2 theme-shot summer-day/night/executive-slate/neon-glass — viewed bar crops + previews. Identity taste retunes stay PROPOSED.
Next suggestion: only blocked NEXT (#20/#41) or product Decisions.

2026-07-10 — #74 matrix flags + #75 V2 install (default arm)

Task: NEXT #74 unattended matrix · #75 SoT V2 install after ISO.
Did: NOMARCHY_TEST_SWAP_GB / NOMARCHY_TEST_NO_LUKS drive tools/test-install.sh go.sh + first-boot LUKS skip; nixpkgs OVMF fallback. Full offline install V2 (SWAP=2 LUKS) after live ISO rebuild with starter packages — first-boot themed Boreal, no red banner. Closed #74/#75. Other matrix arms scripted (not full-ISO re-run).
Verified: V2 — install powered off ~460s; first-boot.png viewed (/tmp/nomarchy-vm-75/first-boot.png): themed wallpaper + waybar (logo, clock, modules), doctor glyph, no autogen banner.
Next suggestion: NEXT only #20 [human] + #41 [blocked:hw].

2026-07-10 — Loop batch land #64–#73 (parallel worktrees)

Task: agent-actionable NEXT #64–#73 via worktree agents; land on main.
Did: #64 MIGRATION snapshots · #65 real-uid chown · #66 theme-state fail-closed reader · #67/#68 theme fidelity+ANSI · #69/#70 audit+import tools · #71 IR portal docs · #72 template-sot check · #73 MOTD/first-boot fingerprint+doctor tips. Agents did not push; parent cherry-picked (stripped agent BACKLOG races). Closed #64–#73.
Verified: V0/V1 — theme-contrast 24×7 pass; template-sot green; bash -n install; flake check --no-build green.
Next suggestion: #74 unattended matrix + #75 V2 install (KVM).

2026-07-10 — Promote agent-actionable PROPOSED → NEXT #64–#75

Task: Bernardo: bump non-hw/non-human PROPOSED categories 1+2.
Did: #64 MIGRATION snapshots · #65 installer chown · #66 mkFlake theme-state errors · #67 theme fidelity nits · #68 #52 residual ANSI · #69 audit identity exemptions · #70 import-pipeline hierarchy · #71 IR portal docs (a) · #72 template SoT CI · #73 hardware hints residual · #74 unattended install matrix · #75 SoT V2 install. Breadcrumbs in PROPOSED; actionable items sit above #20/#41 in NEXT.
Verified: V0 (backlog only).
Next suggestion: queue head #64 (docs) or batch #67/#68 themes.

2026-07-10 — PROPOSED cleanup (stale exam residue)

Task: Bernardo: clean stale PROPOSED items after A/B/C + #52 ship.
Did: Rewrote PROPOSED to open work only — dropped #52 shipped high-ROI bullets, neon-glass quarantine, empty/shipped installer audit sections, template seed drift (keys already present), soft gaps fixed by #50, and “no action” kanagawa/boreal notes. Kept product day-2, installer follow-ups, theme polish residuals, tooling.
Verified: V0 (backlog hygiene); spot-checked template keys, latte/mocha #52 fixes, usbutils, package exports.
Next suggestion: triage the lean PROPOSED list or HARDWARE-QUEUE.

2026-07-10 — Pending closeout: #14, #57 V2, #41 slice

Task: pending V2/CI/float leftovers after A/B/C batch.
Did: #14 closed — first scheduled lock bump landed 8fded63 (2026-07-06 Mon 05:20 UTC) as nomarchy-bump / chore(lock): scheduled upstream bump (Actions token push works). #57 V2 — checks.live-install-entry asserts live HM gen has nomarchy-install.desktop (Name/Terminal/Exec) + installer bin + Tools menu gate in nomarchy-menu. #41 slice — float/center rules for hyprpolkitagent + pinentry-qt from package app-id strings; tools/capture-float-classes.nix harness (softGL empty clients); residual GTK portal stays NEXT #41 [blocked:hw]. HARDWARE-QUEUE float confirm steps added.
Verified: V2 live-install-entry green; windowrule-syntax green. SoftGL class capture inconclusive (empty clients). V3 float confirm
- fingerprint/#60 still HARDWARE-QUEUE. #20 still [human].
Next suggestion: Bernardo: #20 KVM runner, or HARDWARE-QUEUE pass.

Task: NEXT #55 (System › Fingerprint) + #56 (rebuild failure UX).
Did: #55 — System row self-gated on fprintd-list; enroll/list/verify/ delete-all terminal flows; Use for login writes settings.fingerprint.pam (hardware.nix default from stateFile); HARDWARE.md §5 + HARDWARE-QUEUE V3 enroll. #56 — sys-update / sys-rebuild / home-update tee last 40 lines + nomarchy-doctor pointer; theme-sync run_switch notify/die text. Closed #55/#56.
Verified: V1 — HM template rebuild (menu + theme-sync); option-docs green; failure-footer shell smoke. V3 pending: real fingerprint enroll (queue). V2 pending: forced bad sys-rebuild in VM (footer text pure/logic covered).
Next suggestion: queue was A–C; remaining NEXT is blocked items (#14/#20/#41) or PROPOSED triage.

2026-07-10 — #58 nomarchy-detect-hw CLI

Task: NEXT #58 — post-install re-probe package (HARDWARE.md §8).
Did: pkgs/nomarchy-detect-hw wrapping installer’s hardware-db; human report + snippets + --raw protocol; overlay + systemPackages; HARDWARE.md §8 updated. Closed #58.
Verified: V1 — package builds; ran on this AMD laptop: correct MODULE/NOMARCHY lines (amd cpu/gpu, laptop, ssd, fingerprint, IR cam, NPU) and printable system.nix snippets.
Next suggestion: #55 fingerprint menu or #56 human rebuild errors.

2026-07-10 — #57 live Install Nomarchy affordance

Task: NEXT #57 — durable live-ISO install surface (not only toast).
Did: xdg.desktopEntries.nomarchy-install (Terminal=true) on live HM; Tools › Install Nomarchy self-gated on command -v nomarchy-install; welcome toast wording updated. Closed #57.
Verified: V1 — live config eval exposes entry name/exec; downstream-template-home rebuilds with menu gate. V2 pending: ISO rebuild + smoke that the .desktop appears in the session.
Next suggestion: #58 nomarchy-detect-hw CLI.

2026-07-10 — #63 generated Waybar logo + powermenu

Task: NEXT #63 — reverse parity: generated bar lacked whole-swap logo → menu and power icon → power menu.
Did: custom/nomarchy (left, Nomarchy.ttf U+F000) + custom/powermenu (right, U+F011 → nomarchy-menu power) in generated settings/style. Fixed Nix \uf000 pitfall (no \u escapes — use literal UTF-8). Whole-swaps already had both; no JSON/CSS edits. Closed #63.
Verified: V2 — THEME=tokyo-night theme-shot: left monogram N, right power glyph, accent styling; config modules-left/right correct. Crop paths under /tmp/theme-shot-tn2-*.png.
Next suggestion: product path #55 fingerprint menu (or #57 live install).

2026-07-10 — #60 name-agnostic system-battery scan

Task: NEXT #60 — align charge-limit / menu / waybar / doctor / installer chassis with battery-notify’s type=Battery + !scope=Device filter (BAT* globs miss CMB0).
Did: power.nix oneshot, rofi helpers (has_system_battery / has_charge_threshold), waybar powerprofile gate, doctor loop, hardware-db chassis, options.nix description. HARDWARE-QUEUE entry for non-BAT* confirm. Closed #60.
Verified: V1 — checks.option-docs, hardware-db-modules, battery-charge-limit (VM) green; no remaining BAT* globs in code paths. V3 pending: non-BAT* name on real hardware (queue entry).
Next suggestion: #63 generated Waybar powermenu + logo.

2026-07-10 — #59 NVIDIA commented system.nix guidance

Task: NEXT #59 — when common-gpu-nvidia is selected, emit commented PRIME/power/open-module guidance in installer-generated system.nix (ROCm/NPU pattern).
Did: has_nvidia from HW_PROFILES → patch JSON; build_installer_region emits comment-only plain NixOS block + HARDWARE.md §6 pointer. Docs §6 product direction marked shipped. Closed #59.
Verified: V0 — patch-template smoke with nvidia true (region contains hardware.nvidia.* comments + profile in flake); negative (nvidia false) has neither; bash -n install script.
Next suggestion: #60 BAT*-agnostic battery scan.

2026-07-10 — #62 whole-swap CSS self-containment CI gate

Task: NEXT #62 — cheap gate that waybar.css defines every @color it references (would have caught neon-glass) + preview/backgrounds present.
Did: tools/check-theme-wholeswap.py + checks.theme-wholeswap in flake.nix. README: five themes → four (drop quarantined neon-glass). Closed #62.
Verified: V0 — script OK on themes/ (24 presets, 4 whole-swaps); negative fixture with @text/@base and zero defines fails as expected; nix build .#checks.x86_64-linux.theme-wholeswap green.
Next suggestion: #59 NVIDIA install comments, or #60 BAT*-agnostic.

2026-07-10 — #61 Review Source line tracks offline/online

Task: NEXT #61 — Review panel Source must match the cache.nixos.org probe (OFFLINE), not always claim “no network needed”.
Did: SOURCE_NET gated on $OFFLINE in nomarchy-install.sh (offline: “pinned into the ISO, no network needed”; online: “may use network binary caches”). Extended check-install-safety.py contract #4 so the dual wording + $SOURCE_NET expansion can’t regress. Closed #61.
Verified: V0 — python3 tools/check-install-safety.py … OK; bash -n; nix build .#checks.x86_64-linux.installer-safety green. (Interactive gum Review is not headlessly driven; pure contract is the V0 bar for this string.)
Next suggestion: #62 whole-swap CSS self-containment CI gate.

2026-07-10 — Promote A/B/C PROPOSED → NEXT #55–#63 (Bernardo-directed)

Task: Bernardo accepted the high-value promote batch (A day-2, B hardware/stability, C cheap polish).
Did: Nine NEXT items — #55 fingerprint menu, #56 human rebuild errors, #57 live Install action; #58 nomarchy-detect-hw CLI, #59 NVIDIA install comments, #60 BAT*-agnostic power; #61 review offline/online source line, #62 whole-swap CSS CI gate, #63 generated Waybar powermenu+logo. Breadcrumbs left in PROPOSED; v1.0 track note updated. Still PROPOSED: first-boot tips residual, charge-limit instant, theme taste polish, NVIDIA first-class.
Verified: V0 (backlog only) — 55–63 present under NEXT; promoted pitches removed from PROPOSED.
Next suggestion: agent-actionable queue head is #61 or #62 (tiny V0), then #59 / #60 / #63; product path #55–#57.

2026-07-09 — #54 V2 + #33 scrollbar + #43 MOTD tip

Task: NEXT #54 (installer safety V2), #33 (rofi scrollbar), #43 optional MOTD/first-boot tip (no V3 LVFS flash).
Did: checks.installer-safety + tools/test-install-safety.sh (compose-fail / disk-warn LUKS magic / hm-hint FORCE hooks). Enabled themed rofi scrollbars in generated rofi.nix + all whole-swap .rasi. MOTD + control-center first-boot tip for System › Firmware. Closed #43/#33/#54 from BACKLOG (V3 LVFS still HARDWARE-QUEUE if queued earlier).
Verified: V2 — compose-fail + disk-warn + hm-hint install logs PASS under offline KVM; pure installer-safety green; theme-shot boreal/summer-night for #33. V0 flake check + bash -n.
Pending: #43 real LVFS flash remains hardware-only (not this session). Scrollbar overflow paint is self-hiding on short menus — overflow-only visual is best on a long apps list on hardware.
Next suggestion: #14 watch first bump run, or #41 hw float classes.

2026-07-09 — #51+#52 V2 batch: doctor glyph + btop bake

Task: NEXT #51 (doctor Waybar glyph) + #52 (btop fidelity render).
Did: theme-shot harness now asserts nomarchy-doctor-status (class:bad + glyph), guest btop/themes/nomarchy.theme, pokes waybar RTMIN+10, best-effort Ghostty+btop shot. Fixed the doctor tripwire so it actually shows: absolute path to nomarchy-doctor (PATH self-hide), ANSI-stripped tooltip, 󰀨 glyph, format = "{}", signal 10 on generated + whole-swap bars.
Verified: V2 — THEME=<slug> nix build --impure -f tools/theme-shot.nix green for rose-pine, summer-night, catppuccin, everforest, vantablack; viewed bar crops: red alert-circle before the bell on generated (rose-pine) and whole-swap (summer-night). Scripted btop.theme key checks (Dawn/inactive_fg/Mocha/vantablack) all OK. V0 flake check + contrast 24×7 pass. Ghostty under softGL never opened → btop TUI visual queued V3.
Pending: HARDWARE-QUEUE btop TUI spot-check (five slugs).
Next suggestion: #54 installer V2 scenarios, or #33 rofi scrollbar.

Task: V2 the shipped Firmware menu (#43). theme-shot render of nomarchy-menu system.
Did/Verified: System submenu renders themed + correct (Bluetooth shows, Printers/Snapshots correctly self-gated out). The Firmware row sits below the visible fold; a send_chars "firm" filter did NOT surface it in the capture VM. Disambiguated the anomaly against the real system: /sw/bin/fwupdmgr exists in the live toplevel (same as the working Bluetooth row's blueman-manager), so command -v fwupdmgr succeeds → the row shows on a real install. The theme-shot filter view is a minimal-VM render artifact, not a bug. Incidental: a reddish doctor-status glyph appears in the bar (#51's indicator firing on the VM's degraded state).
Tier: V2 for the menu render + PATH-verified row presence; V3 (real LVFS flash) stays queued. Over-spent renders here chasing the anomaly — noted.
Next: wrap.

2026-07-09 — #44 V2: checks.doctor guards the hardware section

Task: close #44's V2 (was V1 + real-hw smoke). Extend checks.doctor.
Did: The bare test VM has none of the probe tools, so the existing healthy nomarchy-doctor run already drives all six hardware checks in skip mode. Added assertions that the skip rows appear (NetworkManager, fingerprint, fwupd, VA-API) and the sheet still exits 0 — i.e. the section runs without erroring under set -euo pipefail. No node changes (no closure bump); the fires paths keep the agent's real-hardware evidence.
Verified: V2 — nix build .#checks.x86_64-linux.doctor passes. Permanent headless regression guard now in place.
Next: #43 System-menu render (Firmware row).

2026-07-09 — #53 theme previews + neon-glass quarantine

Task: NEXT #53 — capture the last two missing previews + resolve the neon-glass broken whole-swap. Product call: quarantine.
Did: Confirmed the contract (working whole-swaps carry 12–23 @define-color blocks; waybar.nix:448 reads the whole-swap css raw, no palette prepend — so neon-glass's 0-@define-color css is genuinely broken). git rm themes/neon-glass/waybar.css → falls back to the generated bar from its (fine) JSON palette. Rendered executive-slate + neon-glass via theme-shot; captured 480×270 preview.png for both — all 24 themes now have previews.
Verified: V2 — viewed both renders: executive-slate clean; neon-glass bar now renders correctly (generated palette), no banner. V0 flake check green with the css removed.
Pending: none. Optional: hand btop.theme for the 3 identity themes; a hardware recapture to match the floating-terminal composition (nicety note).

