Nomarchy/agent/BACKLOG.md

Backlog — the prioritized task queue

This is the only executable work list for agents. Product themes and v1.0 intent live in docs/VISION.md; design history in docs/ROADMAP.md; map in docs/README.md and agent/README.md.

Rules:

• Agents take the topmost actionable item (see LOOP.md). Finished items are deleted here — the journal + git log are the record; durable design notes get a ✓-entry in docs/ROADMAP.md (and/or a note in VISION) if worth keeping.
• Item numbers are stable IDs — never renumbered or reused. A gap in the sequence means shipped (or dropped) work; new items take the next free number regardless of tier.
• Tags: [blocked:hw] needs real hardware (see HARDWARE-QUEUE.md) · [human] needs Bernardo · [stuck] two failed attempts, needs help · [big] must be split before starting.
• Agents may append to PROPOSED and Decisions freely (include VISION § … or ROADMAP § … when relevant); only Bernardo moves items out of PROPOSED into the tiers.

---

NOW

Live ISO hardware findings — Acer Aspire M5-481T (Bernardo, 2026-07-13)

These are separate queue items from one real install/session pass. Preserve that separation when fixing them: the installer blocker, unclear installer copy, and post-install desktop failures have different verification paths.

118. smartd fails where no drive has SMART, so the health icon greets you red

Bernardo, live ISO 2026-07-14: the Waybar doctor module shows a smartd error.

Root cause, reproduced headlessly 2026-07-14 (scratch runNixOSTest, a node whose only config is services.smartd.enable = true — the same block modules/nixos/default.nix:166 mkDefaults ON for every machine):

smartd[567]: In the system's table of devices NO devices found to scan
smartd[567]: Unable to monitor any SMART enabled devices. Exiting.
smartd.service: Main process exited, code=exited, status=17/n/a
Status: "No devices to monitor"

smartctl --scan prints nothing on such a machine — which is both the cause and the ready-made gate. The rest of the chain is each part working correctly: systemctl --failed lists smartd.service, so nomarchy-doctor's generic failed-unit check (nomarchy-doctor.sh:24-28) reports "failed system unit(s): smartd.service", and Waybar's #custom-doctor goes @bad. Fix smartd, not the doctor — the doctor is telling the truth about a unit that genuinely failed.

Scope is wider than the live ISO (which is why it's here and not filed as a live-only nit): services.smartd.enable mkDefaults true on nomarchy as well as nomarchy-live, and QEMU virtio disks expose no SMART — so a plain VM install boots to a red health icon out of the box, as does a live USB whose stick has no SMART. Every V2 QEMU run has been showing this.

Fix direction: self-gate on the hardware, the convention the rest of the distro follows (ExecCondition, like the night-light; "each half self-gates on its hardware"). A condition that runs smartctl --scan and skips the unit when it finds nothing leaves the unit inactive rather than failed, so real SMART failures on real drives still surface. Do not paper over it with SuccessExitStatus = 17 — that also swallows the case where smartd dies on a machine that does have drives, which is the whole reason the daemon is here.

Pass = on a machine with no SMART-capable device the unit is inactive (not failed), systemctl --failed is empty and the doctor is green; on a machine with one, smartd runs as it does today; a permanent checks.* covers both halves (the scratch repro above is most of it — the no-SMART node exists, the with-SMART node needs a QEMU disk that answers SMART, or the gate script tested directly against a stubbed smartctl).

94. Live ISO/install: no default browser observed

Progress 2026-07-13: installed path VERIFIED at V1 — the exact HM generation the ISO pins for offline installs contains the chromium binary, chromium-browser.desktop, and all three HTTP/HTML handlers (mime.nix sets them; template ships the package). The Acer sighting is the LIVE session, which shipped NO browser by design (mime defaults name chromium, GIO skips it while absent).

Update 2026-07-14: the live half is gone — #103 shipped, so the live session now carries chromium (plus office/editor/music/camera) on PATH and in the launcher, and checks.live-baseline-apps asserts HTTPS still resolves to chromium-browser.desktop and that the entry it names is actually present — the exact "mime names a package nothing ships" trap this item found. What remains here is only the installed path: a test-install VM run proving chromium launches post-install (chain #97's Bluetooth-click V2 into the same run).

