The mood page called the plain `pearsonCorrelation` — strength banded from
|r| alone above a floor of five pairs, no significance test — so five days
of noise sloping the right way arrived at the reader as a "Strong" badge,
with n and r behind an info icon. The guarded twin was written for exactly
that failure and names it in its own comment, and the weight, blood-pressure
and mood-status surfaces had already moved onto it. The mood page had not.
Route the page's five cross-metric channels through
`significantPearsonCorrelation` (n >= 20 AND p < 0.05) and carry the reason
a pairing was refused, so the card can distinguish "still building" from
"twenty days and nothing in them" instead of showing the same blank for
both. New copy in all seven locales.
Add a structural guard that fails when a new caller of the unguarded helper
appears outside a written allowlist. Two callers remain on it, each named as
open work rather than as a blessing: the /insights envelope and the AI
feature payload both still band at n >= 5, and moving them changes payload
shapes their own tests pin.
The operator's global provider always built an OpenAI client, so an admin
who saved an Anthropic key and https://api.anthropic.com/v1 (a host the
settings allowlist accepts) had every AI call fail: an OpenAI-shaped body
posted at Anthropic's Messages API, with nothing in the error pointing at
the mismatch.
Route by hostname instead. api.anthropic.com and any subdomain of
anthropic.com get the same AnthropicClient the per-user ANTHROPIC arm
builds, with the same Claude model default; every other host keeps the
OpenAI client, which is what an OpenAI-compatible endpoint needs. The
match parses the URL rather than searching the string, so a lookalike
host cannot claim the Anthropic wire, and an unparseable value keeps the
OpenAI path it had before.
The chain tag stays admin-openai: it names the slot, not the wire, and it
is persisted in user chains and pinned by the OpenAPI enum. The wire is
already visible in the instance type the wide events read.
Settings, Sources listed GOOGLE_HEALTH, OURA, POLAR and STRAVA by their
internal name: the catalogue was written for seven sources and every
provider added since fell through to the raw value. The measurement list
carried the same fall-through in its source filter and its table column,
and the medication intake history read "via APPLE_HEALTH" for a dose
synced from Apple Health.
The three catalogues move into one module keyed by the enum, so an
omission is a type error rather than a raw value on screen, and the sleep
discrepancy tooltip drops the private brand-name map it had grown to work
around the gap.
The published contract still told a client that a Bearer token can never
clear the step-up on account deletion or the record wipe, which is what
an app generates its client from. Describe the elevation arm on both:
which factors qualify, that a password proof does not, that the proof
rides the X-Step-Up header, and that it is spent only when the erasure
runs. The 403 a narrow-scope token gets was never published either;
it is named now, ahead of the second-factor question it precedes.
The mint endpoint lists the surfaces an elevation unlocks, so both
routes join that list.
Also note on the record wipe that API tokens are on its own delete list:
a client that wipes over Bearer signs itself out, which is worth knowing
before the retry rather than after it.
Account deletion and the record wipe both demand a fresh second-factor
proof once a factor is enrolled, and that proof lived on the session row.
A native client has no session row, so an account with TOTP or a security
key could not delete itself from the app on any request it could build:
requireFreshMfaIfEnrolled fell through to requireFreshMfa, which resolves
via getSession and answers 401 to a perfectly valid token. App Store
guideline 5.1.1(v) asks for exactly that deletion.
Add requireFreshMfaOrElevationIfEnrolled beside the cookie-only sibling
rather than softening it. Its cookie arm is the sibling's, unchanged;
its Bearer arm takes the single-use elevation the second-factor
management routes already accept, with the fresh-factor rule pinned on,
so a password-proved elevation cannot erase an account any more than a
password login satisfies the gate on the web. No factor rides the request
body: one mint surface, one redemption path. Scope stays fail-closed, so
a narrow token is refused before the header is read.