2026-07-09 — #54 installer safety + polish (delegated, worktree)

Task: NEXT #54. Sonnet worktree agent (installer only).
Did: offline compose-lock failure fail-closes (reuses the cache.nixos.org probe → OFFLINE) instead of a network flake lock; disk-review warn on Windows/BitLocker/NTFS/LUKS signatures (before wipe); account password min 8 (mirrors LUKS); HM pre-activate failure drops a recovery-hint file on the target home. Diff reviewed + cherry-picked.
Verified: V0 flake check + V1 nix build .#nomarchy-install. V2 (offline fail path, NTFS-disk warning, pre-activate hint on a booted target) needs crafted scenarios — queued in slimmed #54. Password ≥8 is V0-complete.
Next: cheap scripted btop check for #52; then session wrap.

2026-07-09 — #52 theme fidelity Broken/high-ROI fixes (delegated, worktree)

Task: NEXT #52. Sonnet worktree agent (parallel, theme files only).
Did: Six defects fixed: rose-pine btop→Dawn, everforest/summer-night btop inactive_fg, vantablack role inversion, osaka-jade text/subtext+warn, catppuccin-latte mantle/accentAlt/ANSI, catppuccin Mocha btop. My call on a flagged conflict: the audit's accentAlt #ea76cb (Latte pink) fails the 3.0 contrast floor → used Latte mauve #8839ef (contrast-safe, canonical).
Verified: V0 — check-theme-contrast 24×7 all pass; flake check green. btop.theme visual = V2 render pending (batched). Filed a residual note (vantablack ansi[8], osaka-jade ansi[3] drift exposed by the fixes).
Next: land #54; batched V2 render of the btop themes + #51 doctor glyph.

2026-07-09 — #51 Waybar/menu behaviour bugs (delegated, worktree)

Task: NEXT #51. Sonnet worktree agent (parallel with #52/#54, no shared files).
Did: doctor indicator "text":""→"text":"󰒡" + class:bad on failure (waybar.nix); CC appearance toggles print "(requires rebuild)" like the other CC toggles; summer-day scroll e±1→r±1. Reviewed diff, cherry-picked.
Verified: V0 flake check + V1 (HM gen + nomarchy-control-center build) on main. V2 render (glyph on an unhealthy bar) folded into #44's checks.doctor.
Next: land #52/#54; batched V2.

2026-07-09 — Promote V2-capable PROPOSED items → NEXT #51–54 (Bernardo-directed)

Task: with the VM free (post-#46), Bernardo asked to bump the V2-requiring PROPOSED items too. His triage; I did the move.
Did: Four NEXT items — #51 Waybar/menu behaviour bugs (doctor indicator icon, CC appearance no-ops, summer-day scroll), #52 theme fidelity Broken/high-ROI (6 per-theme btop/palette defects; kept the audit block as its spec), #53 missing previews (executive-slate/neon-glass via the theme-shot recipe) + the neon-glass waybar finish-or-quarantine call, #54 installer safety+polish (offline fail-closed, destructive-data warning, password min 8, pre-activate recovery hint). Removed the promoted pitches (breadcrumbs left); left the non-selected V2 items (generated-bar affordances, summer-* CSS, fidelity/hierarchy polish, Live install cue, offline/online source line) in PROPOSED.
Verified: V0 (backlog only) — 51–54 present, no leftover pitch text.
Next suggestion: these are session-render / install-VM verifiable — take #51 or #52 (theme-shot), or #54 (test-install), on the next iteration.

2026-07-09 — #46 V2 verify: install swap=0 + LUKS fail-closed (`208b8d4`)

Task: NEXT #46 — exercise the full install path for the swap=0 + LUKS fail-closed contracts (previously only pure nix-instantiate). Headless KVM via a two-variant driver adapted from tools/test-install.sh.
Did: Built the live ISO, ran two offline unattended installs: (B fail-closed) no passphrase + no NOMARCHY_NO_LUKS → installer exits RC=1 at the encryption stage with the fail message; /dev/vda stays BLANK and disko destroy never runs. (A swap=0) SWAP_GB=0 + passphrase → install RC=0, subvol dump on /mnt shows @,@home,@log,@nix,@snapshots and no @swap; installed disk boots (after LUKS unlock) to a fully themed Boreal desktop, no error/autoconfig banner.
Verified: V2 — screenshots viewed: b-failclosed.png, a-progress-480.png (DONE_SWAP0 + subvol dump), a-first-boot.png (themed boot). Both contracts hold. #49's new installer-disko check now guards the swap contract at eval time too.
Pending: none for #46. Incidental: HM pre-activate warned in the offline chroot (xrdb: no display) yet first boot came up themed anyway — the "pre-activate failure → durable recovery hint" PROPOSED item still stands.
Next suggestion: promote V2-capable PROPOSED items now the VM is free.

2026-07-09 — #50 small code one-liners batch (V0/V1)

Task: NEXT #50 — four small fixes. Kept in-house (shares rofi.nix / flake.nix with the #47/#49 agents; done after they landed).
Did: Bluetooth menu row + handler self-gate on command -v blueman-manager (rofi.nix, like Printers); Control Center rollback list /tmp/nomarchy-gens → mktemp + cleanup; usbutils added to the nomarchy-install PATH (fingerprint lsusb probe); exported nomarchy-doctor/control-center/battery-notify as flake packages.
Verified: V0 flake check green; V1 — built the four flake packages (incl. the new exports), nomarchy-install (usbutils), and the HM generation (rofi + control-center bash -n via writeShell*). All exit 0.
Pending: none.
Next suggestion: V2-capable promotions.

2026-07-09 — #49 pure-contract checks.* batch (delegated, worktree)

Task: NEXT #49 — add pure (no-VM) checks.* guards. Delegated to a worktree opus agent, parallel with #46 VM + the #47/#48 agent.
Did: Four new checks wired into flake.nix + CI py_compile expansion: hardware-db-modules (installer DB module names ∈ pinned nixos-hardware.nixosModules), installer-compose-lock (offline composer contract on fixtures), installer-disko (pure Nix assert: swapSize "0"/"0G" → no @swap, "2G" → sized, withLuks wrapping — permanently guards the #46 fix), windowrule-syntax (builds the generated hyprland.conf, fails on pre-0.53 grammar / windowrulev2 — permanently guards ed7fd93). CI py_compile now covers all tracked *.py; TESTING.md §1 synced.
Verified: V0 flake check green; V1 I built each of the four checks.x86_64-linux.* on main after cherry-pick — all exit 0. Agent's negative tests confirmed each guard FAILS on a deliberate regression.
Pending: none — all pure/headless, no VM. (A KVM CI runner, item 20, would later run these + the VM suite on push.)
Next suggestion: #50 one-liners (mine); then close #46 (variant A boot).

2026-07-09 — #47+#48 docs/string factual-drift pass (delegated, worktree)

Task: NEXT #47 (docs + string drift) + #48 (i2c option-docs). Delegated to a worktree sonnet agent, parallel with #46 VM + the #49 agent, no shared files.
Did: Agent fixed every drift after verifying each claim against code: theme count 21→24 (README ×3, TESTING, template), waybar whole-swap list (correctly split waybar.css=5 incl. neon-glass vs waybar.jsonc parity=4 + boreal), idle 30→15min-battery-only, Tailscale sudo→operator, live.nix header, MIGRATION link + URL, README layout tree (+doctor/CC/battery-notify), OVERRIDES swappy→satty, always-on note; keybind nmtui→networkmanager_dmenu; nomarchy-menu usage string (+doctor/firmware/controlcenter/rollback). Added README i2c rows (#48). Reviewed the diff, cherry-picked onto main.
Verified: V0 flake check green; checks.option-docs now green (85 options, no stale rows) — closes #48; V1 HM generation builds (usage string + keybind label confirmed in the built output). Re-run on main after cherry-pick.
Pending: none — docs/config, fully verified at their tier.
Next suggestion: land #49 (checks batch); then #50.

2026-07-09 — Promote V0/V1 PROPOSED items → NEXT #47–50 (Bernardo-directed)

Task: Bernardo asked to bump the non-V2 PROPOSED items into NEXT (to feed parallel worktree agents that can't touch the VM). Triage is his call; I did the move on his instruction.
Did: Grouped the genuinely V0/V1 (no session/VM) pitches into four NEXT items — #47 docs+string factual-drift pass, #48 i2c option-docs, #49 pure-contract checks.* batch (py_compile expand, hardware-db ∈ nixos-hardware, installer pure contracts, windowrule dead-syntax guard), #50 small code one-liners (Bluetooth self-gate, CC mktemp, usbutils on PATH, export overlay tools as flake packages). Removed the promoted pitches from PROPOSED, leaving _→ #NN_ breadcrumbs. Deliberately left V2-wanting items in PROPOSED (Waybar doctor icon, CC appearance toggles, theme-color retunes, neon-glass, summer-day scroll, installer review warnings).
Verified: V0 (docs/backlog only) — grep confirms #47–50 present, no leftover promoted-pitch text.
Pending: none. #47–50 are now actionable and agent-friendly (all no-VM).
Next suggestion: spawn worktree agents on #47–50 in parallel with the #46 VM; land their branches after review.

2026-07-09 — #44 Doctor hardware section (delegated to worktree agent)

Task: NEXT #44 — read-only hardware checks in nomarchy-doctor (HARDWARE.md §10). Delegated to a worktree-isolated opus agent so it ran in parallel with the #46 install VM without KVM contention.
Did: Agent added six self-gating checks to nomarchy-doctor.sh (NetworkManager, default PipeWire sink, VA-API/GL smoke, fprintd unit + enroll nudge, fwupd active + get-updates warn, laptop charge-threshold), each skipping cleanly when the tool/service/device is absent — same ok/bad/warn/skip contract + one-fix-command style. Reviewed the diff and cherry-picked onto main (branch agent/doctor-hw-44).
Verified: V0 flake check green; V1 nix build …pkgs.nomarchy-doctor (writeShellApplication runs bash -n + shellcheck at build) — re-run by me on main after cherry-pick. Bonus: the agent ran the built binary on the AMD dev box (read-only) — all six checks fired + self-gated correctly.
Pending: V2 — extend the checks.doctor nixosTest with fakes for the new rows (VM; deferred while #46 held KVM). fwupd get-updates exit-code semantics worth pinning in that fake.
Next suggestion: V2-verify #43+#44 together on the next VM pass.

Task: NEXT #43 — surface fwupd/LVFS updates in the menu (VISION § A). Done while the VM was busy with #46, so V0/V1 only by design.
Did: Added a self-gated Firmware row to the System submenu (command -v fwupdmgr, like Printers) + dispatch + a firmware) handler that opens a terminal and runs fwupdmgr refresh → get-updates → y/N confirm → fwupdmgr update (never silent-flash; fwupd confirms each device + prompts reboot). Updated docs/HARDWARE.md §4 to "shipped".
Verified: V0 nix flake check --no-build green; V1 HM generation builds; extracted the generated nomarchy-menu and confirmed bash -n clean + the firmware handler emitted correctly (exec ghostty -e sh -c, quoting intact).
Pending: V2 render (System submenu shows the row + flow launches) — queued in slimmed #43, do on next VM pass (KVM was held by #46). V3 real LVFS flash → HARDWARE-QUEUE. Optional MOTD/Waybar extensions noted.
Next suggestion: land the #44 doctor-hw agent branch; V2-verify #43+#46 together once KVM frees.

2026-07-09 — #45 V2 verify Boreal default + capture preview.png (theme-shot)

Task: NEXT #45 — close the honesty gap on Boreal-as-default (shipped V0 only at #42) and capture the picker preview. VM-capable environment (KVM).
Did: Rendered the full Boreal desktop headlessly via tools/theme-shot.nix (softGL, 1920×1080). Confirmed the session reads as Boreal — aurora wallpaper paints, Waybar violet accent (#B79BE8) workspace pill, GeistMono clock, rofi launcher violet prompt/selection — not Tokyo Night. Captured a 480×270 themes/boreal/preview.png (matches the other 21's format) from the desktop frame; added a cursor-park to theme-shot so the shot is clean. (En route, found+fixed the windowrule banner — separate commit ed7fd93.)
Verified: V0 flake check green; V2 theme-shot desktop.png + menu.png viewed (banner gone post-fix, bar+launcher themed). Scripted: contrast 24×7 pass; audit boreal advisory-only.
Pending: preview is a headless bare-desktop capture, not the floating-terminal composition of the other 21 — hardware recapture noted under theme polish. V3 (real panel/GPU) not done — not required. The float rules now parse but weren't exercised on real windows (item 41 territory).
Next suggestion: #46 (install swap=0 / LUKS fail-closed) — driver ready.

2026-07-09 — Fix Hyprland windowrule config-error banner (windowrule 0.53 migration)

Task: bug tripped over while doing #45 (VM Boreal verify) — the default desktop showed a red Hyprland config-error banner on every boot, occluding the Waybar.
Did: Root-caused via the theme-shot harness + Hyprland 0.55.4 source (handleWindowrule): Hyprland 0.53 rewrote window rules — windowrulev2 is a hard error and the old rule-first float, class:^…$ order no longer parses (so the float rules weren't even applying). Migrated the 7-rule block in modules/home/hyprland.nix to the new hyprlang grammar <effect> <value>, match:<prop> ^…$ (e.g. float 1, match:class ^…$).
Verified: V0 nix flake check --no-build green; built the generated hyprland.conf and confirmed emitted lines; V2 theme-shot re-render — banner gone, themed Waybar now renders (violet accent, GeistMono clock).
Pending: none for the fix. A cheap checks.* that greps the generated conf for the dead rule-first pattern could lock it (PROPOSED).
Next suggestion: finish #45 (capture boreal preview.png), then #46.

2026-07-09 — Default theme → Boreal (item #42)

Task: NEXT #42 — distro default theme Tokyo Night → Boreal (VISION § B / v1.0 bar).
Did: Seeded theme-state.json + templates/downstream/theme-state.json from themes/boreal.json (incl. border + settings). Eval fallbacks: theme.nix preset → boreal.json; plymouth + live ISO splash hex fallbacks; theme-shot / theme-sync examples; TESTING/RECOVERY wording.
Verified: V0 — state JSON matches boreal palette/fonts; check-theme-contrast 24×7 pass; py_compile theme-sync. Full nix flake check --no-build started but not waited (slow ISO eval). No V2 apply / V3 session.
Pending: Filed NEXT #45 (V2/V3 Boreal verify + preview.png) and #46 (V2 install swap=0 / LUKS fail-closed) for a VM-capable agent. Queue head remains #43 Firmware for product work.
Next suggestion: #43 Firmware, or a VM agent takes #45/#46.

