fix(idle): #127 — input wakes the screen compositor-side; revert the clamshell skip

The idle brick was never DPMS. Hyprland ships misc:key_press_enables_dpms
and misc:mouse_move_enables_dpms OFF, so hypridle's on-resume was the ONLY
caller of `dpms on` in the session — an idle daemon with the display held
hostage behind it. hypridle then deadlocks on an unfixed upstream bug
(hyprwm/hypridle#171, "Disconnected from pollfd id 1", open since 2025-09,
absent from 0.1.7 and main) and hangs WITHOUT exiting, so Restart=always
never fires and systemd still reports it active (running). Blank + deadlock
= a black seat no keypress can escape. That was the brick: two mundane bugs,
not one exotic DRM state. 19 disconnects in 11 days on the dev box, incl.
the incident day; the compositor is healthy through all of them.

Both options on, so the wake lives in the compositor where no daemon can
lose it — a dead hypridle now costs auto-lock, not the machine. The #127
clamshell DPMS-off skip (060bf52) is therefore reverted: idle blanks again
in every dock/lid state, and the 27" panels stop being held lit all night
under a static lock screen. #135 (an option to opt back in) is retired
unbuilt — there is nothing left to opt into.

TEST B's first run was invalid and nearly bought the wrong answer: it ran
`hyprctl dispatch dpms off` by hand, which no wake path watches, so "input
did not wake it" was guaranteed on any hardware. The undocked control
returning the SAME result — where the theory demanded a difference — is what
exposed it. Driven through hypridle instead, input wakes it in all three
configurations, including clamshell-docked on the sole live output.

Filed from the diagnosis: #146 (hypridle dies silently, systemd calls it
healthy — the surviving bug) and #147 (a re-login leaves session units dead;
graphical-session.target never re-enters, which is why Waybar already runs
from exec-once).

V3 on the incident hardware (AMD dev box, clamshell): hypridle deliberately
stopped, `dpms off` -> keypress woke it in 6s; the identical test before the
options stayed black the full 45s. V1: nix flake check, checks.option-docs,
checks.clamshell-logind, checks.docking-ux, checks.display-profiles; read the
emitted hyprland.conf/hypridle.conf rather than the source.

V3 pending: the fix is proven via runtime `hyprctl keyword`, not yet from our
own Nix — HARDWARE-QUEUE re-checks it after nomarchy-home + relogin.

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

README.md

@@ -258,7 +258,7 @@ two tables below are split along exactly that line.
 | `nomarchy.batteryNotify.enable` | `true` | Low-battery toasts at the bar's thresholds — 25% low, 10% critical (stays up until dismissed); silent no-op on machines without a battery |
 | `nomarchy.dockAudio.enable` | `true` | On a fresh external-monitor plug, PipeWire/WirePlumber are reprobed after hardware settles and the highest-priority available HDMI/DP/USB sink becomes default, with a toast and journal result. A later manual speaker choice sticks until the next physical plug; unplug falls back to built-in. Manual route: System › Audio |
 | `nomarchy.firstBootWelcome.enable` | `true` | One dismissible “you're set” toast on the first session (SUPER+M / SUPER+T / SUPER+? + network pointer); marker is `settings.firstBootShown` in the flake checkout |
-| `nomarchy.idle.enable` | `true` | hyprlock + hypridle (lock 5 min; display off 10 min **except when docked** — laptop internal present but disabled skips DPMS, #127; suspend 15 min battery-only via `nomarchy-suspend`) |
+| `nomarchy.idle.enable` | `true` | hyprlock + hypridle (lock 5 min; display off 10 min in every dock/lid state — input wakes it compositor-side, #127; suspend 15 min battery-only via `nomarchy-suspend`) |
 | `nomarchy.idle.fingerprint` | `false` | Unlock the lock screen with a fingerprint as well as the password, and say so on the input field. Set it alongside `nomarchy.hardware.fingerprint.pam` in system.nix: hyprlock is configured from Home Manager, which cannot read the NixOS option — and hyprlock does **not** take a fingerprint through PAM at all (its PAM stack runs only on submit), so it uses its own fprintd backend that this switch turns on |
 | `nomarchy.yazi.enable` | `true` | yazi TUI file manager, themed + curated plugins |
 | `nomarchy.osd.enable` | `true` | swayosd on-screen display for volume/brightness/mute |

agent/BACKLOG.md

@@ -31,163 +31,8 @@ post-boot sessions). Preserve separation: the installer bake failure and the
 flake pin are different root causes even when they show up on the same machine.
 (Terminal / Ghostty-on-Acer → shipped as Kitty-only, #95.)
 
