Upstream released after the last landing, so the freshness gate went
red: gatelock 0.16.0, crdt_sync_flutter 0.9.1, data_transfer_flutter
0.2.3, github_device_auth 0.5.1, material_ui 1.11.1, sync_settings_ui
0.8.4, gradle 9.9.1. All additive upstream.
Todo: 364208fc
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The vmbox case grepped case-sensitively for "firing", which only matches
"not firing". Inside the 08:00-09:00 fallback fire window --dry-run says
"WOULD FIRE", so the harness went red whenever it ran in that hour.
Match either verdict. Red at guest 08:10 before, green at 06:00, 08:10,
12:00 after; full harness 16/16 at 08:46 UTC.
Todo: 364208fc
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On 2026-11-16 at 09:35 a session ran an install.sh pinning gatelock 0.13.0, which downgraded system python's gatelock and broke diet-guard's lock window (ImportError: watch_credit). One pin across all consumers stops install.sh from downgrading it again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWfb63jncj9Lca2jeX
rtcwake -t shifts the alarm by the RTC/system clock offset (vmbox: +1 s),
so the exact epoch compare in arm()/disarm_ours() forgot our record and
left the alarm armed. Match within 60 s instead (a local-time RTC is off
by hours and still never matches).
_ring_main now emits ring_backend FAILED when pw-play dies mid-ring, not
only when it never starts.
vmbox: the busy-card holder ran as the ssh user (no ACL on a fresh card)
and was aplay, which handles TERM; it now holds the node as root from a
TERM-ignoring sleep and fails the case as a fixture when nothing holds.
Guest repo path derived from the basename, so worktree runs work.
Records the S5 vmbox result; host S5 check dropped by kuhy.
vmbox harness: 16/16 PASS, ring-alsa-busy-fallback 10/10.
Todo: 364208fc
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
crdt_sync_flutter 0.8.2 imports material_ui, so the passphrase dialog
is material_ui's AlertDialog, not the framework one. The test still
asserted the framework type and failed on a clean checkout of
7a3b767. It now checks the type the dialog really is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWfb63jncj9Lca2jeX
md-naming only allows README*, CLAUDE* and DOCS* markdown, so keep.md
failed the all-files gate. Its only reader, check_textgate_keep.sh,
now points at the new name. The textgate tool does not read the file;
it is this repo's own keep list.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWfb63jncj9Lca2jeX
_sync_transport.py hit 280 lines against the 250 cap and AGENTS.md
258; both failed the all-files gate on origin/main. The token lookup
(config file, then a sibling app's file) moves to _sync_token.py
unchanged. Tests now patch SYNC_CONFIG_FILE and
SYNC_TOKEN_FALLBACK_FILES on _sync_token, where they are read.
AGENTS.md is reflowed to 250 lines with every rule kept.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWfb63jncj9Lca2jeX
Since the material_ui bump the app supplied no framework
MaterialLocalizations, so crdt_sync_flutter's E2ePassphraseButton
showDialog crashed and the phone stayed locked. Nothing synced and the
PC fallback rang on 2026-11-16. The app now provides the localization
delegates. Publish failures now reach logcat in release builds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWfb63jncj9Lca2jeX
The phone uploaded nothing from 11-13 to 11-16, and only the morning
fallback ring revealed it. New check_phone_push flags a phone whose
newest HLC is more than 24 h old. An unreachable Firebase counts as
"could not check", so the audit fails toward reporting.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWfb63jncj9Lca2jeX
The home screen prefixed it to every publish() result, including the
full-screen-intent refusal and sync errors.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Motion tokens hand-copied from DOCS-motion.md into theme.dart; every
page route goes through appRoute() (slow, decelerate in, accelerate
out, instant under Remove animations). mediumImpact on the picker
confirm, heavyImpact on a refused time. The Clock app line shows only
when the SET_ALARM handoff finished before its 3 s timeout. Sound is
skipped on purpose.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On 2026-11-11 the ring stood down for a sick day, the mark was cleared
at 11:38, and the 19:31 stick check judged a miss and sent a lockdown
for a day whose alarm never rang. AlarmReceiver now latches the skip
and StickCheck honours it after a later clear.