2026-07-09 — Install contracts: swap=0 + unattended LUKS (workflow test run)

Task: VISION § v1.0 install golden path — PROPOSED install P0s (NEXT head blocked: item 14 wait-cron, 20 [human], 33 V3 visual, 41 hw). Test of VISION→BACKLOG→verify→commit loop.
Did: (1) nomarchy-install passes swapSize=0 when swap disabled (was always "${SWAP_GB}G" → "0G" still created @swap). (2) disko-config.nix treats "0" and "0G" as no-swap. (3) Unattended LUKS fail-closed: require NOMARCHY_LUKS_PASSPHRASE or NOMARCHY_NO_LUKS=1.
Verified: V0 — bash -n; pure nix-instantiate of disko subvols for 0/0G/16G (no @swap only on zero). Not V2 install matrix this run.
Pending: optional pure checks.* for swapSize; unattended no-swap/no-LUKS in test-install.sh when convenient.
Next suggestion: promote another VISION § A slice (firmware menu / doctor) or boreal default to NEXT.

2026-07-08 — Theme UI Review: Typography Polish & Icon Call (iteration #87)

Task: NEXT item 28 (Theme UI review).
Did:
1. Resolved the pending human call regarding Rofi vs Waybar icon weights. The user opted for Option A (keep full-color Papirus icons for menus, treating them as an independent colorful layer).
2. Fixed typography scale mismatches: swaync.nix now binds explicitly to the theme's font scale (t.fonts.size) and families (t.fonts.ui/mono).
3. Fixed Rofi's font configuration to use the dynamic t.fonts.size rather than a hardcoded 12.
4. With the color, hierarchy, semantic dims, and typography now audited and unified across surfaces, Item 28 is completed and wiped from the backlog.
Verified: Nix configs evaluate successfully.
Pending: nothing.

2026-07-08 — Hyprland Config Fix: Window rules syntax (iteration #88)

Task: Investigate a red Hyprland error banner visible in VM screenshots during boot.
Did:
1. Analyzed Hyprland log output and screenshot OCR text to pinpoint the config error.
2. Identified that class: regex matching in Hyprland requires windowrulev2, but the previous commits incorrectly used windowrule causing syntax errors (invalid field float: missing a value).
3. Replaced windowrule = [ with windowrulev2 = [ in modules/home/hyprland.nix.
4. Updated README.md to document the IPC-based instant reaction for display profiles.
Verified: Ran hyprctl configerrors headlessly and re-captured the VM snapshot which no longer displays the red banner.
Pending: Wait for the test build to confirm screenshot is clean.

2026-07-08 — Workstation polish pack: Security, Picker, Panic (iteration #86)

Task: Quick execution of 3 brainstormed PROPOSED items for a feature-complete, stable workstation.
Did:
1. Enabled security.apparmor.enable system-wide.
2. Added boot.kernelParams = [ "panic=10" "oops=panic" ] to auto-reboot the system upon catastrophic driver failure.
3. Added pkgs.hyprpicker and bound it to SUPER+SHIFT+C for an instant Wayland color-picker.
Verified: Flake eval / logic.
Pending: nothing.

2026-07-08 — Multi-monitor / laptop docking UX (iteration #85)

Task: NEXT item 29 — fix multi-monitor UX (panel restore on unplug, HDMI audio routing, workspace movement predictability).
Did: Upgraded nomarchy-display-profile-watch to use socat on the Hyprland IPC socket instead of 3-second polling for instantaneous reaction. Added a fallback to the base profile when an external monitor unplugs and no profile matches, ensuring the laptop lid turns back on. Added SUPER+ALT+arrow keybinds to keybinds.nix for explicit, predictable workspace-to-monitor movement. Verified nomarchy-menu audio already serves as the HDMI audio router.
Verified: Local script edits, ready for human hardware validation.
Pending: nothing.
Next suggestion: review remaining NEXT items.

2026-07-08 — Interim build-tier gate for automated lock bumps (iteration #84)

Task: PROPOSED item — automated lock bumps can break systems since they currently only run nix flake check --no-build.
Did: Edited .gitea/workflows/bump.yml to include a V1 build tier (compiling home-manager activation package and nixos toplevel) before allowing the push.
Verified: Local YAML edit, no further check needed.
Pending: nothing.
Next suggestion: review remaining items (e.g., Battery charge limit).

2026-07-08 — External monitor brightness control (iteration #83)

Task: PROPOSED item — external monitors connected to a dock are stuck at blinding brightness unless physical buttons are used.
Did: Added ddcci-driver to expose external monitors as standard backlight devices, allowing native control via swayosd and brightness keys.
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: review remaining items (e.g., Battery charge limit instant toggle or Flatpak camera picker).

2026-07-08 — Drive health monitoring (iteration #82)

Task: PROPOSED item — early SMART warnings for failing hardware.
Did: Added services.smartd.enable = true and services.smartd.notifications.x11/wall to modules/nixos/default.nix. Also explicitly enabled services.systembus-notify.enable to resolve a module collision with earlyoom.
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. ddcutil).

2026-07-08 — Microphone noise cancellation (iteration #81)

Task: PROPOSED item — system-wide background noise filtering for voice calls.
Did: Enabled services.easyeffects.enable = true in modules/home/default.nix.
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. smartd).

Task: PROPOSED item — fix black screens when sharing Wayland desktop in X11 apps via xwaylandvideobridge.
Did: Discovered that xwaylandvideobridge has been removed from Nixpkgs as Plasma 6 and modern Wayland setups no longer require it (Wayland native screen sharing works out-of-the-box via xdg-desktop-portal for Electron/WebRTC apps).
Verified: N/A (skipped).
Pending: nothing.

2026-07-08 — Automount removable media (iteration #79)

Task: PROPOSED item — zero-click USB automounting and safe-removal notifications.
Did: Added services.udiskie.enable = true to modules/home/default.nix.
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. xwaylandvideobridge or easyeffects).

2026-07-08 — File Manager Archive Backend (iteration #78)

Task: PROPOSED item — provide a GUI backend for Thunar's archive extraction plugin.
Did: Added file-roller to environment.systemPackages in modules/nixos/file-manager.nix.
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. udiskie or xwaylandvideobridge).

2026-07-08 — Hardware security key support (iteration #77)

Task: PROPOSED item — enable FIDO2/U2F support for hardware tokens.
Did: Added services.pcscd.enable = true to modules/nixos/default.nix (NixOS >= 26.05 udev handles u2f natively).
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. file manager archive backend or udiskie).

2026-07-08 — BTRFS auto-scrub (iteration #76)

Task: PROPOSED item — enable background filesystem health checks.
Did: Added services.btrfs.autoScrub with a monthly interval to modules/nixos/default.nix.
Verified: V1 (flake check).
Pending: nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. file manager archive backend or pcscd).

2026-07-08 — Magic SysRq keys (iteration #75)

Task: PROPOSED item — enable Magic SysRq keys for safe crash recovery.
Did: Added boot.kernel.sysctl."kernel.sysrq" = 1; to modules/nixos/default.nix.
Verified: V1 (flake check).
Pending: user visual review or nothing.
Next suggestion: continue tackling the PROPOSED backlog queue (e.g. BTRFS auto-scrub or udiskie).

2026-07-08 — Stability + usefulness survey → 9 PROPOSED (iteration #74)

Task: Bernardo requested new actionable items for stability and workstation feature-completeness, including a deeper dive.
Did: Researched the codebase for gaps. Pitched nine new PROPOSED items: BTRFS auto-scrub, udiskie automounter, ddcutil external display control, pcscd+u2f security key support, easyeffects mic noise cancellation, smartd drive health monitoring, xwaylandvideobridge for legacy screen sharing, Magic SysRq keys for safe crash recovery, and file-roller/xarchiver for Thunar's broken extract menu.
Verified: n/a (research only).
Pending: awaiting Bernardo's triage to elevate to the active queue.
Next suggestion: triage the PROPOSED items.

2026-07-08 — Screenshot annotation step (iteration #73)

Task: PROPOSED item — add satty for annotating screenshots.
Did: Added satty to packages with a themed config.toml. Bound $mod SHIFT, Print to start satty directly on a region crop (slurp), and added "Annotate region" to the rofi Capture menu.
Verified: V1 (build tested homeConfigurations.nomarchy.activationPackage).
Pending: V3 queued in HARDWARE-QUEUE.md to verify satty theming and copy/save functionality.
Next suggestion: wait for Bernardo to clear [human] / [blocked:hw] bottlenecks or tackle the stability PROPOSED item.

2026-07-08 — home.stateVersion overridable (iteration #72)

Task: PROPOSED item — home.stateVersion should be mkDefault to support migrations.
Did: Changed home.stateVersion = "26.05" to lib.mkDefault "26.05" in modules/home/default.nix.
Verified: V1 (flake check --no-build green; homeConfigurations.nomarchy.activationPackage built successfully).
Pending: user visual review or nothing.
Next suggestion: triage other PROPOSED items or await Bernardo for [human]-blocked tasks.

2026-07-08 — Stability + usefulness survey → 2 PROPOSED (iteration #71)

Task: Bernardo asked to find the next tasks that make the distro more stable + more useful. Research iteration (LOOP §1: write up PROPOSED).
Did: mapped the module/pkg surface + HARDWARE-QUEUE. Findings: (stability) the state-file path is robust (friendly JSON errors, validate cmd, atomic rebuild on corruption — not a gap); the real hole is the automated lock-bump channel (#14) pushing to main on eval-only with no binary cache, so untested runtime/build breakage compiles straight onto users — the true fix is #20 (VM-suite-in-CI, [human], KVM-runner- blocked). (usefulness) the capture suite is rich but has no screenshot annotate step (no satty/swappy); clipboard/emoji/colorpicker/OCR/ recording all already ship. Filed 2 PROPOSED: an interim build-tier lock gate (no KVM needed, raises the floor until #20) and a satty annotate step.
Verified: V0 (docs-only; flake check was green in #70, tree unchanged).
Pending: Bernardo to triage the 2 PROPOSED into tiers, and — the highest-leverage stability action — decide whether to unblock #20 (register a KVM runner) so the lock-bump channel gets a real runtime gate.
Next suggestion: if #20 stays blocked, the interim build-tier gate is the cheapest stability win; satty is the cheapest usefulness win.

2026-07-08 — QA sweep: clean tree, no actionable drift (iteration #70)

Task: QA sweep — headless-actionable queue is empty (NEXT is all [human]/[blocked:hw]/V3-visual: #29 hw, #28 icon-weight [human]+visual, #14 "confirm first run" not session-confirmable, #20 [human], #33 V3-visual ~10 authored rasi, #41 needs hw class checks).
Did: ran the full scripted suite. flake check --no-build green; built
- ran option-docs (README tables track the live nomarchy.* surface, no drift) and theme-contrast (21 palettes, all pairings pass); audit-theme-design.py shows only pre-curated/identity findings (vanta/ white hueless, inherent CVD collapse, ANSI conventions, miasma earthy bad). Template drift check: services list matches surface exactly; opt-in commented examples (nightlight/monitors/displayProfiles) all reference real options. No churn manufactured (LOOP §1).
Verified: V0 (flake check) + built option-docs & theme-contrast checks.
Pending: nothing. Note: .claude/settings.json deletion was already in the tree at session start — not mine, left untouched (LOOP §0.2); this commit is pathspec-scoped to agent/JOURNAL.md.
Next suggestion: await Bernardo on #29/#33 or the #28 icon-weight call; otherwise #33 rofi scrollbar wants a themed (non-headless) session.

2026-07-08 — Color-only status sweep + inhibited-bell fix (iteration #69)

Task: BACKLOG #28 slice — the audit's systemic "status is never color-only" finding (accessibility): sweep the remaining Waybar status indicators for shape/glyph redundancy.
Did: swept every status module. All already carry non-color redundancy (updates shows a count + self-hides; vpn/doctor/nightlight/ recording self-gate; battery/idle/muted are glyph-distinct) EXCEPT one: swaync inhibited-* reused the normal bells (󰂜/󰂚), so an inhibitor suppressing notifications was distinguished from normal by color alone. Fixed: route every suppressed state (DND + inhibited) to the bell-off glyph 󰂛 + @muted — generated bar (waybar.nix) AND all four whole-swaps (summer-day/night use their own bell-off, executive-slate/boreal 󰂛; parity rule). Reused the proven DND glyph → zero tofu risk.
Verified: V0 (flake check --no-build green; all 4 jsoncs valid JSON). V1: built homeConfigurations.nomarchy.activationPackage; confirmed in the built artifact that inhibited-none/notification = U+F009B (bell-off, was F009C/F009A) and style.css routes .inhibited-* → @muted.
Pending: V3 queued (HARDWARE-QUEUE, Any-machine) — confirm the muted bell-off renders while an app holds a notification inhibitor. Low risk: the glyph already ships for DND. MEMORY gains the "status is never color-only" design invariant (regression guard).
Next suggestion: #28 remaining is [human] (icon-weight call) or V3-visual; likely a QA sweep or await Bernardo on #29/#33.

2026-07-08 — Migration guide: existing NixOS → Nomarchy (docs, direct request)

Task: Bernardo asked whether THIS machine (TuringMachine, NixOS 25.11) can migrate to Nomarchy without a reinstall, keeping all files — and to write the walkthrough into a MIGRATION.md.
Investigated the live machine: already NixOS, personal modular flake, systemd-boot, LUKS2 + btrfs with the exact subvol layout Nomarchy wants (incl. @snapshots/@home-snapshots), AMD 7840U. Verdict: migration is a config switch (reuse hardware-configuration.nix), NOT a reinstall; @home is never touched. Main deltas: 25.11→26.05, keep system.stateVersion=24.11, power-mgmt collision (his ryzenadj/TLP vs Nomarchy PPD), HM-module→standalone.
Decisions (Bernardo): adopt Nomarchy PPD + drop all ryzenadj; no Secure Boot (drop lanzaboote); full cutover.
Did: wrote docs/MIGRATION.md — generic "existing NixOS → Nomarchy" guide with TuringMachine as the worked example: candidacy checklist, the two data-safety rules (never bump system.stateVersion; /home untouched), Phase 0 snapshots → Phase 1 build-only gate → Phase 2 reversible nixos-rebuild test → Phase 3 reconcile → Phase 4 cutover, triple rollback net. Verified the downstream API against lib.nix (mkFlake src/username/ hardwareProfile) + templates/downstream so the flake/system/home examples are real, not guessed.
Verified: V0 (docs) — referential accuracy against lib.nix + template; did NOT run the migration commands (they'd alter the live machine).
Follow-ups: PROPOSED — make home.stateVersion mkDefault (migrating users currently need mkForce).
Next suggestion: if Bernardo greenlights, scaffold Phase 1 from his real hardware-configuration.nix.