-### 127. `[blocked:hw]` Docked idle: black screens, no wake, undock does not restore panel
-
-Bernardo 2026-07-15 (AMD dev box, docked): left the laptop docked; after a
-few minutes of idle it appeared to "sleep". Then:
-
-1. External keyboard/mouse did **not** wake any display.
-2. Unplugging the external monitor did **not** turn the laptop panel back
-   on (dock mode has eDP disabled — undock *should* re-enable it).
-3. The machine was still alive: Caps Lock LED toggled on the keyboard.
-
-This also killed a long agent session (looked like a "crash" from the
-outside). Separate from awake undock recovery (HARDWARE-QUEUE round 8):
-the failure is **idle while docked**, then no path back to a usable panel.
-
-**Likely shape (code-side, unconfirmed on the incident):** on AC, hypridle
-does **not** call `systemctl suspend` (`onAc || suspend` — only battery
-suspends at 15 min). Idle path is lock @5 min + `dpms off` @10 min
-(`modules/home/idle.nix`). Caps Lock working fits **DPMS/lock blackout**
-better than deep S3. Dock mode has already disabled eDP; recovery then
-needs either (a) input → hypridle `on-resume` / `dpms on` on the external,
-or (b) undock → `nomarchy-display-transition undock` re-enabling the
-panel. Both failed. `after_sleep_cmd` only does `hyprctl dispatch dpms on`
-and never re-enables a disabled internal.
-
-**Progress 2026-07-15:**
-- Recovery tooling: undock ends with `dpms on`; `nomarchy-display-wake` /
-  `nomarchy-display-dump` (SUPER+SHIFT+D; TTY if it works).
-- **Mitigation (shipped):** do **not** DPMS-off when **clamshell** — if any
-  laptop internal (eDP/LVDS/DSI) exists in `monitors all` but is not
-  enabled, the 10 min listener skips blanking. Lock at 5 min still runs;
-  undocked laptops still DPMS-off as before. Bernardo: Ctrl+Alt+F3 did
-  not yield a usable TTY on the original brick, so prevention > dump path.
-- **Scope settled (#136, 2026-07-16):** the gate is *clamshell*, not
-  "docked" — and Bernardo confirms the incident was exactly that, **laptop
-  closed**. So the code matched the evidence and only the wording
-  overclaimed; that is fixed (idle.nix, ROADMAP, and the journal line, now
-  `dpms-off skipped: clamshell`). Docked with the lid **open** still blanks
-  and is untested — not exempted, not claimed safe.
-- **Revisit later (LATER):** intentional DPMS-when-clamshell once wake is
-  proven trustworthy (monitor power-save without a brick). Tracked as
-  **#135** (state key + the surface to delete this by), downstream of the
-  hardware test below.
-- **Caveat the test must settle:** the mitigation is a cure for an
-  **unconfirmed cause**. "Blanking the only live output caused the brick"
-  is inference from one incident, shipped at V1. If DPMS-off is not the
-  trigger, we are paying a permanent behaviour cost (docked externals never
-  sleep) and the brick is still live — so confirming the cause outranks
-  #135/#136.
-- **Symptom 2 is SOLVED, and it was not DPMS (ff5017e, 2026-07-16).** Bernardo
-  reproduced "unplugging did not turn the laptop panel back on" live, minutes
-  after TEST A — with **no blanking involved at all** (the mitigation had kept
-  the external lit). Cause: `awaiting_lid_open` is set by every successful
-  undock and cleared only when the external is gone **and** the lid is open, so
-  a clamshell undock **latches** it; a re-dock clears its sibling
-  `undock_pending` but not the latch, and the next departure is then silently
-  dropped by the `[ -z "$awaiting_lid_open" ]` guard. His journal shows the
-  first undock logging *nothing*, a release the instant he lifted the lid, then
-  a clean undock. Replaying the sequence with the lid never opened: 1 of 3
-  undocks ran; fixed, 3 of 3. **What this does to the theory:** the original
-  brick may simply have been an ordinary blank plus a recovery that could not
-  run — two mundane bugs, not one exotic DRM state. TEST B is now the only
-  thing standing between that hypothesis and the truth, and it is a much safer
-  test with the latch fixed (an undock is a real escape hatch again).
-
-**Pass (current mitigation):** clamshell-docked on AC, idle past 10+ min →
-external stays lit under hyprlock (journal: `dpms-off skipped: clamshell`);
-undocked still blanks at 10 min and wakes on input. Full original pass (wake
-after DPMS-while-clamshell) deferred to the LATER revisit.
-
----
-
-### The two tests, exactly (2026-07-16)
-
-**There is no SSH lifeline on this box** (`sshd` inactive), which is why test B
-self-recovers instead of relying on one. If you would rather have the net, turn
-`services.openssh.enable = true` on for the day and keep a phone SSH client
-open — then a brick is a nuisance, not a reboot.
-
-**Test A — does the mitigation hold? (safe, ~15 min, unattended)**
-1. Docked, **lid closed**, on AC. Leave it completely alone for 12 minutes.
-2. Expect: at 5 min the external shows hyprlock; at 10 min it **stays lit**.
-3. Then: `journalctl --user -b -t nomarchy-idle --since '-15 min'`
-   → one `dpms-off skipped: clamshell (internal eDP-1 present but off)`.
-4. Now open the lid / undock, idle 11 min → the panel **should** blank, and
-   wake on a keypress. That half must still work; the mitigation is not
-   supposed to have disabled blanking everywhere.
-
-**Test B — was DPMS-off actually the trigger? (the decisive one; bounded)**
-This is the test that outranks everything else here, because the mitigation is
-a cure for an *inferred* cause. Docked, **lid closed**, from a terminal on the
-external:
-
-```sh
-( sleep 45
-  hyprctl dispatch dpms on
-  hyprctl reload            # re-enables every output (the #142 hammer)
-) >/tmp/dpms-test.log 2>&1 &
-hyprctl dispatch dpms off
-```
-
-The screen goes black. Then, in order:
-1. **Press keys / move the mouse** in the first ~40s. Does it come back?
-   (In the incident it did not — this is the exact step that failed.)
-2. **Wait for the 45s auto-recovery.** Does the external return by itself?
-3. If it is still dark, **open the lid** — the `reload` should have re-enabled
-   eDP even if the external stayed dead.
-4. Afterwards:
-   `journalctl --user -b -t nomarchy-idle -t nomarchy-display-watch --since '-5 min'`
-
-**What each outcome means — all three are useful:**
-- **Input wakes it** → DPMS-off is **not** the trigger. The mitigation is a
-  permanent behaviour cost buying nothing, the brick is still live and
-  un-diagnosed → revert the skip, **#135 never exists**, and hunt the real
-  cause (hyprlock? hypridle's own resume path? the dock's DP link?).
-- **Dark until the 45s recovery, then fine** → the compositor was alive and
-  `dpms on` works, so blanking-while-clamshell *is* the trigger and the
-  mitigation is right → **#135 becomes the way to opt back in**.
-- **Dark even after recovery** (lid needed) → the original brick, reproduced
-  **deterministically and on demand**. That is the best outcome: it turns a
-  once-off incident into something bisectable (try it without hyprlock, with
-  the lid open, on the internal only …).
-
-Note the deliberate omission: this runs `dpms off` **by hand** rather than
-waiting for hypridle, precisely because the shipped mitigation would otherwise
-skip it. Same command, same state, no rebuild.
-
-**Fallback if the cause resists (Bernardo 2026-07-16):** diagnosing stays the
-preference — DPMS when docked is the outcome we want. But the mitigation as
-shipped holds a near-static lock screen on the externals indefinitely, so if
-the cause cannot be pinned down, a **screensaver while docked** is the
-consolation prize: it protects the panels (burn-in — the same cost #135
-names) without re-entering the blanking path that bricked the session. Not a
-new item: it is the "if diagnosis fails" branch of this one, and #135's state
-key is where it would land.
-
 ## NEXT
 