The elevation is spent immediately before the erasure, after the typed
confirmation and the last-admin check, so a rejected request does not
burn a proof against a 5-per-15-minute mint ceiling. The Bearer branch
of requireMfaManagementAuth moves into a shared helper so both gates read
and claim in the same place.
The elevation guard grows a section naming the two routes, because the
set of surfaces an elevation can unlock only stays small if adding to it
is a visible edit.
Every background path that resolved a user row's locale collapsed a
NULL column to English: the nightly briefing warm, the status crons,
the period-narrative warm, reaction lines, the morning refresh, the
briefing push, coach nudges, the reminder workers, and the document and
workout summaries each carried their own copy of
`locales.includes(x) ? x : defaultLocale` or called
`normalizeLocale(user.locale)`. None consulted the instance's configured
default language, so an account that never touched the language picker
got English prose and English pushes on a German-language instance.
Add resolveJobLocale (src/lib/i18n/job-locale.ts): stored user locale
when valid, else the operator's app_settings.default_locale when valid,
else English. The settings read lives beside the other app-settings
accessors. Every job and daily path resolves through it once, where it
has the row; the template helpers keep coerceLocale for a value that is
already resolved, and queue payloads keep normalizeLocale.
The briefing push hands its resolved locale down to the digest, and the
digest to the snapshot read, so the push copy and the digest lines are
in the same language instead of two different fallbacks.
A structural test walks src/lib/jobs and src/lib/daily and fails on a
hand-rolled coercion, a validator applied to a row column, or a row
reader that does not import the helper. It was red against the previous
tree on all three counts.
The comprehensive-briefing cache (User.insightsCachedText) is one slot
per account with no record of the language it was written in. Every
reader served whatever was there, so a German-language account whose
nightly warm had written English kept seeing English on the dashboard
lead and the Today briefing. The request path only replaced it once the
row was a day old and the provider answered, which with a provider at
its usage limit was never.
Add User.insightsCachedLocale (migration 0335). Both writers store the
locale they generated in; the timestamp-only refresh on an unchanged
snapshot tags the row too, since the fingerprint covers the generation
locale. Every reset that nulls the text nulls the tag.
The snapshot briefing lift and GET /api/insights/generate treat a tag
that differs from the reader's locale as an empty cache: no prose, no
timestamp, the preparing / no-provider state, and the out-of-band warm
enqueued in the reader's locale through the existing path. The
generator's own short-circuit and the POST short-circuit refuse to call
a foreign-language row fresh. A NULL tag (rows written before this
change) is served as before until the next write tags it. The Today
digest derives from the snapshot and follows.
Covered by unit tests on the lift, the route and the generator, and an
integration test that writes the row through Prisma and reads the
snapshot for a reader in another language.
The Anthropic vision check listed the Claude 3 and 4 families by prefix,
so a key pinned to claude-sonnet-5, claude-opus-5 or a Fable model was
told the provider cannot read images, and lab and document OCR refused
the upload even though the model handles image and PDF blocks.
Replace the per-release prefix list with a rule: any claude-<tier>-<major>
slug from the 4 family upward, plus every claude-fable-* model, counts as
vision-capable. Claude 3 handling is unchanged. The Settings preset list
gains claude-sonnet-5 and claude-opus-5 alongside the existing entries.
The Anthropic client steered JSON replies by appending an assistant turn
containing an opening brace and gluing the brace back onto the reply.
Current Claude models (the 4.6 / 4.7 / 4.8 family, Opus 5, Sonnet 5,
Fable 5 and 5.1) reject an assistant prefill with HTTP 400, so every
JSON surface on the provider (briefing, status cards, document and lab
extraction) failed for a self-hoster on a current model while plain chat
kept working.
Drop the prefill on every model. The request now carries a plain
instruction on the last user turn to reply with a single JSON object and
nothing else, and the reply is narrowed to the outermost object body via
a shared extractJsonObject helper, so a fenced or prefixed answer still
parses. The insights stripJsonFences helper delegates to the same
function rather than keeping a second copy. The tool-use path is
unchanged.