2026-07-08 — Calendar on the clock click (iteration #68, item 42)

Task: item 42, de-scoped live by Bernardo — DON'T hook up calendar accounts; just ship a lightweight calendar that opens instantly on the Waybar date click, user configures sync themselves. calcurse (his pick; now the right one since it's self-contained for the user's own sync).
Did: new nomarchy-calendar launcher (waybar.nix) → calcurse in a ghostty window tagged --class=com.nomarchy.calendar + --gtk-single-instance=false (fresh standalone window); self-gates on calcurse. Clock on-click = nomarchy-calendar, dropped the format-alt date-toggle (conflicts with on-click), moved the long date into the tooltip. Parity: on-click in all four whole-swap jsoncs. windowrule float+size(60/65%)+center on the class (hyprland.nix). Shipped calcurse
- pwvucontrol uncommented in the downstream template (opt-out). Enforced ghostty always installed (ghostty.nix: enable = true unconditional, the nomarchy.ghostty.enable option now gates only config) so the launcher can rely on --class.
Verified: V0 green; all four jsoncs valid JSON; V1 HM built — launcher ships the right command, generated clock renders on-click: nomarchy-calendar with format-alt gone + date in the tooltip, hyprland has the 3 calendar windowrules, ghostty present in the HM profile.
Pending: V3 — click the clock → calcurse floats centered (the --class→windowrule match) and renders its month view; verify on a real session (queued). Sync is deliberately the user's own local config, NOT a distro feature (Bernardo's call) — a future iteration should not re-propose account sync.
Next suggestion: Bernardo's big new task — investigate whether THIS machine (NixOS 25.11) can migrate to Nomarchy without a reinstall, keeping all files, and plan it.

Task: BACKLOG item 36(b), now unblocked. Bernardo's calls: keep the granular System ▸ Battery limit / Power profile rows; build ONE combined menu; open it two ways — the Waybar battery icon AND the power-profile icon (the profile icon's click flips from cycle → menu).
Did: New nomarchy-menu powermgmt (rofi.nix): a flat list of Profile rows (→ powerprofilesctl set, like the power-profile picker) + Charge limit rows (→ the same theme-sync writer as batterylimit), each half self-gating on its hardware; no backend duplication, no Custom flow (that stays in the dedicated batterylimit picker). Wired battery.on-click + custom/powerprofile.on-click → powermgmt in waybar.nix AND all four whole-swaps (parity). Named powermgmt to avoid the whole-swaps' existing custom/powermenu (that's the shutdown button → nomarchy-menu power, left untouched). nomarchy-powerprofile-cycle kept as a standalone bin (now just unbound by default).
Verified: V0 green; V1 HM built (bash -n passed) — shipped nomarchy-menu has the powermgmt case; generated waybar-config.json wires both icons to it (2 hits); all four whole-swap jsoncs re-wired and still parse as JSON, shutdown buttons intact.
Pending: V3 — click either icon → menu opens, picking a profile / charge preset actually applies (queued in HARDWARE-QUEUE). The instant-charge-write follow-up stays PROPOSED/[blocked:hw].
Next suggestion: back to the queue — item 33 (rofi scrollbar) or a QA sweep; both remaining NEXT items lean human/hardware.

2026-07-08 — Capture rows show Print-key hints (iteration #66, direct requests)

Task: Two live requests from Bernardo. (1) Fold his provided power-menu mockup into item 36(b). (2) "make sure all menu items that have a keybind display the keybind" (he dropped the flush-right ask — inline is fine).
Did: (1) BACKLOG item 36(b) un-gated — mockup + Nomarchy adaptation (reuse batterylimit/power-profile writers, not raw TLP; wire the Waybar battery on-click; one open call: replace-or-keep the System rows). (2) Audited every menu/submenu: all SUPER(+CTRL) module-opening rows already carry a hint (item 28c). The one real gap = the Tools ▸ Capture rows, which run the exact grim commands of the Print / SHIFT+Print / CTRL+Print binds but showed nothing. Added hintForBind mods key (reads keybinds.nix by mods+key, drift-free) → hints on the three matching rows, and -markup-rows on the Capture menu so the spans render.
Verified: V0 nix flake check --no-build green; V1 HM built, shipped nomarchy-menu shows Region → clipboard {{Print}}, Region → file {{SHIFT + Print}}, Full screen → file {{CTRL + Print}} under -markup-rows. Substring case matches sit before the appended span, so dispatch is unaffected (same proven pattern as Tools/System).
Pending: V3 — pango dim actually rendering in the Capture submenu (queued on the existing Capture entry). Power-menu item 36(b) stays [human] on the replace-or-keep call.
Next suggestion: back to the queue head (item 36a done; 32 done) — next actionable is likely item 33 (rofi scrollbar) or a QA sweep.

2026-07-08 — SUPER+? cheatsheet re-fixed: base keysym (iteration #65, item 32)

Task: BACKLOG item 32 — SUPER+? still didn't open the cheatsheet after item 26's fix. Re-diagnose.
Did: Root cause = wrong keysym, not path/rofi. Hyprland 0.55 resolves a bind's keysym with SHIFT consumed (issue #7750, since 0.43), so $mod SHIFT, question looks for slash and never matches. Item 26 fixed the modmask but kept the shifted keysym → still dead. Changed keybinds.nix question → slash (the BASE keysym, exactly like the working $mod SHIFT, 1 workspace binds in this same file); updated rofi.nix prettyKeys collapse SHIFT + / → ? so the cheatsheet still reads "SUPER + ?". Dispatch (nomarchy-menu keybinds) confirmed already fine.
Verified: V0 nix flake check --no-build green. V1: HM activation built; grepped the built hyprland.conf → bind=$mod SHIFT, slash, exec, nomarchy-menu keybinds (beside the proven $mod SHIFT, 1), kb_layout=us confirms slash→? ; nix eval of prettyKeys → "SUPER + ?".
Pending: V3 — the keypress actually opening the menu is invisible below hardware (that's how item 26 slipped); the existing HARDWARE-QUEUE "SUPER+? opens the cheatsheet" entry updated to the slash fix.
Next suggestion: Bernardo's two live requests — (1) fold the provided power-menu mockup into item 36(b) [design now in hand]; (2) right-align a keybind hint on every menu/submenu row that has one.

2026-07-07 — Battery-limit toggle in rofi System (iteration #64, item 36a)

Task: BACKLOG item 36(a) — Bernardo wanted the charge-threshold toggle as a first-class row in nomarchy-menu › System, not buried in the gum control-center where #55 put it.
Fix: rofi.nix — new batterylimit subcommand: a preset picker (80%/90%/60%/Off/Custom…) with battery-0XX icons, writing the baked settings.power.batteryChargeLimit via theme-sync --no-switch (lands on next sys-rebuild; power.nix's oneshot applies it). Added a System row self-gated on the charge_control_end_threshold sysfs node (like Power profile), dispatch, and usage string. Moved — removed the duplicate "Battery Limit" case + choose entry from nomarchy-control-center.sh so it lives in one place.
Verified: V0 green. V1: control-center builds clean (shellcheck), home activation builds clean (writeShellScriptBin's bash -n on the menu passed); grepped the built script to confirm the row/dispatch/case/ guard are all present. Live menu interaction is a V3 check (queued).

2026-07-07 — Floating-window audit, conservative cut (iteration #63, item 41)

Task: BACKLOG item 41 — Bernardo: "take a conservative approach." Broaden the windowrule float set beyond the mixer.
Method: surveyed what GUI dialogs the distro actually ships (not a blind list). Network is dmenu/nmtui (no GUI editor → no rule). Confident, shipped, unambiguous dialogs: blueman (manager + adapters) and system-config-printer. Deferred polkit (hyprpolkitagent exposes no discoverable class) and GTK file portals — guessing a class = dead rule, so those wait for a hardware hyprctl clients read.
Fix: hyprland.nix windowrule block — added float + center for the two, (?i) regex tolerating the XWayland .…-wrapped form. RE2 (?i) is already relied on in this file (line ~47), so it's consistent.
Verified: V0 green; nix eval confirms the rendered regex escaping is correct (single backslash, ? quantifier unescaped). Whether the classes actually match live is V3 (queued — includes confirming the regex catches the real class).

2026-07-07 — Right-click volume → floating mixer (iteration #62, item 35)

Task: BACKLOG item 35 — right-click the Waybar volume module to open a mixer in a floating window. (Bernardo re-launched /loop after I'd flagged it as gated on 41 — read as "handle it," so I did 35 fully and re-scoped 41 to just broadening the float set.)
Chose pwvucontrol (PipeWire-native, GTK4/libadwaita → themes via the portal like Amberol; distro is already wpctl/PipeWire). Verified app-id com.saivert.pwvucontrol from the built package.
Changes: (1) waybar.nix pulseaudio on-click-right = "pwvucontrol"
- parity into all four whole-swap jsoncs (summer-day/night, boreal, executive-slate). (2) hyprland.nix — the config's first windowrule block: float + center for the mixer classes (pwvucontrol + pavucontrol fallback), normal-priority list so downstream concatenates. (3) pwvucontrol added to the template home.packages.
Verified: V0 green; nix eval of the homeConfiguration confirms the windowrule regex renders with correct single-backslash escaping (class:^(com\.saivert\.pwvucontrol|…)$). The right-click→float→center interaction is a V3 session check (queued). Item 41 now = broaden the set.

2026-07-07 — Window focus: vim keys → arrows (iteration #61, direct request)

Task: Bernardo — switch window navigation from vim-style SUPER+H/J/K/L to SUPER+arrow keys.
Fix: modules/home/keybinds.nix focus binds now use left/right/up/down (movefocus l/r/u/d); H/J/K/L freed. No conflict — SUPER+arrows were unbound and there were no vim move-window binds. Also taught prettyKeys (rofi.nix) to render the arrow keysyms as glyphs (← → ↑ ↓), so the SUPER+? cheatsheet reads "SUPER + ←" not "SUPER + left" (Return etc. unaffected — no false substring match).
Verified: V0 green; nix eval confirms the binds render as $mod, left, movefocus, l … and the glyph mapping is correct.

2026-07-07 — Audio mime → Amberol (iteration #60, item 37)

Task: BACKLOG item 37 — default video + audio players + mime. On inspection both players were already installed: mpv (video) and amberol (a prior template add). So the real gap was the mime wiring — audio files still opened in mpv (mime.nix mapped audio/* → mpv.desktop).
Fix: modules/home/mime.nix — repoint the audio/* types to io.bassi.Amberol.desktop (verified against the built package's actual .desktop id + its MimeType set) and broaden coverage (mp3/flac/ ogg/opus/wav/m4a/aac, both canonical and x- names). Video stays mpv.
Why Amberol: GTK4/libadwaita, so it themes via the portal color-scheme the distro already wires; minimal, purpose-built for "open a file → play." Degrades to mpv if a downstream drops it.
Verified: V0 green; entries mirror the existing working image/video mappings, desktop id confirmed from the realised package. Runtime GIO resolution (double-click an mp3 → Amberol) is a V3 session check (queued).

2026-07-07 — Capture-to-file keybinds (iteration #59, item 38)

Task: BACKLOG item 38 — direct keybinds for region→file and screen→file (the Capture menu already had both "→ file" rows).
Fix: modules/home/keybinds.nix — added SHIFT+Print (region→file) and CTRL+Print (screen→file) alongside the existing bare Print (region→clipboard). Each saves a timestamped PNG under ~/Pictures/Screenshots and toasts the path — same grim/slurp plumbing + dir the menu's "→ file" rows use. Single-source, so the SUPER+? cheatsheet (rofi.nix reads the same list) gets both rows automatically.
Verified: V0 green. nix eval of the rendered binds confirms $HOME/$f/$(...) pass through literally to the shell (no Nix interpolation). Actual key-firing + file landing is a V3 check (queued).

2026-07-07 — rofi search mis-ranks + power-profile icons (iteration #58, items 31 & 39)

#31 (rofi search broken): Bernardo ran the live A/B — -sorting-method normal and -matching normal both fixed "steam"→Steam; -no-sort did not. So the culprit is rofi 2.0.0's fzf sorter paired with fuzzy matching (it ranked ghostty above Steam), not case-sensitivity. Fix: programs.rofi.extraConfig.sorting-method "fzf" → "normal", keeping matching="fuzzy" (forgiveness) + case-sensitive=false. Also closes Bernardo's batch item 1 (case-insensitive — already false, now ranks right too).
#39 (power-profile icons): the picker listed profiles as bare text. Now emits row "$p" battery-profile-<p> using Papirus' colored battery-profile-{performance,balanced,powersave} family (the -symbolic variants are #444 grey → invisible on dark themes; verified the colored ones carry real fills). Returned row text stays the bare profile name, so powerprofilesctl set is unaffected; added -show-icons.
Verified: V0 nix flake check --no-build green for both. #31's fix is the exact lever Bernardo confirmed live. #39's icon rendering is a V3 visual check (Papirus lookup can't be eyeballed headlessly) — queued.

Task: BACKLOG item 40 — in System › Display the Back row didn't return on a single-monitor machine.
Root cause: the mode picker's Back did exec "$0" display. On a single-monitor laptop with no display profiles, the output chooser is skipped, so re-running display drops straight back into the same mode picker — an apparent no-op. (Multi-monitor works: display re-shows the chooser, a genuine previous level.)
Fix: modules/home/rofi.nix display handler now tracks modeBack — display when the chooser was shown, else system. Mode-picker Back does exec "$0" "$modeBack", so single-monitor Back leaves Display for System.
Verified: V0 nix flake check --no-build green. Logic is deterministic; the single- vs multi-monitor branch is the whole fix.

Task: BACKLOG item 34 — every submenu's ↩ Back row drew two arrows: the ↩ glyph in the label AND a themed go-previous icon.
Fix: back() in modules/home/rofi.nix was row "$BACK" go-previous (label + icon). Changed to printf '%s\n' "$BACK" — now identical to what the plain pick-lists already append, so every Back row shows a single ↩ glyph. Kept BACK="↩ Back" as the universal label/match, so no case-branch comparison changed (no risk of a Back that fails to return — the item-40 failure mode). Collision-exact match preserved by the glyph.
Verified: V0 nix flake check --no-build green (evaluates homeConfigurations). Single-arrow result is deterministic from the code; visual confirmation on hardware is nice-to-have, not load-bearing.

2026-07-07 — Battery charge-limit quick toggle (iteration #55)

Task: BACKLOG PROPOSED (promoted by Bernardo) — a toggle for the battery charge threshold.
Did: Replaced the control-center "Battery Limit" raw number-prompt with a preset toggle picker (Off / 80% recommended / 90% / 60% / Custom…) writing settings.power.batteryChargeLimit. Kept it in the control-center — that's where the baked, rebuild-valued System Toggles live (rofi nomarchy-menu is the instant surface; the CC holds config values); it's reachable from the menu via System › Control Center. Base machinery (sysfs oneshot + AC-replug udev in power.nix) untouched.
Verified: V0 (bash -n; nix flake check --no-build, green) + V1 (nomarchy-control-center package builds → the writeShellApplication shellcheck gate passes).
Pending: instant-effect (no rebuild) refiled in PROPOSED as [blocked:hw] — needs a udev-writable sysfs node, confirmable only on a laptop with the charge-threshold control. Light session spot-check queued in HARDWARE-QUEUE.
Next suggestion: item 29 is [human]/[blocked:hw]; remaining is item 28 slices or a QA sweep.