-### 135. Give the #127 DPMS mitigation a state key (and a way back)
-
-The shipped mitigation is hardcoded policy: docked (clamshell) externals now
-**never** blank, and hyprlock has no timeout of its own, so they stay lit
-indefinitely. Two real costs land on the user — a couple of 27" panels at
-~60–100W all night, and **OLED burn-in** from a near-static lock screen held
-for hours. Today there is no way to opt back in, which also breaks the
-in-flake-state rule that settings are menu-writable.
-
-Add `settings.idle.dpmsWhenDocked` (default **false** = today's safe
-behaviour), read by `modules/home/idle.nix`'s `dpmsOff` gate, with a
-Preferences row. The real point is not the option: it gives the mitigation a
-**named surface to delete** once wake is proven, instead of an unowned LATER
-note that quietly becomes the product. Whoever removes the mitigation removes
-the key in the same commit.
-
-**Order:** after the #127 hardware test. If DPMS-off turns out not to be the
-trigger, the mitigation is reverted wholesale and this key never exists —
-building it first risks shipping an option for a policy we withdraw.
-
 ### 143. Rofi keeps the selected *row number* when the search changes, so the highlight lands on an unrelated entry
 
 Bernardo 2026-07-16, and his repro is exact: open the app menu, type `ca`, move
@@ -246,14 +91,6 @@ the next lock bump, that is fine — nothing here rots meanwhile.
 