Reported independently by @sweidinger.
The v1.38.0 notice that warns when the configured OAuth callback origin
differs from the origin the browser is on never rendered in the published
image, and the copyable callback address in the redirect guide showed a
bare path. `integrationCallbackUrl()` read `process.env.NEXT_PUBLIC_APP_URL`
inside a "use client" module. Next.js inlines NEXT_PUBLIC_* into client
bundles at build time, and the GHCR image is built without that variable
(only the version, build-sha and built-at args reach the build), so the
base compiled to an empty string, `new URL("/api/<provider>/callback")`
threw, and the notice returned null on every card for every self-hoster.
The value now comes from the server at request time. A new map
(`src/lib/integrations/callback-urls.ts`) calls each provider's own
redirect-URI function (the four private `getRedirectUri()` helpers are
exported under provider-specific names; Polar, Oura and Strava already
exported theirs), so the page shows exactly what the handshake registers,
explicit `*_REDIRECT_URI` override included. The settings route resolves
the map and passes it as a prop through `IntegrationsSection` and
`ConnectionsPanel` to every card. Fitbit and Google Health validate their
URI and throw on a bad configuration; that surfaces as `null`, the notice
skips the comparison and the guide falls back to the fixed path.
Tests: a map test proving the explicit override wins and the derived form
is used otherwise, a component test rendering the notice from a passed
value (mismatch renders, match does not), and a structural guard that
fails when any "use client" file under src/components/settings/integrations
reads NEXT_PUBLIC_APP_URL.
A lab report's reference range reaches the record by four paths. Three of
them validate that the floor does not exceed the ceiling; the document
path did not, and the catalog store writes what it is handed. An
integration run against a real Postgres confirmed the consequence: a
100/30 pair landed on the row and on the freshly minted biomarker, so
every later reading of that analyte inherited a window that cannot be
true and a high value came back reassuring.
Dropping beats swapping. A transposed pair says the transcription is
wrong, not which of the two numbers is, and inventing an order would
state something the report never printed. The verbatim string survives
either way, so nothing the document actually said is lost.
Three points, answering according to what each one is:
- the extraction schema drops the pair, because every field there is
catch-guarded so one bad value cannot sink a 120-fact document
- the review-screen edit refuses with the wording the lab schemas
already use, because a person can see both numbers and fix them
- the commit path drops it as the last gate, which is what covers facts
staged before this rule existed and confirmed afterwards
parseReferenceRange has answered a transposed PRINTED window this way
since it was written, and the commit path already counts a window it
could not read at all. This extends both to the case that was readable
and cannot be true.
The mint used requireAuth() with no scope, which refuses a narrow token
but still admits a cookie-equivalent one. The reach is not the problem —
a wildcard token can already do everything this endpoint does. The
lifetimes are: a native access token lives a day, and what it mints here
lives a year. So a credential that leaked for an afternoon could leave
behind one that outlives revoking it, and the revocation would look like
it had worked.
requireCookieAuth() closes that. It is the helper the passkey
registration and trusted-device routes already use for the same reason,
and the sharing guard recognises it as cookie-only.
The prose said more than the code did, in three places: the route header
and the OpenAPI description both claimed the endpoint "cannot mint
another token", which held only for narrow tokens, and the 403 arm
described a scope refusal that a Bearer never reaches. All three now say
what is true — no Bearer credential reaches this endpoint at any scope,
and a Bearer is refused 401 before the operator switch is read.
Tests follow the status change: the integration case asserting a scoped
token cannot mint now expects 401 rather than 403, since the refusal is
authentication and not scope, and a companion case pins the wildcard —
the credential class this change is actually about — and asserts no
token row is created.
`stdResponses` already carries 401, 422 and 429. The three response
blocks added with the measurements:write scope each declared their own
422 (and, on the mint route, a 429) BEFORE spreading it, so the generic
wording overwrote the specific wording and TypeScript refused the
duplicate key. The build failed at auth.ts with "'422' is specified more
than once".