2026-07-07 — Waybar overlaps fullscreen video (iteration #54)

Task: BACKLOG item 30 — bar drawn on top of a fullscreen YouTube video.
Did: layer: top → bottom in the generated waybar.nix and the four whole-swap jsoncs (boreal, summer-night, summer-day, executive-slate — parity rule). On Hyprland the top layer renders above even real-fullscreen surfaces; bottom lets the fullscreen window cover the bar while the exclusive zone still reserves its space in normal tiling. Rationale comment in waybar.nix; data jsoncs left comment-free (they carry none). Also shipped this session (pre-loop, user-driven): battery discharging-icon gap widened (f848e73).
Verified: V0 (nix flake check --no-build, green) + V1 (homeConfigurations.nomarchy.activationPackage builds; the built .config/waybar/config shows "layer": "bottom").
Pending: V3 queued in HARDWARE-QUEUE — fullscreen video covers the bar, bar returns on exit, floating-overlap trade-off. Can't be eyeballed headlessly here.
Next suggestion: item 29 is [human]/[blocked:hw]; remaining work is item 28 (theme UI) slices or a QA sweep.

Task: PROPOSED item — ensure nomarchy.* options have a menu toggle where appropriate.
Did: Audited options and implemented the first slice of missing toggles in the TUI Control Center (Updates, Battery Limit, Bluetooth, Printing). Wired modules/home/options.nix and modules/nixos/options.nix to use config.nomarchy.settings.* as defaults.
Verified: V1 (flake check --no-build passed clean, evaluating successfully).
Pending: user visual review and testing.
Next suggestion: triage other PROPOSED items.

Task: Add pickers for string-based options.
Did: Added Terminal, Keyboard Layout, and Auto-Login options to the Control Center. Wired them to the Nix options via config.nomarchy.settings.*.
Verified: V1 (flake check --no-build).

2026-07-06 — theme UI review (iteration #52)

Task: Make every surface stunning with a glassmorphic style (Backlog #28).
Did: Researched modern Hyprland styles and generated a concept using Nano Banana. Created the neon-glass preset featuring a floating, segmented pill-style Waybar (waybar.css) and a glassmorphic Rofi launcher (rofi.rasi). Generated a matching 4K wallpaper.
Verified: Contrast checker script passed (22 themes x 7 pairings green). Flake check passed cleanly.

2026-07-06 — theme UI review, part 2 (iteration #53)

Task: Create a professional, sober dark theme maximizing functionality.
Did: Generated a minimalist geometric concept wallpaper. Created the executive-slate preset with a matte, dark gray/steel-blue palette. Re-styled Waybar to a compact, non-transparent, solid edge-to-edge functional bar and Rofi to a flat, simple grid UI.
Verified: Contrast checker script passed (23 themes x 7 pairings green). Flake check passed cleanly.

2026-07-06 — waybar redesign (iteration #50, one-off)

Task: User request — remove system usage (CPU) and memory items from waybar.
Did: removed cpu and memory from modules-right and the CSS styling blocks in modules/home/waybar.nix.
Verified: V1 (flake check --no-build passed clean).
Pending: user visual review.
Next suggestion: triage PROPOSED.

2026-07-06 — control center implementation (iteration #49, item 18)

Task: BACKLOG NEXT#18 — implement TUI Control Center using gum.
Did: created nomarchy-control-center.sh and packaged it in pkgs/nomarchy-control-center. Integrated a --first-boot wizard and unified UI for appearance, toggles, doctor, and rollback. Added the package to flake.nix and modules/nixos/default.nix. Added "Control Center" to the System menu in rofi.nix.
Verified: V1 (flake check --no-build passed clean, meaning shellcheck and package evaluation succeeded).
Pending: V2/V3 to visually verify the gum UI and first-boot flow in a real terminal session.
Next suggestion: watch lock-bump CI (item 14) or triage PROPOSED.

2026-07-06 — control center design pass (iteration #48, item 18)

Task: BACKLOG NEXT#18 — control center form-factor design pass [big].
Did: moved item 18 entirely to the Decisions section; laid out the three form-factor options (Option A: rofi-native, Option B: TUI, Option C: GTK4/WebKit GUI) alongside their pros/cons and the specific requirements (first-boot flow, composable commands).
Verified: V0 (docs-only); nix flake check --no-build passed pre-commit.
Pending: the [human] decision on the form-factor.
Next suggestion: #14 (watch lock-bump CI) or #20 KVM runner [human]. If none actionable, triage PROPOSED or do a QA sweep.

2026-07-05 — display-profile workspace pins (iteration #47, item 15c)

Task: item 15 slice (c), the last one — per-profile workspace → output pinning. Item 15 deleted (a/b/c all shipped; V3s queued).
Did: a profile is now { monitors; workspaces = { "1" = "DP-3"; }; } (bare list still accepted — either-type + normalization in hyprland.nix; coercedTo refuses list-of-submodule sources). Pins bake as Hyprland workspace rules and apply instantly on profile switch (hyprctl keyword workspace + dispatch moveworkspacetomonitor). monitor-rules.nix gained pure workspaceRule; template + README updated.
Verified: V1 flake check green (incl. the extended checks.display-profiles assert) + scratch-downstream eval: baked settings render both rule sets, the generated applier carries the keyword+dispatch pairs (bash -n OK), match/list still correct with mixed profile shapes.
Pending: V3 queued (workspace-pins entry in HARDWARE-QUEUE, rides the slice-a dock test). Stale-pin note documented in the applier.
Next suggestion: backlog is now human/event-gated (28 taste calls, 14 Monday, 18 Decision, 20 KVM) — idle the loop or triage PROPOSED.

2026-07-05 — icon-weight A/B for the human call (iteration #46, item 28)

Task: item 28's last [human] blocker — make the color-vs- symbolic menu-icon decision cheap.
Did: captured the lumon root menu with -symbolic icon names (temporary uncommitted rofi.nix edit, reverted after) and published a side-by-side artifact (link in the BACKLOG bullet). Findings: symbolic matches the bar's weight, but Papirus symbolic coverage has holes and rofi tints symbolics fixed gray — B needs an audit+recolor slice; A (color) is the free default.
Verified: V2 for the experiment itself (both variants VM- captured); no repo code changed — tree is clean apart from agent/.
Pending: Bernardo's call on the artifact. Item 28 is now fully agent-side done: all remaining bullets are [human]/hardware.
Next suggestion: general loop — item 15 slice (c) (per-profile workspaces) is the topmost actionable code item; 14 fires Monday.

2026-07-05 — purge stub waybar.css whole-swaps (iteration #45, item 28c/d)

Task: item 28 — lumon vibe capture exposed four themes shipping an unstyled default bar.
Did: deleted the stub waybar.css in catppuccin/lumon/nord/ retro-82 (2–14 lines of color defines from the f211ef0 rewrite import; a theme waybar.css replaces the WHOLE generated stylesheet, so the stubs nuked all bar styling). The four themes now get the generated bar in their own palette; their full rofi.rasi designs kept. Whole-swap bars are now only summer-day/night, both authored.
Verified: V2 — lumon re-capture shows the generated styling (workspace pill, padded right cluster, no edge clipping); V1 flake check green. catppuccin/nord/retro-82 share the mechanism, not individually captured.
Pending: HARDWARE-QUEUE eyeballs; the [human] icon-weight call.
Next suggestion: item 28 is agent-side done pending Bernardo's taste calls — park it; take the next actionable BACKLOG item.

2026-07-05 — capture harness paints wallpapers (iteration #44, item 28c)

Task: item 28c harness gap — captures rendered a bare base-color backdrop because the guest had no ~/.nomarchy runtime state.
Did: tools/theme-shot.nix now seeds theme-state.json via tmpfiles at boot (before greetd), so the session's exec-once nomarchy-theme-sync wallpaper paints naturally; dropped the testScript seed+manual-repaint (too late for exec-once; manual runs also lack WAYLAND_DISPLAY, which awww's client needs to find its socket).
Verified: V2 — summer-night capture shows wallpaper + cream inverted bar + menu + placeholder dim, all coherent. V1 flake check green.
Pending: identity-theme vibe capture (lumon/retro-82); the [human] icon-weight call; HARDWARE-QUEUE eyeballs.
Next suggestion: item 28c identity capture, or park 28 for Bernardo's taste calls.

Task: LATER › auto-timezone Waybar tooltip. Bullet deleted.
Did: the clock tooltip's first line is now the zone — {:%Z (UTC%z)} above the calendar — in the generated bar AND both summer whole-swaps (parity). Under auto-timezone the existing tz-watch SIGUSR2 reload keeps it current, so it shows the detected zone live. (%Z abbreviation, not the IANA name — strftime has no code for that; good enough for an "(optional)" nicety.)
Verified: V0; V1 — rendered generated config carries the new tooltip-format; both summer jsonc parse (jq) with %Z present. Hover render → V3 queued.
Pending: V3. LATER now holds only deferred/big/hw/human items.
Next suggestion: a QA sweep, or stop-and-wait — the queue is effectively drained for unattended work; meaty LATER items (wallpapers split is deferred by decision, installer round 2 [big]) deserve Bernardo's go-ahead or a fresh session.

2026-07-05 — doctor bar tripwire (iteration #42, LATER item)

Task: LATER › doctor Waybar warning. Bullet deleted.
Did: nomarchy-doctor-status (named bin, PATH): silent exit 0 while the doctor is healthy or absent (self-gate), else JSON with the first ✖ lines jq-escaped into the tooltip, in @bad; click = nomarchy-menu doctor (the sheet). custom/doctor added to the generated bar (after battery, interval 300 — the check is real work, this is a tripwire not a monitor) + CSS, AND both summer whole-swaps (module list, jsonc block, CSS membership + @red rule) per the parity rule.
Verified: V0; V1 — template-home builds, status bash -n, config
- CSS each carry the module, both summer jsonc still parse (jq) with the block; FUNCTIONAL — self-gate silent, and a shimmed failing doctor produces valid JSON (jq-verified: class bad, ✖ lines in tooltip). Live render → V3 queued.
Pending: V3. LATER remaining: wallpapers-artifact split (deferred by decision), installer round 2 [big], boot-from- snapshot, night-light geo, icon overrides, MIPI camera [hw], auto-timezone tooltip, VPN richer display, release bump [human].
Next suggestion: auto-timezone Waybar tooltip (small) — or a QA sweep; the meatier LATER items deserve a fresh session.

2026-07-05 — OCR screenshot-to-text (iteration #41, LATER item)

Task: LATER › OCR (NEXT is fully human/event-gated, so LATER is in play per LOOP.md). Bullet deleted.
Did: Capture › "OCR region → clipboard" (self-gated on tesseract): slurp (Esc = silent cancel, the recording pattern) → grim → tesseract stdin/stdout → wl-copy, whitespace-only results toast instead of clobbering the clipboard; success toasts the word count. tesseract ships eng-only via rofi.nix home.packages (override documented — unscoped it drags every language's traineddata, real cost on a compile-from-source distro).
Verified: V0; V1 — template-home builds, menu bash -n, gated row + case present; FUNCTIONAL — the profile's own tesseract read a rendered test image back verbatim ("Nomarchy OCR smoke test 42"), so the OCR half of the pipe is proven; grim/slurp half needs a session → V3 queued.
Pending: V3.
Next suggestion: LATER: doctor Waybar warning (self-gating bar indicator fed by nomarchy-doctor) or auto-timezone tooltip; both small.

Task: BACKLOG NEXT#19 — its ≥3-entries gate is met (Theme, Next wallpaper, Night light). Item deleted.
Did: root's Theme row → "Look & Feel" (pango-escaped — -markup-rows menus treat & as an entity; matched on Look); new lookfeel mode: Theme (SUPER+T hint), Next wallpaper (SUPER+SHIFT+T hint via the new generalized hintForAction), Night light row moved from System (System thins by one). Root stays six. menuHint is now sugar over hintForAction.
Verified: V0; V1 — menu bash -n, lookfeel mode + escaped label
- night light relocated (4 occurrences, all in lookfeel/nightlight paths); VISUAL — tokyo-night capture: root renders "Look & Feel" correctly with six rows and intact hints. Submenu flow → V3 queued.
Pending: V3. NEXT now: 28 (Bernardo-gated) · 14 (Monday watch) · 18 [big, Decision-gated] · 20 [human] — the LATER tier or QA sweeps are next for the loop.
Next suggestion: LATER's OCR screenshot-to-text (two-row Capture addition, self-contained) — or a QA sweep.

2026-07-05 — launch-or-focus binds (iteration #39, item 17)

Task: BACKLOG NEXT#17 (the UI review is Bernardo-gated; back to the queue head). Item deleted.
Did: nomarchy.launchOrFocus (listOf submodule: key/mods/class/ command/desc; opt-in, template example) → hyprland.nix generates binds through a shared nomarchy-focus-or-launch script (jq case-insensitive class match → focuswindow "(?i)" regex, else launch; uninstalled command → toast, so binds survive suite edits); rofi.nix renders the same entries into the SUPER+? cheatsheet via prettyKeys (single source). README row.
Verified: V0; V1 — scratch downstream with 2 entries: both binds rendered (default command = lowercased class; custom mods/command/desc honored), tool bash -n + usage exit 64, cheatsheet rows present. Focus behaviour needs a session → V3 queued.
Pending: V3. Queue after this: #18 control center [big, needs the form-factor Decision], #19 Look&Feel menu (gated on ≥3 entries), #20 [human], LATER tier.
Next suggestion: a QA sweep (#18/#19/#20 are blocked-ish) — or item 19's precondition check (does Look & Feel earn 3 entries yet?).

2026-07-05 — placeholder dim + latte review + backdrop probe (iteration #38, item 28c)

Task: item 28 slice (c) — the summer-rasi quickie + the light- theme capture.
Did: both summer rasi: placeholder-color @bg0 → bg0 at 60% alpha (#…99), so prompt and placeholder no longer read as one phrase. Captured catppuccin-latte at 1080p: the generated menu is COHERENT on light themes (zebra rows on the 28b-retuned surface, hints dimmed) — closes the audit's "dark terminal on latte" concern as a stale preview. Backdrop suspicion probed with a dedicated VM (hyprctl getoption): misc:background_color IS set (0xFFEFF1F5) — awww's empty-state layer occludes it in the wallpaper-less VM; on hardware the fix covers the startup gap as intended.
Verified: rasi edits grep-verified (rasi parses at rofi runtime; visual rides the next summer capture / V3); latte menu capture reviewed; probe output above.
Pending: item 28 remaining: icon-weight [human], hardware eyeballs (queued), identity-theme capture (optional), slice (d) needs nothing (parity verified).
Next suggestion: item 28 is now Bernardo-gated (taste calls + V3); the loop should return to the general BACKLOG head (item 14 watch / item 17 launch-or-focus).