 ## LATER
 
-- **DPMS when docked (revisit #127):** today we **skip** display blanking
-  while any laptop internal is present-but-disabled (dock/clamshell), so
-  idle only locks. That avoids blanking the sole external and the brick
-  where wake/TTY failed (2026-07-15). Revisit when we want monitor
-  power-save on dock again: re-enable DPMS-off for that mode only if
-  `nomarchy-display-wake` (or better) reliably restores the external and
-  undock-while-black restores eDP on the AMD dock setup — without
-  depending on SSH. Until then, leave the skip in `modules/home/idle.nix`.
 - **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
@@ -299,6 +136,64 @@ high-ROI, etc.) live in the journal + ROADMAP — not here.*
 
 ### Product / day-2
 
+### 146. hypridle dies silently and systemd calls it healthy
+
+Fell out of the #127 diagnosis (2026-07-16, ROADMAP § "#127 solved"), and it is
+the bug that was actually behind the brick — the DPMS gate we just reverted was
+only ever a symptom shield. hypridle hits an unfixed upstream deadlock
+(hyprwm/hypridle#171, "Disconnected from pollfd id 1" — open since 2025-09, in
+neither 0.1.7 nor main; upstream's own reporter saw 10-hour hangs) and then
+**hangs without exiting**, so `Restart=always` never fires and `systemctl
+--user is-active` still says `active (running)`. On the dev box: **19 times in
+11 days**, including the incident day. Jul 14 is the clean example — locking
+normally at 13:02, `CRITICAL` at 13:08:17, then eleven hours of silence, with
+D-Bus receive queues visibly backing up (48 KB unread) because nothing was
+reading them. The compositor is *fine* through all of it (no other session unit
+dies in the same second), so this is not the compositor's death taking it down.
+
+Now that input wakes the screen compositor-side this is no longer brick-class,
+but it is not cosmetic either: a silently dead hypridle means **no auto-lock**
+(Bernardo sat unlocked for 25 min on 2026-07-16 without knowing), no blank, no
+idle suspend. Cost: systemd can't see it — `WatchdogSec` needs `sd_notify`
+hypridle doesn't implement, and the process never exits — so detection has to be
+external: a guard watching for the `CRITICAL` line, or a liveness probe on its
+Wayland fd, then `systemctl --user restart hypridle`. **Sequence it after #147**:
+today's disconnect was the compositor dying, and restarting hypridle into a dead
+compositor just fail-loops into `start-limit-hit` exactly as cliphist does. The
+trigger is still unknown — *not* dock/undock (no display-watch activity in the
+3 min before any of them) and *not* a D-Bus quota disconnect (no broker errors).
+Worth a reproducer + an upstream comment on #171: it has sat ten months on
+vaxerski's "that will cause a segfault no?", and we have better data than the
+reporter did.
+
+### 147. A re-login leaves the session's user units dead, and nothing brings them back
+
+Bernardo, 2026-07-16: after re-logging in, Waybar took 5+ seconds and
+nomarchy-doctor reported cliphist, network-manager-applet and udiskie failing.
+Root cause is the session lifecycle, not those apps. Hyprland exited at
+17:25:39 and the replacement came up at **17:25:52** — a 13-second gap — but
+`graphical-session.target` never re-entered (its `ActiveEnterTimestamp` still
+read 08:03:32, from the *morning's* session). So: units with a restart policy
+burn their start limit inside that gap against a Wayland socket that does not
+exist yet (cliphist's journal is explicit — `wl-paste: Failed to connect to a
+Wayland server`, 5 restarts in one second, `start-limit-hit` at 17:25:40,
+twelve seconds before the new compositor existed); units without one
+(nm-applet, udiskie) simply die once and stay dead forever. Nothing re-enters
+the target, so nothing recovers.
+
+We have already paid for this once without naming it: `modules/home/hyprland.nix`
+launches Waybar from `exec-once` rather than a unit, and the comment there says
+why — "bound to graphical-session.target the unit raced Hyprland's IPC on
+relogin". That is this bug, worked around for one service; the 5-second Waybar
+start is the workaround being slow instead of failing. Fix is the standard
+wiring the repo never adopted: the compositor imports `WAYLAND_DISPLAY` +
+`HYPRLAND_INSTANCE_SIGNATURE` into the user manager and starts/stops a
+`hyprland-session.target` on entry/exit (Home Manager's
+`wayland.windowManager.hyprland.systemd.enable`, or uwsm). Cost: touches how
+every session service is bound, so it wants a VM pass and a real relogin on
+hardware — but it would let Waybar go back to being a unit and unblock #144's
+guard.
+
 ### 120. A netinstall ISO, next to the fat offline one
 
