Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
35 commits
Select commit Hold shift + click to select a range
ee8d9f0
l10n(es): translate the supplement catalogue, so 300 strings stop fal…
Nov 15, 2026
4db0a9a
l10n(fr): translate the supplement catalogue, so 300 strings stop fal…
Nov 15, 2026
8e9650d
l10n(zh): translate the supplement catalogue, so 300 strings stop fal…
Nov 15, 2026
b03e71b
l10n(hi): translate the supplement catalogue, so 300 strings stop fal…
Nov 15, 2026
faabe1f
l10n: translate three strings es, fr and th still held in English, an…
Nov 15, 2026
41cbce2
l10n(es): singular "1 noche"/"1 día", pool "largos", and a missed dos…
Nov 15, 2026
ce8d0a1
l10n(fr): singular "1 nuit"/"1 jour", and pool "longueurs" in the swi…
Nov 15, 2026
eaf9414
l10n: translate the current HRV, HRV-stability and breathing-rate met…
Nov 15, 2026
71fb04d
test(l10n): check every catalogue on disk for English copies, not a f…
Nov 15, 2026
aef9d4d
l10n: translate the ChatGPT sign-in flow and an alarm status, which s…
Nov 15, 2026
52c2a59
Revert the Spanish catalogue commit that landed on this branch by mis…
Nov 15, 2026
eda15f4
Revert the method-text commit that landed on this branch by mistake
Nov 15, 2026
ce544bf
l10n: translate the current HRV, HRV-stability and breathing-rate met…
Nov 15, 2026
3271d48
test: the literal scan also reads calls whose first argument is AppLo…
Nov 15, 2026
b2de4b6
l10n(es): the dose reminder says a dose is due, not to tap one
Nov 15, 2026
4ac24ed
context model: feed the overnight detectors residuals (actual - expec…
mistral-vibe Nov 15, 2026
94c85e3
fix analyzer findings from CI: spread-of-nullable tag merge, generic …
mistral-vibe Nov 15, 2026
9a09c55
fix two more analyzer errors: null-aware index ?[] is not valid here,…
mistral-vibe Nov 15, 2026
c73f6a8
freeze golden: strip only the context model's additive keys from the …
mistral-vibe Nov 15, 2026
ac7a6b7
l10n: add journalTagVaccinated to all seven non-English locales (each…
mistral-vibe Nov 15, 2026
8afab21
fix fixture assertions against the detectors' real scales: illness re…
mistral-vibe Nov 15, 2026
84b9814
fix last two fixture bugs: travel test index out of range (n=42), and…
mistral-vibe Nov 15, 2026
9ea00d7
sleep plan, weekend shift: print durations in the display language
Nov 15, 2026
e326fd4
fr: say '1 longueur' for a single pool length
abdulsaheel Nov 15, 2026
c0b161d
zh: check the supplement counts at a second value
abdulsaheel Nov 15, 2026
6f62ad6
Merge pull request #726 from kurexw/l10n/chatgpt-signin
abdulsaheel Nov 15, 2026
ad9968f
Merge pull request #725 from kurexw/l10n/metric-method-copy
abdulsaheel Nov 15, 2026
86011d9
Merge pull request #724 from kurexw/l10n/english-copies
abdulsaheel Nov 15, 2026
ae514de
Merge pull request #723 from kurexw/l10n/hi-supplement
abdulsaheel Nov 15, 2026
e1bd9d3
Merge pull request #722 from kurexw/l10n/zh-supplement
abdulsaheel Nov 15, 2026
dc6fe8f
Merge pull request #721 from kurexw/l10n/fr-supplement
abdulsaheel Nov 15, 2026
b9373e5
merge main
abdulsaheel Nov 15, 2026
a73618d
Merge pull request #720 from kurexw/l10n/es-supplement
abdulsaheel Nov 15, 2026
0cd3f8b
Merge pull request #731 from kurexw/fix/sleep-plan-duration-locale
abdulsaheel Nov 15, 2026
a3ef2a5
Merge remote-tracking branch 'refs/remotes/upstream/main' into feat/c…
mistral-vibe Nov 15, 2026
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions docs/FEATURES.md
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,23 @@ so a workout counts the same for a fit person with a low resting HR as for anyon
It needs three days of your data before it shows a number. Nightly HRV is RMSSD from one
estimator, and a night where the beat detector over-counts is refused rather than scored.

### Deviations from your own baseline

The overnight watches (resting-heart-rate trend, unusual-overnight-physiology, skin
temperature) compare each night with **your own** rolling baseline over valid nights only —
never against a population table, and never against a night that was not measured. The
baseline unlocks in stages and says so while it is learning: below 7 valid nights the
watches stay quiet ("learning your baseline"), and a wear gap longer than two weeks steps
the stage back down. Each night is also checked against **context you already know about** —
training load, accumulated sleep debt, journal entries (alcohol, late caffeine, a late
meal, vaccination, travel, feeling unwell) and, when cycle tracking is on, the second half
of the logged cycle — so an expected deviation (a hard workout, a couple of drinks, a
vaccine) is not reported as an anomaly. What remains after that subtraction is what the
watches report: a deviation from your baseline, stated as a pattern, never as a diagnosis.
These are approximations based on published research (e.g. Mishra et al., *Nature
Biomedical Engineering* 2020; Natarajan et al., *npj Digital Medicine* 2020) — not
medical-grade, not validated against a lab, and not a diagnosis.

## Activity

Auto-detected workouts, live workout tracking with GPS routes, heart-rate zones you can
Expand Down
2 changes: 1 addition & 1 deletion lib/ai/journal_ai.dart
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ import '../coach/coach_engine.dart';
const List<String> kJournalPresetTags = <String>[
'caffeine', 'alcohol', 'late meal', 'stress', 'poor sleep', 'travel',
'screens late', 'meds', 'sick', 'sauna', 'cold plunge', 'social',
'workout', 'rest day',
'workout', 'rest day', 'vaccinated',
];

/// One assistant turn: the conversational reply + the CUMULATIVE proposed log.
Expand Down
Loading
Loading