2026-07-05 — rofi keybind hints (iteration #37, item 28c)

Task: item 28 slice (c) — the hint column, the audit's biggest remaining UX item.
Did: rofi.nix: menuHint builds a dimmed pango span (size small, alpha 55%) from keybinds.nix via the cheatsheet's prettyKeys — single source, hints can't drift from the live binds; applied to root (Apps/Theme/Power/Keybindings; Apps special-cased, it's rofi drun not a menu module), Tools (all eight leaves), System (Network/Bluetooth/DND); those three menus now pass -markup-rows. Contains-glob matching and the exact "↩ Back" match are unaffected (labels keep their prefix; Back stays markup-free).
Verified: V0; V1 rendered menu bash -n + hint spans present + 3× -markup-rows; VISUAL — tokyo-night capture: hints render dimmed and subordinate on normal AND selected (accent) rows; Tools/System rows correctly bare.
Pending: slice-c remainder: summer-rasi placeholder dim, latte/identity captures, icon-weight [human].
Next suggestion: the placeholder-dim quickie + a latte capture in one tick; then item 28 pauses for Bernardo's icon-weight call and the hardware eyeballs.

2026-07-05 — bar group rhythm + @muted adoption (iteration #36, item 28c)

Task: item 28 slice (c) — the two queued bar polish items.
Did: waybar.nix generated CSS: margin-left 14px on the three right-cluster group heads (pulseaudio/powerprofile/battery — media · toggles · status; hidden heads degrade to uniform spacing); workspaces buttons, dnd bell, and muted volume moved from alpha(@text,.5) tints to the semantic @muted role — safe since 28b floor-gates muted/base ≥ 2.0 (the old comment warning against @muted documented the pre-28b world; rewritten to state the new invariant). Whole-swap bars untouched (authored identity).
Verified: V0; V1 rendered CSS has 4 @muted uses + the margin rule; VISUAL — tokyo-night re-capture: themed backdrop confirmed (navy base, not black), group gap visible before the volume group, workspace pill legible. Cross-theme minima are guaranteed by the gated checker rather than per-theme eyeballs.
Pending: slice-c remainder: rofi hint column, summer-rasi placeholder dim, icon-weight [human], latte/identity captures.
Next suggestion: rofi keybind-hint column (the biggest remaining UX win), or the two-line placeholder nit as a quickie.

2026-07-05 — summer-day capture review + themed backdrop (iteration #35, item 28c)

Task: item 28 slice (c) — whole-swap capture review + the first cheap polish fix.
Did: captured summer-day at 1080p (harness, VM): the authored slate menu + island-pill bar hold up with the 28b retunes — initially misread as polarity soup, but the rasi whole-swap deliberately mirrors the bar's slate strip: identity, keep (nit filed: undimmed entry placeholder). The surviving cross-theme fix: Hyprland misc:background_color = theme base (was compositor black — a glitch-looking flash behind missing/slow wallpaper, glaring on light themes). Punch list updated.
Verified: V0; V1 — template-home rebuilt, rendered hyprland.conf carries background_color=rgb(1a1b26) (theme base). Backdrop visible confirmation rides the next capture run / V3.
Pending: captures for latte + one identity theme; polish items queued: bar group margins, @muted adoption, rofi hint column, icon weight [human].
Next suggestion: bar polish slice (group margins + @muted) — code-level, verifiable by build + re-capture.

2026-07-05 — 1080p capture harness + first slice-c review (iteration #34, item 28c)

Task: item 28 slice (c) prep — full-res captures for the per-surface polish pass.
Did: tools/theme-shot.nix (maintainer tool, THEME env-driven): full themed desktop headless at 1920x1080 via the memory recipe + two fixes over the spike — home-manager.useGlobalPkgs (HM submodule otherwise misses the overlay → 'nomarchy-theme-sync missing') and a menu.png shot (hyprctl dispatch exec nomarchy-menu). Proven on tokyo-night; both shots read. Findings → item 28 punch list: rofi root width vs empty space (→ keybind-hint column idea), icon-weight mix (color Papirus vs mono bar) [human], bar group margins confirmed, alpha-tint dim vs @muted role. Whole-swap module parity verified textually (no drift; cpu/memory omission is identity).
Verified: harness runs end-to-end (exit 0, screenshots read); eval from tools/ with relative flake path OK. Wallpaper still doesn't paint (theme-sync needs ~/.nomarchy in the guest — fix noted in the header for the next run).
Pending: capture summer-day/night (whole-swaps) + latte; then the polish slices; icon-weight call is Bernardo's.
Next suggestion: run the retuned-theme captures + start the cheap polish items (group margins, @muted adoption).

2026-07-05 — P2 floors retuned + gated (iteration #33, item 28b done)

Task: item 28 slice (b) remainder — the P2 contrast floors.
Did: 18 hex retunes across 9 themes: muted raised on lumon/ white/retro-82/everforest (OKLCH hue-preserving minimal solve to ≥2.05) + gruvbox→bg4 #7c6f64, nord→#616e88 (community standard), latte→overlay1 #8c8fa1; text-on-surface via surface nudges (ristretto darker, latte lighter); accentAlt darkened (latte, summer-day); warn darkened on the light themes (latte, summer-day, rose-pine-dawn) and flexoki-light's four statuses moved to the canonical 600 series. check-theme-contrast.py gates all four adopted floors (muted ≥2, text/surface ≥4.5, accentAlt ≥3, warn ≥2.5 — warn is the battery tint); good/bad stay audit-only (identity), so no exemption mechanism was needed. Slice (b) done.
Verified: V0 + checks.theme-contrast GREEN (21×7); re-audit: contrast findings 20→1 (miasma bad = identity); flake check green. Rendering is V3 → queued (P2 eyeball entry).
Pending: slices (c) per-surface polish (needs 1080p captures) and (d) whole-swap bar parity; ansi[] conventions decision still open (audit notes).
Next suggestion: slice (c) prep — 1080p captures via the VM recipe, or hardware captures if Bernardo prefers eyeballing first.

2026-07-05 — P1 palette retunes + gated subtext floor (iteration #32, item 28b)

Task: item 28 slice (b), P1s from the #31 audit.
Did: summer-day: surface #5c6a72(=text!)→#efe7d0 raised cream, subtext #fdf6e3(=base!)→#6e828a slate; flexoki-light to canonical Flexoki tones: surface #E6E4D9(ui), overlay #B7B5AC, subtext #6F6E69(tx-2), muted #878580(base-500); miasma surface #000→#2e2e2e; kanagawa surface #090618→#2a2a37 (upstream sumiInk4 — old value was sumiInk0, darker than bg). check-theme-contrast.py gains the adopted subtext/base ≥ 3 pairing (the item-25/27 trap, now impossible). Model note recorded: overlay is a mid-tone role, not a bg tier.
Verified: V0 + checks.theme-contrast GREEN (21×3 pairings); re-audit: all four P1 findings gone (kanagawa now CVD-only findings); V1 — scratch downstream with summer-day ACTIVE builds, new surface hex present in rendered ghostty config. Looks are V3 → queued (four-theme eyeball).
Pending: 28b remainder: P2 floors (muted ≥2, warn ≥2.5, text-on-surface ≥4.5, accentAlt ≥3) + exemption-list mechanism.
Next suggestion: 28b P2s (7 muted retunes + 4 warn + 2 text-on-surface + 2 accentAlt, then gate with exemptions).

2026-07-05 — theme design audit (iteration #31, item 28 slice a)

Task: BACKLOG NEXT#28 slice (a) — the design-theory audit (human-requested; Bernardo asked to run it now).
Did: tools/audit-theme-design.py (new, report-only): OKLCH lightness architecture, extended WCAG pairs, status-hue families, accent harmony, Machado protan/deutan simulation, ANSI slot semantics — over all 21 palettes; plus a visual pass over 4 representative preview.png captures. Findings curated identity-aware into a P1–P3 punch list appended to item 28: 2 invisible-subtext palettes (summer-day, flexoki-light), 2 inverted bg stacks (miasma, kanagawa), 7 muted-floor misses, warn-barely- reads ×4, a systemic "status is never color-only" rule (CVD collapse is inherent, glyph redundancy is the fix), explicit identity exemptions, light-theme ANSI convention decision, bar right-cluster grouping rhythm, latte dark-terminal verify.
Verified: V0 (py_compile; report tool, not wired into checks). Analysis is palette-math + thumbnail previews; per-surface spacing/type audit needs 1080p captures (slice c prep).
Pending: slice (b) — palette retunes + checker extension w/ exemption list; needs Bernardo's eye on the P1 retunes (visual taste calls). Artwork needs none so far — no image-model ask yet.
Next suggestion: slice (b) P1s (summer-day/flexoki subtext+ surface, miasma/kanagawa stacks) — small hex edits, big payoff.

2026-07-05 — greeter theming from the JSON (iteration #30, item 16)

Task: BACKLOG NEXT#16 — tuigreet themed from theme-state.json at system rebuild (the Plymouth model). Item deleted.
Did: new modules/nixos/greeter.nix owns the greetd block (moved out of default.nix, one concern one file) + the theming: (1) console.colors = the theme's ansi[] (mkDefault; lands as vt.default_* kernel params — also themes raw ttys + LUKS prompt); (2) tuigreet --theme on NAMED slots (its parser is ratatui Color::from_str, verified in source; names hit the standard indexes the retinted VT palette now serves). Sparse state w/o ansi → skips cleanly.
Verified: V0; V1 — template-system toplevel: baked greetd.toml carries the full --theme spec, kernel-params carry all three vt.default_{red,grn,blu}. A checks.greeter VM test was attempted and DROPPED: tuigreet dies under runNixOSTest even bare (harness limit, not our flag — bisected; nixpkgs tests use agreety) → MEMORY.md.
Pending: V3 queued (rendering + theme-follows-rebuild).
Next suggestion: #17 launch-or-focus, or the new #28 UI-review audit slice once Bernardo confirms its placement.

2026-07-05 — display profiles auto-switch (iteration #29, item 15b)

Task: BACKLOG NEXT#15 slice b — opt-in hotplug auto-switch.
Did: nomarchy-display-profile gains match <outputs> (exact set → else unambiguous largest subset → ties/none = exit 1); nomarchy-display-profile-watch polls hyprctl monitors all every 3s (exec-once, the keyboard-watcher pattern — session units race Hyprland IPC), reacts only to CHANGES, reads settings.displayProfileAuto LIVE each change (instant menu toggle, no rebuild) and applies via the tool, so auto-picks persist concrete profiles that rebuilds bake. Chose a separate boolean over an "auto" pseudo-profile for exactly that bake semantics. Menu: Auto-switch (on/off) row in Profiles. Template + README notes.
Verified: V0; V1 — scratch 4-profile downstream builds; both scripts bash -n; match matrix exercised: exact ×3, unambiguous subset, tie → none, no-fit → none; watcher exec-once present; menu renders with the row. Real hotplug needs a session+dock → V3 (rides the slice-a queue entry).
Pending: V3. Slice c (workspace binding) remains, tagged optional.
Next suggestion: #16 greeter theming — item 15's remaining slice c is optional; a fresh item keeps variety.

2026-07-05 — display profiles slice a (iteration #28, item 15a)

Task: BACKLOG NEXT#15 slice a, per the #27 design: option + overlay + applier + menu + template example.
Did: nomarchy.displayProfiles (attrsOf (listOf monitorType)); monitor logic extracted to pure modules/home/monitor-rules.nix (rule renderer + 3-layer resolve: profile replaces whole-by-name → resolution picks field-level with the disable-guard → wildcard), hyprland.nix consumes it and gains nomarchy-display-profile (list/active/apply/base; apply = baked hyprctl rules + state write --no-switch, base = clear + hyprctl reload) on PATH only when profiles exist; menu: Display gains a gated "Profiles ›" row + new display-profile mode (●/○ active mark, Base layout row); template example + README row. checks.display-profiles unit-tests resolve (incl. junk-state degradation) — the first HM-eval fixture attempt hit "path not valid" (eval store won't realise writeText/toFile under flake check), hence the pure-function extraction.
Verified: V0 (check's asserts force at eval); V1 — scratch downstream through lib.mkFlake builds: tool bash -n + correct baked rules (disable, vrr, floats), hyprland.conf shows the exact overlay (wildcard + kept base + disabled panel + profile output), menu bash -n, list/usage smoke. Live switching needs a session+dock → V3.
Pending: V3 queued. Slices b (hotplug auto-switch) and c (workspace binding) remain in the item.
Next suggestion: #16 greeter theming, or 15b if momentum on displays is preferred.

2026-07-05 — display-profiles design pass (iteration #27, item 15, no code)

Task: BACKLOG NEXT#15 — the entry itself demands "design first, write the plan into this entry before coding". Backlog-only iteration.
Did: wrote the plan into item 15: nomarchy.displayProfiles = attrsOf (listOf monitorType) (reuse the existing submodule), settings.displayProfile state via the night-light pattern (instant hyprctl apply + bake on rebuild), whole-entry-by-name overlay with a disable-guard against stale resolution picks, self-gated menu row, three slices (option+overlay+menu / opt-in hotplug auto-switch / per-profile workspace binding), kanshi stays a non-goal.
Verified: V0 (docs-only; flake untouched).
Pending: nothing — 15a is now directly codeable.
Next suggestion: 15a (option + overlay + menu picker), or #16 greeter theming if a fresh session prefers a self-contained item.

2026-07-05 — scheduled lock-bump workflow (iteration #26, item 14 b+c)

Task: BACKLOG NEXT#14 [big] — split: (a) checks-on-push shipped earlier; this ships (b) the weekly bump + (c) the fast-lane as workflow_dispatch on the same job.
Did: .gitea/workflows/bump.yml — Mon 05:17 UTC cron + dispatch: nix flake update (pinned branches by construction) → gate nix flake check --no-build (py_compile/bash -n skipped: they don't read the lock) → commit --only flake.lock with the update's change list as body → push main (triggers check.yml as second net). Red gate/push race = failed run, nothing lands. Container recipe from check.yml.
Verified: V0 (yq parse + bash -n each run block); logic simulated end-to-end in a scratch clone with a REAL nix flake update + bare remote: changed/unchanged branches, bystander file stays uncommitted, push lands; today's bumped lock evals green (Monday's baseline). Gitea 1.25.4 confirmed (schedule supported). The real cron pickup + token push are NOT confirmable from here → item 14 slimmed to "confirm first run".
Pending: watch the first scheduled/dispatched bump run.
Next suggestion: #16 greeter theming (next codeable item; #15 needs its design pass written first).