 **Deferred to PROPOSED 2026-07-16 (Bernardo): not now.** Nothing below is

agent/HARDWARE-QUEUE.md

@@ -16,7 +16,7 @@ Run these first when you have a laptop session (AMD dev box unless noted):
 | # | Item | Why first |
 |---|------|-----------|
 | 1 | **#115** suspend-then-hibernate | Just shipped; quick path is minutes (battery vs AC + `nomarchy-suspend`) |
-| 2 | **#127** docked idle blackout | Real incident; capture path before it bit-rots |
+| 2 | **#127** fix from config (not runtime keyword) | Cause found + fixed 2026-07-16; confirm our Nix emits it and it survives a relogin |
 | 3 | **Docking recovery round 8** + headphone jack-follow | Same dock session; relogin once, then undock ×5 + jack |
 | 4 | **#104** airplane mode | Fast radio smoke after `nomarchy-home` |
 | 5 | **#101** charge-limit USB-C burst | Dock plug/unplug storm; pairs with #3 |
@@ -49,36 +49,36 @@ Everything else below stays open; order is convenience, not a gate.
       restores **only** the radios that were on before. If Wi-Fi was already
       off, it must stay off after disengage. Glyph must **not** appear when
       airplane is off.
-- [ ] **#127 clamshell idle — TEST A, the mitigation holds (AMD dev box)** —
+- [x] **#127 TESTS A + B — ANSWERED 2026-07-16 (AMD dev box), both closed.**
-      **Context:** original brick = clamshell-docked idle → black, no input
+      TEST B ran and DPMS-off is **not** the trigger: with hypridle driving the
-      wake, undock no eDP, Caps Lock alive; Ctrl+Alt+F3 gave no usable TTY.
+      blank, input woke it in all three configurations including clamshell-docked
-      **Shipped mitigation:** skip DPMS-off while clamshell (internal
+      on the sole live output. The mitigation was reverted and #135 retired
-      present-but-disabled); lock at 5m still.
+      unbuilt, which also voids TEST A (its premise — the skip — no longer
-      **After `nomarchy-home` + relogin:**
+      exists). Real cause: Hyprland's input→DPMS options ship off, so hypridle's
-      1. Dock, **lid closed** (eDP in `hyprctl monitors all -j`, absent from
+      `on-resume` was the only caller of `dpms on`, and hypridle deadlocks
-         `hyprctl monitors`).
+      (hyprwm/hypridle#171) without exiting. Full record: ROADMAP § "#127
-      2. Leave it entirely alone past 10+ min.
+      solved"; the surviving bug is BACKLOG #146. **Method note worth keeping:**
-      3. **Pass:** external still shows hyprlock (not black power-save);
+      the *first* TEST B was invalid — it ran `dpms off` by hand, which no wake
-         `journalctl --user -t nomarchy-idle --since '-20 min'` has
+      path watches, so its "input did not wake it" was guaranteed regardless of
-         `dpms-off skipped: clamshell`.
+      hardware. Drive the real code path or measure nothing.
-      4. Undocked control: eDP only → at 10 min it *does* blank; input wakes.
+- [ ] **#127 fix from config, not a runtime keyword (AMD dev box)** —
-      Optional: `nomarchy-display-dump` / SUPER+SHIFT+D for other display faults.
+      the fix was proven with `hyprctl keyword` (runtime-only); this checks our
-- [ ] **#127 TEST B — was DPMS-off the trigger at all? (AMD dev box)** —
+      Nix actually emits it and that it survives a session. **After
-      **This one outranks #135**: the mitigation is a cure for an *inferred*
+      `nomarchy-home` + relogin:**
-      cause, and if the inference is wrong we are paying for it forever while
+      1. `hyprctl getoption misc:key_press_enables_dpms` and
-      the brick stays live. Bounded and self-recovering (there is no sshd on
+         `misc:mouse_move_enables_dpms` → both `int: 1` with **no** manual
-      that box). Full protocol + how to read each outcome: **BACKLOG #127 §
+         keyword set.
-      "The two tests, exactly"**. In short — docked, lid closed:
+      2. Docked, **lid closed**, idle 10+ min → the external **blanks** (it no
-      ```sh
+         longer stays lit; that is the revert working).
-      ( sleep 45; hyprctl dispatch dpms on; hyprctl reload ) &
+      3. Press a key → it comes back. `journalctl --user -t
-      hyprctl dispatch dpms off
+         nomarchy-display-wake --since '-15 min'` shows a `done enabled=…` line
-      ```
+         (hypridle's on-resume also ran).
-      then (1) try to wake it with input in the first ~40s, (2) let the 45s
+      4. The one that matters — the brick condition on purpose:
-      recovery run, (3) open the lid if still dark. **Input wakes it** → the
+         `systemctl --user stop hypridle`, then
-      mitigation is pointless, revert it and #135 never exists. **Dark until
+         `( sleep 45; hyprctl dispatch dpms on ) &` and
-      recovery** → mitigation confirmed, #135 is the opt-in. **Dark after
+         `hyprctl dispatch dpms off`. **Pass:** a keypress wakes it with
-      recovery** → the brick, now reproducible on demand, which is the best
+         hypridle stopped (it did in 6s on 2026-07-16 via runtime keyword).
-      outcome available.
+         Then `systemctl --user start hypridle`.
 - [ ] **#96 battery-limit row on threshold-less firmware (Acer M5-481T)** —
       on the Acer (live or installed session with the commit carrying this
       entry): open System (`SUPER+CTRL+I`). **Pass:** a "Battery limit" row