Chromium is the resolved default-browser decision and is present in the downstream template, but this hardware pass found no usable default browser. Trace the live-to-installed Home Manager path rather than merely checking the template source. Pass = Chromium launches after installation and HTTP/HTTPS mime defaults resolve to chromium-browser.desktop (and the intended live-ISO browser posture is explicit).

119. text/plain names vscode, which the live ISO does not ship

Found while shipping #103 (2026-07-14), deliberately not fixed there — same bug class as #94, one mime key over. modules/home/mime.nix:48 assigns text/plain = code.desktop, and vscode ships only in the template, never on the live ISO. GIO silently skips an entry whose desktop file is absent, so opening a text file in the live session finds no handler at all — the exact shape of the no-default-browser bug, and now the only one left of its kind since #103 put chromium on the live medium.

Two ways it bites, and a fix should settle both: (1) the live session, which now ships gnome-text-editor (org.gnome.TextEditor.desktop) — the obvious live handler; (2) any installed machine where the user takes the template at its word and deletes the vscode line (opt-out = delete the line, per the option-surface convention), which silently breaks text/plain with no warning.

The general trap is worth fixing once rather than per-key: a mime default naming a package the profile does not carry is invisible until a user double-clicks a file. checks.live-baseline-apps now guards exactly this for HTTPS (asserts the named .desktop is present, not merely named) — the cheap version of this item is extending that assertion to every key in defaultApplications, which would have caught both #94 and this. Pass = no mime default names a desktop entry the profile lacks, on the live ISO and on a template install, and a guard fails on a regression.

95. Live ISO/install: Ghostty does not open

Reproduce from the launcher and a terminal, capture its stderr/journal, and fix the packaging/session/runtime cause. Pass = the default terminal opens in the installed graphical session and the SUPER+Return path works.

98. Boreal: button text renders black and is difficult to read

Audit GTK button foreground/background contrast under Boreal, including the installer surface where it was observed. Follow THEME-DESIGN's two-theme visual protocol. Pass = normal, hover, focused, and disabled button labels remain legible and palette-consistent.

Progress 2026-07-13: scripted checks pass all 24 themes — not palette-level. Stylix gtk.css for Boreal audited: fg roles are light (#d3dae0); the near-black values are accent/warning/error_fg_color (#21272f) on their pastel bgs (deliberate, ~4.4:1). Candidate surfaces: adw-gtk3 (light base) + dark recolor edge cases, or gum's TUI confirm buttons in the installer (VM gum screens under Boreal look legible — artifacts in /tmp/nomarchy-v2-swap-93/). Needs Bernardo's screenshot or an Acer repro to pin the actual widget before fixing.

113. Live session: offline theme switch rebuilds the world from source

Surfaced by #99's evidence run (2026-07-13, /tmp/nomarchy-v2-theme-99/): in the live ISO session, offline nomarchy-theme-sync apply gruvbox fails — its home-manager switch tries to build 1176 derivations from the stage0/hex0 bootstrap up, then dies on offline source fetches (vala, yasm, config.jsonc, ghostty-config, easyeffects.svg, typogrify…), ending error: Build failed due to failed dependency. This contradicts docs/TESTING.md item 6 ("apply gruvbox → switch runs offline"). The failing generation is bvl30ggn…-home-manager-generation.drv; it is not the drv the ISO pins (the pin is the default-theme homeConfigurations.${username}.activationPackage, i.e. Boreal — confirmed drv-identical to the failing one only for gruvbox state, so a non-default theme's generation is unpinned). NOT caused by #99's git-seed: the drv is byte-identical path-seeded vs git-seeded. Decide the contract: either pin every baked theme's live activationPackage into the ISO (size cost) or scope the offline-switch promise to the shipped theme + correct TESTING.md/MEMORY. Pass = a fresh live ISO either switches to any baked theme offline, or fails with a clear "needs network for this theme" message and accurate docs. Not caused by anything in today's batch — separate pre-existing ISO-pinning gap.