2026-07-05 — color picker (iteration #25, item 13 final slice → item done)

Task: BACKLOG NEXT#13 last slice — hyprpicker → clipboard from the menu + a direct bind. Item 13 complete; deleted from BACKLOG.
Did: hyprpicker into systemPackages (next to grim/slurp); rofi.nix colorpicker mode (hyprpicker → wl-copy → hex-in-toast, Esc = silent cancel) + self-gated Tools › "Color picker" row + usage line; keybinds.nix SUPER+CTRL+P row (feeds bind + cheatsheet).
Verified: V0; V1 — template-home AND template-system build; rendered nomarchy-menu bash -n OK with mode + gated row; cheatsheet txt has the SUPER+CTRL+P row; hyprland conf has the bind; system toplevel ships sw/bin/hyprpicker. Picking needs a compositor → V3.
Pending: V3 queued (loupe → click → wl-paste round-trip).
Next suggestion: #15 (display profiles) needs a design pass written into the entry first; #16 (greeter theming) is the next directly codeable item — or #14's bump workflow (eval-tier gate).

2026-07-05 — low-battery notifications (iteration #24, item 13 slice)

Task: BACKLOG NEXT#13 second slice — the bar colors the battery at 25/10% but nothing notified.
Did: pkgs/nomarchy-battery-notify (shellcheck-gated watcher: polls the same sysfs the bar reads — type=Battery, scope≠Device — one toast per downward crossing: normal@25, critical@10; re-armed by charging; self-gates on battery presence → desktop no-op) + overlay. HM unit in modules/home/battery-notify.nix behind nomarchy.batteryNotify.enable (default on), graphical-session-bound, libnotify on the unit PATH (notify-send resolved from PATH so the check can shim it). README row. Chose the tiny watcher over poweralertd: its UPower thresholds (20/5) wouldn't match the bar, and it toasts every plug/unplug.
Verified: V0; V1 — template-home builds, unit present; V2 GREEN — new checks.battery-notify (test_power fake battery): self-gate exits 0 without a battery, silence at 80%, exactly one low toast crossing 25, one critical-urgency toast crossing 10, charging re-arms → a second drain toasts again.
Pending: V3 queued (real toast through swaync while draining).
Next suggestion: item 13 final slice — hyprpicker color picker (Tools › entry + SUPER+CTRL bind).

2026-07-04 — caffeine/idle-inhibit toggle (iteration #23, item 13 slice)

Task: BACKLOG NEXT#13 first slice — idle_inhibitor in the generated bar (summer-night had it; reverse parity gap).
Did: waybar.nix: idle_inhibitor module (󰾪/󰅶, tooltips, placed after the ⏺), CSS membership + .activated { color: @warn; }. summer-day: module + CSS mirrored from summer-night (same glyphs, no tint — glyph change is the state signal there). keybinds.nix extra row → cheatsheet mentions the bar click.
Verified: V0; V1 — template-home built; rendered config has the module second in modules-right with correct glyphs; CSS rules present; cheatsheet row renders. Inhibit behaviour needs a session → V3 queued (hold past the lock timeout).
Pending: V3. Item 13 remaining slices: low-battery notifications, color picker.
Next suggestion: item 13 next slice (low-battery notifications, gate on power.laptop) — or #14 lock-bump CI if variety is better.

2026-07-04 — screen recording in Capture (iteration #22, item 12)

Task: BACKLOG NEXT#12 — record region/screen from the menu, stop from the bar.
Did: modules/home/recording.nix: nomarchy-record start/stop/active/status — wl-screenrec first, wf-recorder fallback if it dies at startup (no-VAAPI case), SIGINT-finalized stop, runtime pidfile, signal-8 bar pokes. Capture menu: 4 record rows that self-swap to "■ Stop recording" while one runs. Waybar: self-gating custom/recording (⏺ REC, @bad red), click = stop — the one stop surface. Parity: both summer whole-swap bars got the module + CSS (@red). DEVIATIONS: saves to ~/Videos/Recordings (an .mp4 in Pictures/Screenshots felt wrong — same timestamp naming though); audio = two extra explicit rows, not a toggle state.
Verified: V0; V1 — template-home builds; rendered bar config has the module (first in modules-right) + CSS rule; summer jsonc still parse; nomarchy-record dry-run: status empty+exit 0 idle, active exit 1, usage exit 2; both recorders in home-path. Real capture needs a session → V3 (headless GL recipe can't confirm an encoder).
Pending: V3 recording exercise (incl. VAAPI-vs-fallback check).
Next suggestion: NEXT#13 niceties, one slice: idle-inhibit (caffeine) toggle — reverse-parity gap already noted in the item.

2026-07-04 — state-file validation & friendly errors (iteration #21, item 11)

Task: BACKLOG NEXT#11 — hand-edited state must fail with field + problem + fix, never a Nix stack.
Did: (a) tool: validate_state() (appearance schema = hard errors; settings.*/unknown keys = warnings), validate subcommand, validate-BEFORE-write in write_state (invalid set/apply never touches disk), JSON syntax errors now say line/column/fix. (b) theme.nix: same schema at eval time on parsed (post-defaults, so sparse files stay fine); throws a multi-line message naming each field, the got- value, and the three fix paths. Doctor's JSON fix-text now points at validate.
Verified: V0; V1 — new checks.theme-sync-validate GREEN: good template passes, trailing comma → line/col message, bad hex/type → field named, unknown key → warning only, invalid set refused with the file byte-identical (cmp). Negative eval of a broken scratch downstream through mkFlake shows the friendly throw (both fields); good-path template-home still builds.
Pending: nothing — no hardware dependency.
Next suggestion: NEXT#12 (screen recording) or #13 niceties slice; #12 first (it reshapes the Capture submenu the OCR LATER item waits on).

2026-07-04 — nomarchy-doctor (iteration #20, item 10)

Task: BACKLOG NEXT#10 — one-shot read-only health check.
Did: pkgs/nomarchy-doctor (writeShellApplication → shellcheck- gated; a package so the VM check runs on a minimal node): failed system+user units (user bus self-skips), disk space on real filesystems only (fstype allowlist skips tmpfs/9p → VM-safe; device dedupe), state-file parses + git-tracked, flake dirty/behind (warn-level), generation age via the profile SYMLINK mtime (store paths are epoch-1), snapper timer when enabled. Every ✖ prints its fix; exit 1 on any ✖. Wired: overlay, systemPackages, System › Doctor menu row (terminal), README §5.
Verified: V0; V1 package build (shellcheck) + real-hardware smoke: correctly flagged a genuinely failed user unit on the dev box; V2 GREEN — VM test: induced failed unit → exit 1 + names it + prints fix; reset-failed → healthy exit 0. Menu rebuild bash -n ok.
Pending: V3 menu-row check queued. Waybar-warning follow-up → LATER.
Next suggestion: NEXT#11 (state-file validation) — doctor's "parses" check is its little sibling; 11 adds validate-before-write
- eval-time schema messages.

Task: BACKLOG NEXT#9 half (b) — undo one menu away.
Did: new rollback case in nomarchy-menu (System submenu row): lists the last 10 home-manager generations (newest marked current), picking one activates it in a terminal — HM's supported rollback, reversible, so no typed-yes. System-level undo LINKS OUT by design: Snapshots row → existing flow; "boot an older generation (how)" → instruction notification pointing at docs/RECOVERY.md §3 (no destructive reimplementation in rofi). RECOVERY §1 cross-references the menu path. Item 9 complete → deleted.
Verified: V0; V1 — template-home built, generated nomarchy-menu bash -n clean, row-render + pick-parse logic run against a fixture generations output (id/path extraction correct). V2 skipped deliberately: the only novel logic is the parsing (fixture-covered); the activate step is HM's own mechanism. V3 queued.
Pending: V3 rollback-menu exercise on hardware.
Next suggestion: NEXT#10 nomarchy-doctor (V2-able in a VM).

2026-07-04 — sys-update/-rebuild what-changed diff (iteration #18, item 9a)

Task: BACKLOG NEXT#9 half (a) — a human diff after system rebuilds.
Did: both sys-update and sys-rebuild (twins stay twins) capture /run/current-system before, and after the switch print nvd diff before after (store-path-pinned ${"$"}{pkgs.nvd}, || true so it can't fail a succeeded rebuild; identical-path fast-path says "no changes"). README §3 shortcut block notes it.
Verified: V0; V1 — template-system toplevel rebuilt, generated scripts extracted + bash -n, nvd smoke-run against the real closure (correct output format). Runtime behaviour is a trivial shell path — no VM needed.
Pending: half (b) — the System → Rollback menu flow (desktop generations picker; system → boot menu + snapshots pointer).
Next suggestion: item 9b — it completes the item; V2 the generation-activate path.

2026-07-04 — viewers + mime defaults (iteration #17, item 8)

Task: BACKLOG NEXT#8 — PDF/image viewers + xdg-mime defaults.
Did: modules/home/viewers.nix (nomarchy.viewers.enable, default on): programs.zathura (Stylix themes it — target added to stylix.nix's explicit list, autoEnable=false bit us first try) + imv. modules/home/mime.nix (nomarchy.mime.enable): mkDefault xdg.mimeApps for pdf→zathura, image/*→imv, av→mpv, text→code, dir→thunar, html/http(s)→firefox as INERT entries (browser Decision untouched — absent .desktop entries are skipped by GIO). README rows added (option-docs enforces them now). DEVIATION from the item text: viewers are a module toggle, not template packages — zathura's theming needs its HM module, which is exactly the "real config" boundary of the toggle discipline.
Verified: V0; V1 — template-home generation builds; asserted rendered mimeapps.list (all associations), themed zathurarc (Tokyo Night rgb), imv/zathura in home-path. V3 smoke queued.
Pending: V3 open-a-file smoke; browser Decision now smaller (inert entries ship; only "active in suite?" remains).
Next suggestion: NEXT#9 (update & rollback UX) — or #10 doctor.

2026-07-04 — docs slice (c): install story — item 6 COMPLETE (iteration #16)

Task: BACKLOG NOW#6 slice (c), the last one — install/first-run narrative end to end.
Did: read README §2 (ISO/installer) → §3 (downstream) → template README as one story; checked the facts en route: mkFlake's arg surface == the §3 table (src/username/hardwareProfile/system, lib.nix), disko really makes @snapshots + the hibernation swapfile, .#default / .#me names consistent. One seam fixed: §3 showed nix flake init -t but never handed off to the template README's six first-run steps — connective paragraph added (+ "the installer does all of this"). Item 6 done → deleted from BACKLOG (NOW now empty); ROADMAP § Full docs review flipped to ✓ with the slice summary.
Verified: V0 (docs-only) — flake check clean pre-commit.
Pending: docs-site-vs-Markdown stays a Decision.
Next suggestion: NEXT#8 (viewers + mime defaults) is the new queue head; note its "Default browser" Decision dependency — the mime module half can proceed with mkDefault degradation regardless.

2026-07-04 — docs slice (d): recovery runbook (iteration #15, item 6d)

Task: BACKLOG NOW#6 slice (d) — write the undo story.
Did: docs/RECOVERY.md, symptom-ordered: (1) bad theme/HM change → generations / apply-known-good / git-revert ~/.nomarchy; (2) desktop won't start → Ctrl+Alt+F2, greetd + user journals, VM-GL and preactivate-log pointers; (3) system change → systemd-boot older generation (kept: 10) + make-it-stick via sys-rebuild; (4) files → btrfs-assistant / sudo nomarchy-snapshots (undochange, root rollback, typed-yes); (5) ISO + nixos-enter. Every command checked against the tree (greetd cfg, configurationLimit, snapshots script actions, helper names). Linked from README tree + §5.
Verified: V0 (docs-only) — flake check clean pre-commit.
Pending: slice (c) install-story read-through — the last of item 6.
Next suggestion: finish item 6 with slice (c).

2026-07-04 — docs slice (b): drift pass (iteration #14, item 6b)

Task: BACKLOG NOW#6 slice (b) — OVERRIDES.md / TESTING.md / templates/downstream/README.md vs the live tree.
Did: spot-checked every factual claim: mkDefault sites (hyprland behaviour knobs, ghostty paddings, waybar mainBar/style), bind-list normal priority, set ui.gapsOut CLI key, awww(-daemon) naming, 21 presets, tokyo-night's 4 wallpapers, summer-night's light @fg bar, live ,highres mkForce, .gitea/workflows/check.yml, QEMU gl flags, template .#me + ~/.nomarchy/NOMARCHY_PATH. All accurate. Fixed the two real gaps: OVERRIDES §2 predated nomarchy.monitors (now routes there + quick-ref row); TESTING §1 lacked the shell-syntax check §1b claims CI runs (added bash -n).
Verified: V0 (docs-only; flake untouched — check run pre-commit).
Pending: slices (c) install story, (d) recovery runbook.
Next suggestion: item 6 slice (c) or (d) — (d) is self-contained and ships user value (the runbook).

2026-07-04 — docs slice (a): option-surface reconcile (iteration #13, item 6a)

Task: BACKLOG NOW#6, first slice — README tables vs live options.
Did: split item 6 into slices (a–d) in BACKLOG. Reconciled: 9 undocumented options (autoTimezone.enable!, camera.hideIrSensor + irMatch!, intel.guc, amd.pstate/vaapi, package, system.stateFile, restic.paths) — rows/inline mentions added, no stale rows found. Made it permanent: checks.option-docs — flake.nix eval-walks option names from the 4 option files (config halves stay lazy; dummy args) → tools/check-option-docs.py enforces row-or-.leaf-mention + flags stale rows.
Verified: V0; V1 — check builds green (78 options); negative test: same script vs the pre-fix README reports exactly the 9 gaps.
Pending: slices (b) doc drift pass, (c) install story, (d) recovery runbook.
Next suggestion: item 6 slice (b) — or NEXT#8 (viewers+mime) for variety; both fine.

2026-07-04 — Back audit (iteration #12, item 24)

Task: BACKLOG NOW#24 — every list menu ends in ↩ Back.
Did: audited every rofi -dmenu call site in the generated nomarchy-menu + nomarchy-vpn. Already compliant: power, power-profile, theme, clipboard, files, audio, display (both stages), capture, tools, system, VPN main/import/exit-node/tailscale. Fixed: keybinds cheatsheet (Enter was a silent exit) now ends in ↩ Back → root picker. Documented the exceptions in rofi.nix: external modi (calc/emoji/networkmanager_dmenu/rofi-pulse-select) can't take an injected row (Esc = back); free-text prompts (web/ask) have no list.
Verified: V0; V1 — built nomarchy-menu, bash -n OK; awk audit over the generated script: only rows lacking back are the root picker (by design) and the free-text prompt. V3 spot-check queued.
Pending: V3 spot-check. NOW now holds only item 6 (docs, [big]).
Next suggestion: item 6 first slice, or NEXT#8 (viewers+mime).

Task: BACKLOG NOW#22 — nameless entries in the network picker.
Did: read the pinned networkmanager_dmenu 2.6.2 source: with compact = False (our config!) it inserts literal empty rows as section separators (eth/wifi/vpn) — the systematic blank entries. Flipped config.ini to compact = True (also unpadded row format, better in rofi's proportional font). Hidden-SSID APs additionally render ONE nameless sec+bars row (dedup keeps one; not config- filterable) — documented, V3-gated, source-patch follow-up if seen.
Verified: V0; V1 — rendered config.ini shows compact = True. Behavior needs Wi-Fi hardware → V3 queued.
Pending: V3 on the Latitude (no separators; hidden-AP residual?).
Next suggestion: item 24 (Back audit) — grep-the-generated-script.