agent/JOURNAL.md

@@ -19,6 +19,39 @@ Template:
 
 ---
 
+## 2026-07-16 — #127 SOLVED: the brick was never DPMS, and TEST B's first run lied
+- **Task:** #127 TEST B (Bernardo ran it), then the whole sweep his result forced.
+- **Read this if nothing else — the test measured nothing, and said so.** His run
+  reported "input never wakes it" **docked *and* undocked alike**, and that tie is
+  the tell: the theory demanded those differ. It ran `dpms off` **by hand**, which
+  no wake path watches, so "input didn't wake it" was guaranteed on any hardware
+  in any dock state. Re-run *through* hypridle so `on-resume` is actually armed:
+  input woke it in all 3 configs, **including clamshell on the sole live output**
+  (`done enabled=1 first=DP-2`) — the incident's exact geometry. Put the mechanism
+  under test inside the circuit, or you are measuring your own scaffolding.
+- **Real cause:** Hyprland ships `misc:{key_press,mouse_move}_enables_dpms` **off**,
+  so hypridle's `on-resume` was the ONLY caller of `dpms on` anywhere in the session
+  — and hypridle deadlocks (hyprwm/hypridle#171, open since 2025-09, in neither
+  0.1.7 nor main) **without exiting**, so `Restart=always` never fires and systemd
+  still reports it active. Blank + deadlock = a black seat no key escapes. **19× in
+  11 days** on his box, incl. the incident day. Two mundane bugs, not one exotic one.
+- **Did:** both options on — structural: the wake moves into the compositor where no
+  daemon can lose it, so a dead hypridle costs auto-lock, not the machine. Reverted
+  the clamshell skip; **#135 retired unbuilt**; ROADMAP § "#127 solved"; filed #146
+  (hypridle mortality) + #147 (relogin leaves session units dead) → PROPOSED.
+- **Verified: V3 on the incident hardware.** Clamshell, hypridle **deliberately
+  stopped** (brick condition on purpose), `dpms off` → keypress woke it in **6s**;
+  the identical test before the options stayed black the full 45s. V1: `flake check`
+  + `option-docs`/`clamshell-logind`/`docking-ux`/`display-profiles` pass, and I read
+  the **emitted** configs, not the source (`key_press_enables_dpms=true`;
+  `on-timeout=hyprctl dispatch dpms off`, skip script gone).
+- **Pending:** V3 — proven via runtime `hyprctl keyword`, NOT from our Nix yet;
+  HARDWARE-QUEUE entry re-checks after `nomarchy-home` + relogin. Ctrl+Alt+F3 dead
+  during the incident is still unexplained (upstream co-symptom, hypridle#145).
+- **Next suggestion:** #147 before #146 — a guard that restarts hypridle into a dead
+  compositor just fail-loops into `start-limit-hit` exactly like cliphist does. Both
+  are PROPOSED; Bernardo triages.
+
 ## 2026-07-16 — #129/#130 buttons; #133; #132; branding; and a latch bug from TEST A
 - **Task:** Bernardo triaged the queue (#136 settled, #130 decided, #143 → wait,
   #120 → PROPOSED), then #133, #132, the codename drop, and #129+#130.