NEXT

120. A netinstall ISO, next to the fat offline one

Bernardo 2026-07-14, after seeing the measured size: keep the current ISO exactly as it is — the guaranteed offline install is the feature it buys — and ship a much lighter netinstall variant alongside it. Two products, one distro: "works on a plane" and "8 GiB is absurd to download" are both true, and a second target settles them without compromising either.

Measured facts (2026-07-14), so this starts from numbers, not vibes. (Taken before #121 landed; it removed a duplicate 687 MiB chromium, so the uncompressed figures below are each ~0.67 GiB lower now — the live desktop is 13.34 GiB, not 14.01. The shape of the argument is unchanged. Re-measure before acting rather than trusting these.)

Read this before using the numbers below. They are closure arithmetic, and #121 proved the hard way that closure size is neither disk size nor image size: removing a 687 MiB path shrank the ISO by 8 KiB, because mksquashfs dedupes duplicate files and auto-optimise-store hardlinks them on disk. So a change that looks like it sheds gigabytes of closure can shed nothing off the actual image. Measure the artifact — build the ISO and stat it. The corollary cuts the other way and is the good news for this item: what dedupe cannot help is the wire, so a netinstall's download is the one figure closure/NAR size predicts honestly (nix path-info --store https://cache.nixos.org --json gives the real downloadSize).

• Current ISO 8.078 GiB compressed; 18.03 GiB of store uncompressed (zstd -19, 2.23:1 — compression is already near-max, not the lever).
• The offline pin (system.extraDependencies, 60 roots: a representative installed system + the template HM closure + all flake inputs) is 4.02 GiB uncompressed of that — only ~22%. Dropping it entirely still leaves a ~13.3 GiB desktop → roughly 6 GiB compressed at the same ratio. So "no pin" alone is NOT the lighter ISO — this is the trap to avoid.
• The desktop's own top weights: libreoffice 1457 MiB, initrd 1369, linux-firmware 770, chromium 704 (was 1391 before #121), llvm-lib 540, bibata-cursors 322, mesa 264, mbrola-voices 259, nerd-fonts ~420 combined. Note what that list implies: no single lever gets a desktop ISO under ~4 GB — which is the case for (b) below.

So the real decision is what a netinstall ISO IS, and it should be settled first ([human]): (a) the full try-before-install desktop minus the pin (~6.3 GiB — barely lighter, probably not worth a second target); (b) a TUI installer only, no desktop (~1 GiB, the actual "netinstall" in the Debian sense) which drops "try before install" from that medium — the fat ISO still offers it; (c) a middle desktop (no libreoffice/chromium — but note #103 just put those there deliberately, and a demo desktop that can't browse is the bug #103 fixed).

The gotcha that decides feasibility: without the pin, a netinstall target fetches from cache.nixos.org for stock nixpkgs paths — but Nomarchy's own derivations are in no binary cache, so they would build from source on the user's machine during install. That is the same failure tools/vm/gap-analysis.py exists to diagnose (and #113 is a live instance of). So this item probably depends on a public binary cache (cachix) for the flake's own outputs, or it trades an 8 GiB download for a 40-minute install. Establish that before building the target.

Pass = a second, documented ISO target that is substantially smaller (state the measured number, both ISOs built from one tree), installs successfully with a network in a QEMU run, says clearly at boot that it needs one, and leaves the offline ISO's behaviour untouched (checks.* for the offline path stay green).

115. Suspend-then-hibernate, with a way to set it up

Bernardo, 2026-07-14: a suspended laptop should be able to fall through to hibernate after a while, so a bag-carried machine stops draining, and setting that up must be easy — not a systemd-sleep man-page trip.

The mechanism is systemd's suspend-then-hibernate plus HibernateDelaySec; the work is the surfacing, and the constraints are already in the tree. Hibernate needs the resume offset + swap the installer detects (resumeOffset/rootUuid in patch-template.py) — on a machine without them this must self-gate, not fail at suspend time, the way Battery limit gates on a battery. modules/nixos/default.nix already declines to hyprlock before an encrypted hibernate (the LUKS resume gates it); check that reasoning still holds when the suspend leads to hibernate. Menu placement: it belongs with the power settings, not the root (§ menu placement convention).