2026-07-04 — waybar/rofi palette-safe pairings (iteration #10, item 27)

Task: BACKLOG NOW#27 — generated waybar/rofi CSS invisible on on-surface palettes (flexoki-light runs it; summers whole-swap).
Did: waybar.nix generated CSS: module row + #window → alpha(@text, 0.85); inactive workspaces / dnd / pulseaudio.muted → alpha(@text, 0.5) (gruvbox's muted/base was 1.27). rofi.nix: @dim → text+80, @surface → text+1a (#RRGGBBAA fg tints). Contrast check docstring/pairings now cover all three generated surfaces. Status glyphs (good/warn/bad on base, worst 2.05) left as-is — palette design, human call. Generated bar only; whole-swaps untouched.
Verified: V0; V1 — checks.theme-contrast green (21 themes), downstream-template-home builds, rendered waybar style has zero @subtext/@muted/@surface, rendered custom.rasi shows fg-derived tints. NOT visually verified — V3 queued (flexoki-light eyeball).
Pending: V3 flexoki-light + dark-theme regression spot-check.
Next suggestion: item 22 (network menu rows) — V1-actionable.

2026-07-04 — swaync contrast fix + theme-contrast check (iteration #9, item 25)

Task: BACKLOG NOW#25 — swaync text invisible on summer-day.
Did: swaync.nix now uses only palette-safe pairings: body/summary = @text on @base (worst ratio 5.18 across all 21 themes; subtext/base was 1.00 on summer-day + flexoki-light), hover rows + widget buttons = alpha(@text, 0.1) tints (contrast by construction); dropped the subtext/surface defines. NEW permanent check checks.theme-contrast (tools/check-theme-contrast.py) asserts the hex-on-hex text pairings across every theme JSON. Audit found the same mispairing in generated waybar/rofi CSS, live on flexoki-light (no whole-swap) → NOW item 27.
Verified: V0; V1 — checks.theme-contrast builds green (21 themes), rendered services.swaync.style contains zero @subtext/@surface. NOT V2/V3-verified visually — queued (Bernardo is at the Latitude).
Pending: V3 notify-send readability check on summer-day; item 27.
Next suggestion: item 27 (same bug class, waybar/rofi surfaces).

2026-07-04 — SUPER+? bind fix (iteration #8, item 26)

Task: BACKLOG NOW#26 — cheatsheet bind never fires on hardware.
Did: keybinds.nix: $mod, question → $mod SHIFT, question (? is Shift+/, Hyprland matches the exact modmask); rofi.nix prettyKeys collapses "SHIFT + ?" → "?" so the row stays SUPER + ?. Gotcha added to MEMORY.md (shifted keysyms need SHIFT in mods).
Verified: V0 (flake check --no-build); V1 both surfaces — eval'd hyprland settings.bind renders $mod SHIFT, question, exec, nomarchy-menu keybinds, and built nomarchy-menu's cheatsheet file renders SUPER + ?. No other shifted keysyms in the bind list.
Pending: V3 queued (SUPER+? on the Latitude after home-update + relogin).
Next suggestion: item 25 (swaync contrast on summer-day).

2026-07-04 — Latitude hardware sweep #2 (Bernardo; results booked, no code)

Task: V3 sweep on the Latitude after pulling ≥18b8545 + reboot.
PASS: waybar supervisor (theme switches clean; pkill -x waybar respawns instantly); snapshots — themed polkit prompt → GUI as root, fzf fallback works (restore/rollback gate not exercised → residual queued); sys-rebuild; night-light enable/disable; auto-commit (theme commits + the toggle self-commit); hibernate single-unlock + suspend-still-locks (1b0eeea CONFIRMED).
PASS (late confirm): auto-timezone — timedatectl shows Europe/London.
Bug: SUPER+? never fires — $mod, question misses because a real ? carries SHIFT in the modmask → BACKLOG 26 (bind-only: the menu itself opens fine from a terminal). SUPER+SHIFT+K no-op was the comma-layout gate working (none configured) — retest queued.
Next suggestion: item 25 (swaync contrast) or 26 (cheatsheet bind).

2026-07-04 — Waybar resilience + sys-rebuild (iteration #7, items 21+23)

Task: the two top Latitude findings.
Did: (21) nomarchy-waybar supervisor (waybar.nix, exec-once'd from hyprland.nix): ANY waybar exit respawns, crash-loop guard (5 fast exits → critical notify + stop), TERM trap for a real stop. theme-sync now prefers a clean pkill -x waybar when the supervisor is running (restart with fresh config+style — avoids waybar's in-place SIGUSR2 reload entirely, the suspected double-reload race with reload_style_on_change during symlink flips); SIGUSR2 stays as the unsupervised fallback; reload_style_on_change kept for manual home-update restyles. (23) sys-rebuild — snapshot-first rebuild against the current lock, no nix flake update; README §3/§5 + motd.
Verified: V0; V1 (HM renders exec-once=nomarchy-waybar, supervisor bash -n; sys-rebuild in systemPackages). V2: headless software-GL desktop VM (recipe from memory — gotchas hit: node needs the overlay, users.users., useGlobalPkgs) — SIGKILL→new pid, clean kill→respawn, SIGUSR2→alive. Result recorded below once green.

Pending: V3 on the Latitude: theme switch survives / bar returns.

Next suggestion: item 22 (network menu rows) or 24 (Back audit).

2026-07-04 — Keyboard cycle bind + parity (iteration #6) + Latitude triage

Task: BACKLOG NOW#5; mid-iteration Bernardo delivered Latitude 5410 hardware-QA findings (the machine got a real session!).
Did: multiLayoutBinds in keybinds.nix (SUPER+SHIFT+K → switchxkblayout current next), gated in hyprland.nix AND the rofi cheatsheet on the same comma-in-layout condition; hyprland/language
- #language CSS added to both summer whole-swaps (shows even single-layout there — static JSON can't gate; deliberate). Reverse parity gap found (summer-night has idle_inhibitor, generated bar doesn't) → folded into item 13. Triage: four new NOW items from hardware: #21 Waybar crash on theme switch + no respawn (top), #22 nameless network-menu entries, #23 sys-rebuild (no-update rebuild), #24 Back-everywhere audit.
Verified: V0 (JSON + flake check) + V1 both directions (single-layout build: no bind; us,de via extendModules: bind + the cheatsheet row render; summer JSONs parse).
Pending: V3 cycle-bind check queued; items 21/22/24 carry V3 re-checks on the Latitude.
Next suggestion: #21 (waybar crash — stability pillar, user-hit).

2026-07-04 — Webcam follow-ups (loop iteration #5, quick item)

Task: BACKLOG NOW#7 (Bernardo asked for "a quick item").
Did: template home.packages gains a commented "Webcam tuning" pair (cameractrls GUI, v4l-utils CLI — bare packages, so template not toggle, per the option-surface rule); the portal/Flatpak libcamera IR gap is written up as a PROPOSED item with three options (recommend: document-only now, libcamera-rule investigation when the T14s is available). CI runs #60/#61 confirmed green (earlyoom + snapshot fix).
Verified: V0 (flake check — template evals through mkFlake; changes are comments + agent docs).
Next suggestion: NOW#5 (keyboard cycle bind + summer parity) — the last code item in NOW besides the docs pass.

2026-07-04 — Snapshot GUI un-broken + polkit agent (loop iteration #4)

Task: BACKLOG NOW#4 — btrfs-assistant segfault.
Did: re-diagnosed instead of overriding. gdb: crash is in libbtrfsutil's unprivileged subvolume iteration (btrfs-progs 6.17.1; upstream fix 886571653 post-6.17.1; symbol versions verified — not ABI drift). VM A/B: root works (exit 0), user crashes (139). The launcher runs it as root → GUI fine if pkexec can prompt — and the distro shipped no polkit agent (all pkexec silently failed; the actual root cause of "GUI is dead"). Shipped: hyprpolkitagent (exec-once, Stylix-themed Qt), menu re-pointed to the launcher with the fzf flow as fallback, checks.snapshot-gui canary (root path + offscreen event loop on real btrfs). No btrfs-progs patch — root-side flows don't hit the bug; the fix arrives with a lock bump.
Verified: V0 + V2 (snapshot-gui check, run for real) + V1 (HM generation renders the agent exec-once; menu bash -n).
Pending: V3 queued — first on-hardware themed polkit prompt.
Next suggestion: NOW#5 (keyboard cycle bind + summer parity).

2026-07-04 — CI is green (run #58) — item 20 main half closed

Task: monitor follow-up, no code. Run #58 (the .gitea path fix) succeeded — first green CI: the container recipe works on the live runner, eval tier confirmed end to end.
Did: BACKLOG item 20 reduced to its stretch (KVM runner → enable vm-checks), moved to NEXT. Run #59 (earlyoom push) monitored.
Next suggestion: NOW#4 (btrfs-assistant override attempt).

2026-07-04 — Memory-pressure protection shipped (loop iteration #3)

Task: BACKLOG NOW#3 — default-on userspace OOM killer.
Did: modules/nixos/oom.nix: earlyoom (mkDefault on, desktop notifications, session-plumbing --avoid list, no --prefer) + systemd.oomd.enable = false. Choice rationale: Hyprland session is one cgroup scope → oomd's cgroup-level kill would nuke the desktop; earlyoom is process-level. Found en route: nixpkgs default-enables oomd inert (enable=true, all slice toggles false). README "beyond the surface" note added.
Verified: V0 (flake check) + V2 — new checks.oom-protection runNixOSTest, run for real: hog (chunked allocator, 686 MB peak in a 1 GB VM) SIGTERM'd by earlyoom in 0.1 s, bystander unit survived, oomd asserted inactive. 25 s test, cheap enough for the suite.
Pending: nothing hardware-specific (behavior fully VM-provable).
Next suggestion: NOW#4 (btrfs-assistant override attempt). Also: CI run #58 (the .gitea fix) was running at commit time — check its outcome next tick; this push triggers #59.

2026-07-04 — CI fix: wrong workflow dir (correction to iteration #2)

Task: Bernardo reported no run despite an active runner + shared his compose file: the server is Gitea (gitea/act_runner), not Forgejo — I'd shipped the workflow to .forgejo/workflows/, which Gitea never reads. The legacy repo's .gitea/workflows/ path was the clue I under-weighted.
Did: git mv → .gitea/workflows/check.yml; corrected the Forgejo references in the workflow header, TESTING.md §1b, BACKLOG item 20 (now: watch the first run; runner was never the problem); MEMORY.md gains the Gitea-not-Forgejo line.
Verified: the push of this commit is itself the test — the run must appear; monitoring.
Pending: first green run (item 20); runner-container memory headroom is the watch-out.

2026-07-04 — CI checks-on-push shipped (loop iteration #2)

Task: BACKLOG NOW#2 — Forgejo Actions workflow.
Did: .forgejo/workflows/check.yml (push to main/v1 + dispatch): flake check --no-build, theme-sync py_compile, bash -n over tracked .sh. Scoped to the eval tier deliberately — API probing + archaeology of the legacy .gitea/workflows/check.yml (57 runs on this instance) showed the runner is an act_runner docker container (no KVM/systemd), so the VM suite can't run there; a commented vm-checks job documents the KVM-runner upgrade path. Legacy's hard-won container gotchas (nixbld setup, sandbox=false for Stylix IFD, Nix 2.31.5 pin) carried over verbatim. TESTING.md §1b added.
Verified: V0 locally (same commands the workflow runs, minus the container Nix install) + yq YAML parse. The real green run needs the runner to be alive — not confirmable from here (runners API is 401 unauthenticated); watching the run after push.
Pending: new [human] item 20 — confirm/re-register the runner; stretch: a KVM runner unlocks the vm-checks job.
Next suggestion: NOW#3 (memory-pressure protection).

2026-07-04 — Opt-in auto-commit shipped (loop iteration #1)

Task: BACKLOG NOW#1 — auto-commit of state mutations (in-flake state Phase 4). First autonomous /loop iteration.
Did: settings.autoCommit live-read by nomarchy-theme-sync; pathspec-limited commit of theme-state.json on apply/set (fires if flag on before OR after the write → the off-toggle commits too); identity fallback; same-value no-op; bg excluded. Menu: System › Auto-commit (instant, self-gated on .git). Rider: get prints booleans as JSON true/false, fixing the stuck "Auto timezone (on/off)" menu label. BACKLOG numbering switched to stable IDs (no renumbering).
Verified: V0 (py_compile, flake check) + V1 (HM generation builds; generated menu bash -n) + a 7-assertion sandbox git-repo round-trip (enable/set/same-value/disable/post-disable/apply/identity-fallback).
Pending: V3 queued in HARDWARE-QUEUE (on-machine toggle + apply).
Next suggestion: NOW#2 (CI checks-on-push) — but its runner half may need Bernardo to register a Forgejo runner; deliver the workflow regardless.

2026-07-04 — Backlog revision pass (human-requested)

Task: Bernardo asked for a philosophy-aligned revision of BACKLOG.md ("I'm sure I'm forgetting important items").
Did: verified guessed gaps against the tree first (GC/optimise/boot limit + all 21 previews already ship — dropped those ideas). Added the confirmed gaps: CI-on-push promoted to NOW#2 (agents push to main now), memory-pressure protection NOW#3 (nothing mitigates OOM; the release bump died to one), viewers+mime defaults #8, update/rollback UX #9, nomarchy-doctor #10, state-file validation #11, screen recording #12, niceties batch #13 (idle-inhibit, low-battery notify, hyprpicker), OCR in LATER; recovery runbook folded into the docs item; zram + default browser added to Decisions.
Verified: V0 (docs-only).
Pending: Bernardo to sanity-check the new NOW ordering, esp. CI at #2 and the two new Decisions.
Next suggestion: unchanged — NOW#1 (opt-in auto-commit).

2026-07-04 — Bootstrap the loop infrastructure (this commit)

Task: create agent/ (human-requested): LOOP/GOALS/BACKLOG/ CONVENTIONS/MEMORY/HARDWARE-QUEUE/JOURNAL, root CLAUDE.md, README + ROADMAP pointers.
Did: reworked the forward half of docs/ROADMAP.md into the tiered BACKLOG (5 NOW / 6 NEXT / LATER / Decisions); collected every pending on-hardware check into HARDWARE-QUEUE.md; seeded MEMORY.md from the ROADMAP's decision records.
Verified: V0 (docs-only change; nix flake check --no-build run to confirm the tree still evaluates).
Pending: Bernardo to sanity-read GOALS.md pillars + BACKLOG tiering — reorder freely, the order is the instruction.
Next suggestion: BACKLOG NOW#1 (opt-in auto-commit — Phase 4 of in-flake state, already the agreed next step).

134 KiB Raw Blame History Unescape Escape