docs/ROADMAP.md

@@ -622,6 +622,10 @@ Design/decision records and a running log of shipped work (items marked
   that could not run — two mundane bugs, not one exotic DRM state. The shipped
   DPMS mitigation may therefore be treating the wrong cause; #127's TEST B is
   what decides, and it is far safer now that an undock works again.
+  **It did decide, the same day: the hunch was right on both counts** — the
+  mitigation *was* treating the wrong cause and is now reverted, and the brick
+  *was* two mundane bugs (an ordinary blank + a wake path that had died). See
+  "#127 solved" below.
 - ✓ **The #107 compat shim is pinned (#133, 2026-07-16):** `mkFlake` prefers
   `state.json` and still accepts the legacy `theme-state.json`, because a
   machine only migrates on its next menu write — but nothing evaluated a
@@ -723,19 +727,44 @@ Design/decision records and a running log of shipped work (items marked
   `nomarchy-suspend` (hypridle + Power menu). LUKS: unlock session before
   encrypted hibernate phase so resume is LUKS-only (not hyprlock twice).
   Menu row self-gates on CanHibernate. V3: HARDWARE-QUEUE bag-carry test.
-- **Clamshell idle DPMS (#127 mitigation):** after an idle brick (alive
+- ✓ **#127 solved — the idle brick was never DPMS (2026-07-16, AMD dev box):**
-  machine, black external, undock no eDP, no usable Ctrl+Alt+F3), hypridle
+  the clamshell DPMS-off skip (060bf52, and its #136 wording fix) is
-  **skips DPMS-off** when any laptop internal is present but not enabled.
+  **reverted**; idle blanks again in every dock/lid state. Hyprland ships
-  Lock at 5 min unchanged; undocked still blanks at 10 min.
+  `misc:key_press_enables_dpms` and `misc:mouse_move_enables_dpms` **off**, and
-  **Say clamshell, not "docked" (#136, settled 2026-07-16):** the gate is
+  nothing else in the session calls `dpms on` — so hypridle's `on-resume` was
-  "an internal exists and is switched off" — the lid shut — and Bernardo
+  the *only* thing that could un-blank a screen: an idle daemon with the
-  confirms the incident was exactly that, laptop closed. Docked with the lid
+  display held hostage behind it. hypridle then deadlocks on an unfixed
-  **open** has every output enabled, falls straight through, and still blanks;
+  upstream bug (hyprwm/hypridle#171, "Disconnected from pollfd id 1", open
-  whether *that* can brick is untested and not claimed either way. The earlier
+  since 2025-09, absent from 0.1.7 *and* main) and hangs **without exiting**,
-  wording ("skip when docked") promised twice the coverage the code has, which
+  so `Restart=always` never fires and systemd still reports it
-  is how a recurrence gets misread as a regression.
+  `active (running)`. Blank + deadlock = a black seat no keypress can escape.
-  Intentional DPMS-when-clamshell is LATER (BACKLOG) once wake is proven.
+  That is the brick: two mundane bugs, not one exotic DRM state — the same
-  Tooling kept: `nomarchy-display-wake` / `nomarchy-display-dump`.
+  shape as the #100 latch above.
+  **Evidence, on the incident hardware.** With hypridle driving the blank,
+  input woke it in all three configurations — including clamshell-docked on the
+  sole live output (`nomarchy-display-wake done enabled=1 first=DP-2`), which
+  is the incident's exact geometry. So blanking-while-clamshell is innocent and
+  the gate guarded nothing. Then the decisive run: same box, still clamshell,
+  hypridle **deliberately stopped** (i.e. the brick condition manufactured on
+  purpose), `dpms off` → a keypress woke it in **6 s** with the two options on.
+  The identical test before them stayed black the full 45 s.
+  **Why the first TEST B said the opposite, and the lesson.** It ran `hyprctl
+  dispatch dpms off` **by hand**, which no wake path watches, so "input did not
+  wake it" was guaranteed on any hardware, in any dock state. The tell was the
+  undocked control returning the *same* result where the theory demanded a
+  different one. A test must put the mechanism under test inside the circuit —
+  this one measured a wake path that was never armed, and nearly bought a
+  permanent behaviour cost with the answer.
+  **Why the fix is structural.** The wake now lives in the compositor, where no
+  daemon can lose it; a dead hypridle costs auto-lock, not the machine. #135
+  (an option to opt back into DPMS-when-clamshell) is therefore **retired
+  unbuilt** — there is nothing left to opt into, and the 27" panels stop being
+  held lit all night under a static lock screen.
+  **Still open:** hypridle's silent death is its own bug (~2×/week on this box;
+  no auto-lock is a security cost even when the screen is recoverable), and the
+  incident's dead Ctrl+Alt+F3 remains unexplained — a known co-symptom upstream
+  (hyprwm/hypridle#145) but not accounted for here. Tooling kept:
+  `nomarchy-display-wake` / `nomarchy-display-dump`.
 - ✓ **Generation prune (#128):** weekly `nomarchy-gen-prune` (system + HM):
   delete only gens older than **14 days** that are also beyond the **3 most
   recent past** gens (current always kept). Stock `nix.gc` no longer uses