Move the explicit keys after the spread, which is what the MCP mint next
door already does and the only order in which a per-route description
survives.
Worth stating because the failure mode is quiet: had TypeScript allowed
it, the published contract would simply have carried "Request validation
failed" everywhere instead of naming
measurement.create.source_not_permitted and its siblings.
Since v1.30.17 there has been no self-service path to a token that can
write a measurement. That release made the Bearer scope default
fail-closed, which closed a real privilege-escalation hole, and removed
the generic mint at POST /api/tokens along with it. The generic mint had
been the only way to get a token that could reach /api/measurements, so
a local sensor pushed through Home Assistant or Node-RED lost its door
while the OAuth integrations kept theirs.
Add a narrow scope accepted by exactly two routes, both writes, both on
the holder's own record.
The scope constant is a leaf module beside the surface that consumes it,
matching FHIR_READ_SCOPE and the MCP scopes. requireRecordAuth grows an
options object rather than a sibling function, because the sharing
surface guard keys every leg on that one identifier and a second name
would mean two copies of the delegable declaration. AuthContext carries
the admitting scope as a single nullable string rather than the token's
permission array, so nothing downstream can re-derive an authorisation
decision from it.
A scoped credential is refused a record selector before any grant is
read. Placing the refusal above the sharing check rather than beside it
means no future widening of what a grant permits can reach one of these
tokens.
Rows are attributed MANUAL. An entry that explicitly names APPLE_HEALTH
is refused 422 rather than relabelled: that source is half the
(userId, type, source, externalId) dedup key the iOS client writes into,
it participates in the cross-source merge, and it gates the HealthKit
sync fields. For the same reason a scoped write no longer advances
User.lastSyncedAt, which the phone reads to size its undelivered window.
Also closes a narrower gap found while building this. The idempotency
wrapper resolved a replayed delegated response against a live grant
without considering which credential was presenting it, so a cached
response about a shared record could be returned to a token the handler
itself would refuse. No write was reachable that way; the response body
was.
Guards: T3 gains the mint site, T4's vocabulary gains the scope and
learns to read the requireRecordAuth form (without which this change
would open a way to widen a route that the closed vocabulary could not
see), and a new T5 freezes the set of delegable routes naming a scope.
The sharing surface guard's AST matcher reads only the first two
arguments, so it needed prose but no code.
The constant was left over from an earlier shape of the guard, which
located the fixture by path before it imported it directly. Nothing reads
it now, and the dead-code gate says so.
Three descriptions in the OpenAPI registry still describe the floor that
v1.38 removed, and one of them contradicts its own error-response
description two dozen lines below it.
- The score-config PATCH still said the write is refused below three
distinct areas and one physical measurement. The route refuses the
empty selection and nothing else; its 422 description already said so.
- The dashboard snapshot's healthScore still said the block is null until
three eligible domains include a physiological one. It is null only
when no pillar is readable at all.
- The pillar domain field still pointed at a three-distinct-domains rule
that no longer exists, where what a client needs to hear is that domain
count comes resolved on scoreBasis and must not be counted out of
composition.
Regenerated docs/api/openapi.yaml.
The user-facing description of the score lives in the docs site's own
repository, not here; nothing under docs/ describes the composite.
The score once graded weight against 22 x height squared, widened by 2 kg
either side: a band nobody set, no surface named, and the only producer
of that pillar's verdict. v1.34 removed it and the composite rebuild left
a pillar set where every scored band cites somebody. Nothing but review
has kept it that way since, and review is what let the original in.
So the rule is written down. PillarReference.kind becomes a runtime list
carrying the reason there is no fourth kind, reference.source is
documented as a citation a reader could go and look up, and a guard
beside the pillars checks all seven modules three ways.