Settle first ([human]): whether the delay is a preset list (30 min / 1 h / 2 h / never) or a free-form entry, and whether it applies on battery only or always — the same on-AC question modules/home/idle.nix's suspend listener already answers with ${onAc} ||.

In-flake state like the rest (settings.*), menu-writable, and read via theme-state-read.nix on the NixOS side — NOT config.nomarchy.settings, which does not exist there (ROADMAP § NixOS-side state bridges (#116); copy the bluetooth.enable shape in modules/nixos/default.nix and add a case to checks.state-bridges). Pass = the toggle survives a rebuild, a suspended machine hibernates after the configured delay and resumes, and a machine without hibernate support hides the row instead of offering a suspend that never wakes.

107. Rename theme.json to reflect that it is the system state

The state file long ago stopped being about themes: it carries night-light, keyboard memory, display resolution and profiles, auto-commit, services and more, so theme.json now misnames its own contents and misleads anyone reading the flake. Rename it to state.json (Bernardo's call, 2026-07-14 — settled, do not revisit), and carry the nomarchy-theme-sync tool name along with it.

Standalone task — do not fold it into a feature. Ships with a compatibility shim that keeps reading an existing theme.json so downstream checkouts do not break on a pull, plus a migration note. Pass = a fresh install and an existing downstream checkout both work, every in-repo reference (modules, tools, template, docs) uses the new name, and the shim is documented with the release it can be dropped in.

104. Runtime Airplane mode

Add Airplane mode under System connectivity controls. It must disable Wi-Fi and Bluetooth together, remember their prior states, and restore those states when disengaged. Pass = the runtime round trip works, and a Waybar indicator is visible only while engaged with parity across every whole-bar swap.

105. [big] System-menu information architecture

Keep exactly six root menu entries while reorganizing System into Connectivity, Devices, Recovery, and Preferences. This item must be split into scoped implementation/verification tasks before work starts. Pass = every current route has one deliberate home, navigation remains shallow, and no root-level entry is added or lost.

106. Internal menu Back/Left navigation contract

Reproduce the reported internal leaf with neither Back nor Left behavior, then fix any breach of the already-intended navigation contract. Add a permanent guard covering every internal leaf; external GUIs and free-text prompts retain their explicit Esc exceptions. Pass = no internal leaf can strand the user and the guard fails on a regression.

108. Keybindings menu presentation and completeness audit

Group the menu as Window, Workspace, Menu, and Media, then prove every live Hyprland binding appears from the canonical source. Do not repeat the obsolete claim that float/move bindings are absent. Pass = presentation is scannable and an automated comparison detects omissions or stale displayed bindings.

110. [big] Retire Control Center safely

Split this into two phases before implementation: first build a migration/drop matrix for every unique setting, then remove the package and all references. Keyboard and terminal replacements must reject unsafe free-text values. Pass = no supported setting silently disappears and the second phase leaves no stale launcher, menu, package, documentation, or generated-artifact reference.

Facts for the matrix, from #116 (2026-07-14): its Bluetooth and Printing rows write settings.{bluetooth,printing}.enable, and those keys became live bridges to the NixOS config in a9f3a64 — until then they wrote JSON nothing read, so these two toggles have never actually worked for any user. That cuts both ways: "preserve existing behavior" is not a reason to keep them (there is no behavior to preserve), but the state keys and their bridges are real now and outlive the TUI — dropping the rows must not drop the keys, or a machine whose theme-state.json already says bluetooth.enable = false will silently flip back on at the next rebuild. checks.state-bridges covers those two keys and will fail loudly if the bridges go with the tool; keep it green, and if a key is deliberately retired, remove its case in the same commit. Whatever inherits the rows should rebuild on toggle rather than print "requires rebuild" (#117). Updates (settings.updates.enable) is HM-side, where nomarchy.settings genuinely exists, and needs none of this.