modules/home/hyprland.nix

@@ -1097,6 +1097,20 @@ in
         # instead of compositor black — on light themes the black flash
         # reads as a glitch (item 28c, seen in the capture harness).
         background_color = rgb c.base;
+
+        # #127, and the whole answer to it. Hyprland ships both of these
+        # OFF, so nothing in the compositor turned a blanked screen back
+        # on: hypridle's on-resume was the ONLY caller of `dpms on` in the
+        # session — an idle daemon as a single point of failure with the
+        # display behind it. hypridle then deadlocks on an unfixed upstream
+        # bug (hyprwm/hypridle#171: "Disconnected from pollfd id 1", open,
+        # not in 0.1.7 or main) and hangs WITHOUT exiting, so Restart=always
+        # never fires and systemd still reports it active. Blank + deadlock
+        # = a black seat no keypress could escape. That was the brick.
+        # These put the wake in the compositor, where no daemon can lose it;
+        # a dead hypridle now costs auto-lock, not the machine.
+        key_press_enables_dpms = lib.mkDefault true;
+        mouse_move_enables_dpms = lib.mkDefault true;
       };
 
       # First boot of every installed image was showing "Hyprland updated to

modules/home/idle.nix

@@ -51,34 +51,6 @@ let
       inherit pkgs;
       displayTransition = miniTransition;
     }).displayWakeTool;
-
-  # #127 mitigation (2026-07-15): blanking the *only* live output with the
-  # lid shut (eDP disabled) left a black seat that sometimes would not
-  # recover (and even VT switch looked dead). Skip DPMS-off when any laptop
-  # internal is present but not enabled — i.e. CLAMSHELL, which is narrower
-  # than "docked" and is deliberately what the incident was (#136, Bernardo:
-  # laptop closed). Docked with the lid OPEN keeps every output enabled and
-  # still blanks; that case is untested, not exempted. Lock still runs at 5 min.
-  # Revisit: intentional DPMS-on-dock once wake is trustworthy — see
-  # BACKLOG #127 and LATER "DPMS when docked".
-  dpmsOff = pkgs.writeShellScript "nomarchy-dpms-off" ''
-    set -euo pipefail
-    jq=${pkgs.jq}/bin/jq
-    internals=$(hyprctl monitors all -j 2>/dev/null \
-      | $jq -r '.[] | select(.name | test("^(eDP|LVDS|DSI)")) | .name' \
-      2>/dev/null || true)
-    if [ -n "$internals" ]; then
-      for m in $internals; do
-        if ! hyprctl monitors -j 2>/dev/null \
-              | $jq -e --arg m "$m" 'any(.[]; .name == $m)' >/dev/null 2>&1; then
-          ${pkgs.util-linux}/bin/logger -t nomarchy-idle -- \
-            "dpms-off skipped: clamshell (internal $m present but off)"
-          exit 0
-        fi
-      done
-    fi
-    hyprctl dispatch dpms off
-  '';
 in
 {
   config = lib.mkIf cfg.idle.enable {
@@ -182,8 +154,14 @@ in
           { timeout = 300; on-timeout = "loginctl lock-session"; }
           {
             timeout = 600;
-            # Docked: skip blanking sole output (#127). Undocked: DPMS off.
+            # Blanks in every dock/lid state — the #127 clamshell skip is
-            on-timeout = "${dpmsOff}";
+            # gone: it was a cure for a cause that does not exist (proven on
+            # hardware 2026-07-16, see ROADMAP). The wake no longer depends
+            # on this daemon surviving — misc:{key_press,mouse_move}_enables_dpms
+            # in hyprland.nix means input wakes the screen compositor-side.
+            on-timeout = "hyprctl dispatch dpms off";
+            # Same rescue path as after_sleep (#127): re-enables a disabled
+            # internal, which plain `dpms on` cannot do.
             on-resume = "${lib.getExe displayWake}";
           }
           # Suspend only on battery, and sooner than the old fixed 30 min