The arm worth arguing about is the second. An invented target is a
continuous function of a profile field; a guideline band is a step
function into a published table. So the guard sweeps every pillar across
a matrix of heights and ages and requires the scored bands it emits to
land inside a small enumerated set transcribed from the guideline that
pillar cites. No finite enumeration can hold a per-person curve, which is
what makes the arm able to fail rather than merely able to pass. The
other two arms cover the citation string and freeze which pillar may read
which profile field, with the reason written beside it and stale entries
failing as loudly as missing ones.
The proof is in the file: the pre-v1.34 weight target is restored as a
fixture in today's shape and each arm is shown refusing it.
A pillar can leave the set behind the number without anybody choosing it:
a window rolls past its floor, a module goes off, a source goes quiet. The
delta already refuses to narrate that as a health event, which stops a
false "you dropped six points" and puts nothing in its place. The level
moves, the panel shows a different number, and the cause is nowhere on the
surface. Under the adaptive minimum this matters more rather than less,
because compositions are smaller now and sit closer to their floors.
The report carries `compositionNotice` when today's set differs from the
last local day the account has a stored score for. Optional and nullable,
the additive shape the score's wire has used six times, so a payload
cached before it existed stays valid. It names what left and what joined
as pillar ids; the surfaces already own the labels, and the reason a
departed pillar stopped counting comes from that pillar's own row in the
same report rather than from a second vocabulary.
Two changes deliberately do not raise it. A method move and a recipe
change each already have their own note, and telling someone twice about
one event teaches them to ignore both.
The key is `health-score:v<version>:composition:<hash of the sorted set>`
and rides the existing dismissed-item ledger, so a dismissal holds until
the set moves again rather than until tomorrow. The read excludes today on
purpose: the row for today is written by the same request that computed
the report, so comparing against the newest row would find today's own
composition on the second load of the same page and the note would vanish
between two views of it.
This gives `HealthScoreRecord` its first production reader. Since it
shipped it has been written on every scored read and consumed by nothing
but backup and restore, which is the write-only shape the contributor
notes call a liability and which nothing in the gate can notice, because
every check proves one end. The round trip is proved against real
Postgres: raise the note, wipe the table, restore it through the admin
route, get the same note back, and hold the dismissal across it.
Rendering is one muted line on the score card and a card of its own on the
settings page, beside the recipe note rather than inside it. Deliberately
not a push.
Four fake Prisma clients gain the new read. They are doubles of a client
the reader really does call now; leaving them short and guarding the call
site instead would have been a shim for a caller that does not exist.
Six suites pinned the old rule as the rule. Each inversion, and why it is
one:
`breadth.test.ts` — the case table asserted a refusal reason per set; it
asserts a tier and a `physiological` flag now. Every row in it except the
empty set used to be a refusal, and each is a number somebody gets to see.
Adds the tier boundaries (one pillar, two areas, physiological-absent), a
case proving the same pillars produce the same number at `partial` and at
`full`, and a block whose only job is the counter-case: the empty verdict
still carries a reason and no tier, and a composite whose selected pillars
all came back short still refuses with `no_usable_data`. A rule that has
stopped refusing anything is a deleted rule, not a graded one.
`composite.test.ts` — "requires three domains and one measured
physiological domain" asserted the refusal that no longer happens; it now
asserts the score and the basis block on the same fixture. The coverage
case keeps its fixture and its arithmetic and flips its outcome: two of
three used to be a refusal and is a scored account with a real coverage
fraction. Adds the one-area case and the zero-pillar counter-case, and the
`SCORE_VERSION` floor moves 2 -> 3 because widening who gets a score is
exactly the kind of change that must not ship without a version bump.
`record.test.ts` — "records nothing when the composite did not resolve"
used a two-pillar fixture to prove a below-the-floor day stored nothing.
That day has a number now and the table has to keep it, or an account that
starts narrow and widens later has a hole where its early days were. The
assertion is now that a one-pillar day builds a draft AND reaches the
database; the absence case survives on the fixture that genuinely produces
no score.