111. Scope-first Recovery menu

Replace confusing Rollback/Snapshots duplication with explicit Desktop generation, System boot generation, and Files/root BTRFS scopes. Pass = labels state what is restored and from where before action, with destructive or reboot effects made clear and each existing recovery path represented once.

112. Installer disk-picker safety

Exclude floppy, pseudo, tiny, and otherwise non-installable devices such as /dev/fd0; never select them by default. Reproduced in QEMU 2026-07-13 (#93's V2 run): with OVMF the guest exposes /dev/fd0 and the picker listed it FIRST — a blind Enter selected it (artifacts: /tmp/nomarchy-v2-swap-93/20-review.png, Disk: /dev/fd0 (WILL BE ERASED)). Pass = the real install disk is clear, invalid devices cannot reach destructive setup, a permanent guard covers the filter/default logic, and a pre-destructive KVM run proves the picker behavior.

LATER

• Wallpapers artifact split (ROADMAP § Faster switches — decided, deferred): pinned Nomarchy-wallpapers input so a state write stops re-copying 86 MB. Follow-on: pre-built theme variants if switches are still slow after.
• Installer round 2 (ROADMAP § Installer): multi-disk BTRFS RAID, impermanence, BIOS/legacy boot.
• Boot-from-snapshot: a systemd-boot equivalent of grub-btrfs.
• MIPI/IPU software-ISP camera support (no-UVC machines).
• NixOS release bump → v2 [human]: deliberate, hand-edited, never automated; the previous attempt was discarded (2026-06-22) over a Hyprland OOM blocker — see MEMORY.md before retrying (NOW#3 should also soften that blocker class).

FUTURE (decided deferred — not the agent queue head)

Work we intend someday but explicitly not NEXT. Agents do not pick these unless Bernardo promotes one into NEXT/NOW.

20. KVM runner → VM suite in CI [human]

Status (2026-07-10): keep eval-only CI on the current Gitea stack (act_runner in docker-compose on the 4c/4 GB IONOS VPS). Nested KVM + RAM headroom on that host are a poor fit next to Gitea; full checks.* VMs stay local / promotion-time until a separate KVM-capable machine exists.

When ready: register a second runner (host-mode nix + /dev/kvm, label nix-kvm — not the existing docker eval runner), then uncomment the vm-checks job in .gitea/workflows/check.yml (runs-on: nix-kvm, nix flake check + toplevel/HM builds). Do not enable the job until that label is online (Gitea queues forever otherwise).

Formatter — adopt later [human]

Intent: add a Nix formatter (likely nixfmt-rfc-style) in a dedicated pass: reformat the tree once, document in CONVENTIONS, optional CI check. Not the queue head — no drive-by reformats until that pass.

PROPOSED (agent suggestions — await human triage)

Agents: append here with a one-paragraph pitch (what/why/cost). Do not implement. Bernardo moves accepted items into a tier.

Open work only. Shipped exam/A–C items (#47–#63, #14, #52 theme high-ROI, etc.) live in the journal + ROADMAP — not here.

Product / day-2

117. Control Center says "requires rebuild" and leaves the user to do it

Filed as an input to #110 (Retire Control Center safely), not as standalone work — Bernardo 2026-07-14: the Control Center is going away, so fixing its toggles would be building on a condemned surface. Recorded because the observation outlives the tool: it is a fact the #110 migration matrix needs, and whatever menu inherits these settings must not repeat it.

Noticed while fixing #116 (2026-07-14). nomarchy-control-center.sh's Bluetooth, Printing and Updates toggles set the state and print "…​ (requires rebuild)", so the setting only lands whenever the user next thinks to run nomarchy-rebuild. Every menu toggle written since does the rebuild itself and toasts the outcome (nomarchy-autologin, nomarchy-fingerprint, nomarchy-autotimezone) — the state-write-then-rebuild shape. A user who toggles Bluetooth off and sees Bluetooth still running cannot tell "needs a rebuild" from "the toggle is broken again", which is exactly the symptom #116 removed. Migrating these rows into the rofi menu resolves it for free, since that shape is the convention there.