Docs: the 11-01/10-26 PC no_session rings were correct (no session
started before A+15); the focus-owner on-device check is still open.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The phone pushes alarm.json sealed under wake-alarm-sync/devices-e2e
(Argon2id key from the shared crdt-sync passphrase, keystore + lock
icon in Settings). The PC reads it through open_encrypted_client on
both Firebase and the GitHub mirror (now via GitHubSyncClient); no or
wrong passphrase reads as nothing published, never plaintext. The old
plaintext prefix is copied once per process, never deleted.
crdt-sync 0.11.0, crdt_sync_dart 0.14.1, crdt_sync_flutter 0.8.1,
sync_settings_ui 0.8.2, freedays 0.4.2, sickdays 0.5.3. install.sh
now fails if crdt_sync/pynacl/argon2 do not import.
Todo: fda71ed1
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
PC half: editable install + timers + trigger dry-run health check.
Phone: release APK via phone_deploy, health = installed base.apk
matches the built one. install.sh stays manual (interactive sudo).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The free-day hook ran the system flutter (3.47.9). On a fresh
checkout its pub get rewrote pubspec.lock to its own test_api, so
pre-commit failed with "files were modified". Hooks and
test_changed.sh now put the pinned fvm SDK on PATH.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Flutter 3.50 deprecates package:flutter/material.dart, so the app
moves to material_ui 1.11.0 together with sync_settings_ui 0.8.0
and github_device_auth 0.5.0 (crdt_sync 0.13.1 rides along).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Top-right always-on-top strip while the signed morning session is
in_progress: title, "Next: <step>", "Locks return at HH:MM".
wake-alarm-strip.path starts it on every file rewrite; it re-reads
the file each second, asks for a refresh every 30 s, exits 0 when
the session ends. No display/output: warning, exit 1.
Bump gatelock pin to 0.12.0 (corner_strip).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWZd3smuxDGmEgXkYP8
Conflicting git-tag pins in the shared user site let the last install downgrade focus_project/sickdays (no harvest, no focus locker).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stakes and history move behind two tiles that open bottom sheets;
the page no longer scrolls at Pixel 6a geometry (home_fit_test).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWWcbzoov1KFZJSFcSs
check.sh now runs Dart->Python (keep, unsorted/duplicate input, a cleared
keep day, a keep-only edit that must not re-stamp state) and
Python->Dart (0.2.0 keep sets plus a 0.1.0 record with no keep, a stale
keep and a malformed one), Python 0.2.0 as the oracle with explicit
paths and HOME pointed at a scratch dir. DOCS-sick-days.md and the
SickDays.kt contract describe the alarm-only native cache.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PAhbNszs5JKdtnD4uBi3xp
Tapping today or a later normal day opens a sheet with one switch per
locker (Workout lock, LeetCode, Book, Diet, Wake alarm), all off, and
"Mark sick"; a sick day reopens with its switches, "Save" (keep-only
edit, nothing written when unchanged) and "Clear sick day". Past days
stay inert. A sick day keeping any locker carries a dot in the calendar
and a "Kept: ..." line in the list. DayCell moves to day_cell.dart for
the 250-line cap.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PAhbNszs5JKdtnD4uBi3xp
SickDayPool gains keep: mark/clear write state+actor+keep on one tick,
setKept re-stamps only keep+actor, read rules 1-5 in sick_day_keep.dart.
The native SickDays cache now holds only days on which the alarm stands
down, so a sick day that keeps the alarm rings like a normal day.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PAhbNszs5JKdtnD4uBi3xp
All PC enforcement (fallback ring, RTC self-wake, trigger, workday penalty
write/retract, audit) now asks sickdays.stands_down(LOCKER_ALARM, day) via
_sick.alarm_stands_down instead of is_sick_day, so a sick day that keeps the
alarm behaves exactly as a normal stick day. Bump sickdays pin to v0.2.0.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PAhbNszs5JKdtnD4uBi3xp
The test staged a 2026-10-01 event through stageLocalSettings, which called
stageEnforcementData on the real clock; on 2026-10-16 the event aged past
boundEvents' 14-day window and the test went red for everyone.
stageLocalSettings now forwards an optional `now` to stageEnforcementData
(callers unchanged), and the test passes its fixed clock. phone_app suite:
402 passed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Xfj1sxKH9VBPqpcAni8Jry