run_number counts workflow RUNS, not code. Two failure modes, one of which already bit: it resets to 1 if this file is renamed (the existing NOTE warned about that), and it can fall BEHIND a build produced outside this workflow. On 2026-08-09 diet-guard's phone held a manually sideloaded vc 8 while CI's next build was vc 6, so `adb install -r` refused every CI build with INSTALL_FAILED_VERSION_DOWNGRADE. The only non-destructive escape was re-running the workflow until the counter passed 8 -- forcing a downgrade needs an uninstall, which wipes app data. `git rev-list --count HEAD` only ever increases, tracks the code rather than the CI file, and is already far ahead of every installed build (diet-guard 62 vs 9, screen-locker 118 vs 5, wake-alarm 41 vs 5, todo 51). Needs fetch-depth: 0 -- checkout@v4 is shallow by default and would count 1. Verified the count is identical from the repo root and from app/, so the job's `working-directory: app` default does not affect it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GZH89WEqAjuJT8WZdqc7AP |
||
|---|---|---|
| .github/workflows | ||
| docs/explanations | ||
| phone_app | ||
| scripts | ||
| wake_alarm | ||
| .gitignore | ||
| .pre-commit-config.yaml | ||
| CLAUDE.md | ||
| DESIGN_AUDIT_TODO.md | ||
| install.sh | ||
| MIGRATION.md | ||
| pyproject.toml | ||
| README.md | ||
| requirements.txt | ||
| shutdown-wrapper.sh | ||
| sleep-hook.sh | ||
| wake-alarm-fans.sh | ||
| wake-alarm-trigger.service | ||
| wake-alarm-trigger.timer | ||
| wake-alarm.service | ||
wake_alarm
A phone-driven wake alarm for an always-on PC. The alarm time is whatever the
companion phone app last synced (via the shared crdt_sync transport over a
private GitHub repo); a per-minute systemd user timer fires at exactly that
time, showing a fullscreen, challenge-dismissed alarm with fan ramp and
optional TP-Link Tapo P110 smart-plug control.
Change the alarm on the phone at any hour and the PC follows within ~5
minutes. If the PC does happen to be powered down, a shutdown wrapper
hibernates with rtcwake set to the same synced time, as a fallback.
Install
bash install.sh
This installs the package + dependencies into system Python's user
site-packages (the systemd units run /usr/bin/python directly, not a venv —
see CLAUDE.md), installs and enables the per-minute trigger timer, the
systemd-sleep resume hook, the shutdown wrapper that hibernates with
rtcwake as a fallback, the fan-ramp script, and (optionally) python-kasa
for smart-plug control.
Reading the phone's alarm needs a GitHub token with repo scope, at
~/.config/wake_alarm/sync.json as {"token": "..."}. If that file is
absent, a sibling app's ~/.config/{diet_guard,screen_locker}/sync_token is
used instead, so one token covers every app syncing through kuhyx/syncs.
Usage
python -m wake_alarm._alarm --demo # test the alarm window (safe, closeable)
python -m wake_alarm._sync --show # what alarm time the PC resolves
python -m wake_alarm._trigger --dry-run # what the next tick would do
journalctl --user -fu wake-alarm-trigger.service # watch it live
The alarm fires automatically once installed; no manual invocation is needed in normal operation.
Development
python -m venv .venv && .venv/bin/pip install -r requirements.txt
.venv/bin/pre-commit install && .venv/bin/pre-commit install --hook-type pre-push
.venv/bin/python -m pytest wake_alarm/tests/ --cov=wake_alarm --cov-branch --cov-fail-under=100
See CLAUDE.md for scheduling details, the hibernate/rtcwake mechanism,
the cross-repo workout_log.json read, and production deployment gotchas.