114. Greeter ignores per-device keyboard layouts

Found by Bernardo 2026-07-14: logging out while docked lands on tuigreet, where his external keyboard (remembered as us via settings.keyboard.devices) types the session layout gb — so the password prompt fights him. Not a regression and not docking-related: per-device layouts are applied with hyprctl keyword device[<name>]:kb_layout, which only exists inside a running Hyprland session, while tuigreet draws on a kernel VT whose single keymap comes from console.useXkbConfig ← services.xserver.xkb.layout. A VT structurally cannot do per-device layouts, so this is a design gap, not a bug to patch. Options, cheapest first: (a) document it and stop there; (b) a greeter layout-cycle key (tuigreet has no such feature — would need the keymap swapped under it); (c) host tuigreet inside a small Wayland compositor (cage/labwc), which does get per-device XKB and would let the greeter honour the same in-flake state the session uses — real work, and it changes the greeter's whole rendering path (modules/nixos/greeter.nix themes tuigreet through the 16 ANSI console slots, so (c) is not a drop-in). Worth deciding whether the greeter is meant to be layout-aware at all before costing it.

• NVIDIA first-class options — deferred past v1 (Bernardo 2026-07-10). Keep #59 commented install guidance; no nomarchy.hardware.nvidia.* until a hybrid maintainer + queue.

• Post-install hardware hints (VISION § B) — After the general “you're set” card (#81), optionally fire one additional self-gated notify when the machine actually has the hardware: (a) fwupdmgr on PATH → “System › Firmware to check LVFS updates”; (b) fprintd-list on PATH → “System › Fingerprint to enroll”. One-shot markers in settings.* (same in-checkout discipline as firstBootShown); never a permanent MOTD nag. Cost: small — extend nomarchy-first-boot or a sibling oneshot + checks.first-boot fixture. Control-center / MOTD already mention these; the gap is the silent first graphical session for people who never open those.

(#80–#83 + #85–#88 shipped 2026-07-11. Theme A day-2 + neon-glass finish shipped — VISION ✓. Dock/hibernate V3 → HARDWARE-QUEUE. Parallel fingerprint-or-password shipped 2026-07-12 (Bernardo promoted it live; fingerprint.parallel, pam-fprint-grosshack) — reader V3 → HARDWARE-QUEUE.)

v1.0 pointer

See VISION. Open PROPOSED: post-install hardware hints; NVIDIA deferred past v1; IR portal (b)/(c) need T14s (HARDWARE-QUEUE § T14s). Standing calls: browser = Chromium; power = PPD.

Decisions [human]

Open calls only Bernardo can make; agents add options/evidence but never decide. Resolved entries stay for history; agents treat them as closed.

Resolved (2026-07-10)

• Docs site vs Markdown-in-repo — markdown in-repo for now (docs/, README). A rendered docs site is FUTURE if wanted.
• Default browser — ship Chromium in templates/downstream/home.nix; mime → chromium-browser.desktop. Opt out: delete the line / override mime.
• Default power backend — keep PPD (nomarchy.system.power.backend default). TLP remains the one-line opt-in. Rationale: stability + live profile API for menu/Waybar; Omarchy’s TLP experiment reverted.

Resolved (2026-07-10, more)

• Formatter adoption — yes, but not now. Tracked as FUTURE (below). Nix-source style only (nixfmt-rfc-style or similar); one bulk reformat + CI/check when promoted. Until then: hand-aligned style per CONVENTIONS.

• Hibernation — want by default (product intent). Needs a disk-backed swap (file or partition) sized for resume; not zram alone. Shipped as #76; V3 power-cycle PASSED on TuringMachine 2026-07-12 (ROADMAP § Hibernation + zram by default).

Resolved (2026-07-10, #76 design)

• Swap sizing — exactly RAM (installer default, unchanged). Hibernate image ≤ RAM; zram takes day-to-day paging. swapSize=0 stays no-swap.
• Migration — docs runbook (docs/MIGRATION.md), not a tool.
• No-swap Hibernate — keep the menu row; notify on failure.