`composition-source.test.ts` — three recipes asserted as refused at read
time. Two of them score. What the block proves is unchanged — the reader
publishes the shared rule's verdict on the reader's own resolved set, so a
second copy of the rule would still surface here — but the agreement is
over a tier rather than over a refusal.
`route.test.ts` (score config) — the two too-narrow refusals become
acceptances, because a write that refuses what the read is willing to serve
is the two layers disagreeing. The empty selection keeps its 422 and gains
the assertions the old cases carried: errorCode, reason, and the sentence.
The errorCode driver loses the two selections that no longer 422.
`score-refusal-consumer-guard.test.ts` — its non-zero control anchored on
`three_domains_required`; it anchors on `no_pillars_selected`.
New: `health-score-basis-additive-contract.test.ts` pins that `scoreBasis`
is published, is optional on all three carriers, and points at the one
shared component — a required field on a cached surface would invalidate
every cached payload with nothing failing anywhere. It also pins that the
walk does not confuse the composite's basis with the older per-pillar
`scoreBasis` that blood pressure uses for its axis, since they share a key.
Fixtures: the notice-key assertions derive the method half from
`SCORE_VERSION` rather than restating 3. Three fixtures that named
`three_domains_required` now name `no_usable_data`, a reason the server can
still emit. The settings e2e drove its refusal through a two-pillar
selection that now succeeds; it deselects every pillar instead.
`scoreBasis` rides the composite, the dashboard snapshot's flattened hero
block and the daily digest's score. It says how many distinct areas of
health the number counted, how many the method recommends, which tier that
lands in, and whether any counted pillar was a physiological measurement.
Optional everywhere, because two of the three surfaces are served out of a
cache and a payload written before this release cannot grow a field. The
snapshot and digest blocks have taken six additive fields this way already;
this is the seventh.
Resolved on the server for the same reason `configured` is. Counting
`composition` client-side would get it wrong more often than right — three
of the seven pillars share the cardiometabolic area, so a three-pillar
score can rest on one.
One shared component, `HealthScoreBasis`, declared once in the insights
schema module and referenced from the other two registry modules, rather
than three inline copies that drift the first time the block gains a field.
Declared through the assigning `.meta()` form: a bare `.meta()` statement
registers an instance nothing references, and the id then never becomes a
component.
The composite's `insufficient` reason narrows to the single value
`no_usable_data` and says so in the published description. That field is an
open string in the document, so the change is additive on the wire even
though the value space shrank.
The score-config 422 description drops the two retired reasons and states
plainly that the empty selection is now the only refusal, so a client does
not keep a three-area rule of its own in front of a write that no longer
has one.
Three distinct areas of health stops being the price of admission and
becomes what it always described: a recommendation. `evaluateScoreBreadth`
returns a tier instead of a pass/fail — `full` at three areas or more,
`partial` at two, `minimal` at one — and refuses only when not one pillar
came back usable, with the reason `no_usable_data`.
The arithmetic does not change with the tier. Same mean, same worst-pillar
band drag, same noise floor. A one-area score of 82 is an honest 82 that
rests on one area, and discounting it for the pillars somebody does not
track would have been a new invented target, which is the thing this score
spent two releases removing.
The physiological requirement becomes a labelling fact rather than a veto.
An account whose only readable pillar is activity or wellbeing gets its
number with `physiological: false` on the basis block; the old refusal told
that person nothing they could not already see.
`SCORE_MIN_ELIGIBLE_DOMAINS` is now `SCORE_RECOMMENDED_DOMAINS` and still
feeds `deriveCoverage`, so the coverage fraction and the confidence blend
keep telling the truth with no new machinery: a one-area score lands at low
confidence through the path that was already there. The fraction used to be
a floor every scored account had cleared; it is a real moving number now.
`SCORE_VERSION` 3 -> 4 and `SCORE_ALGORITHM_CHANGED_AT` moves to the
release date. Both are method-identity duties: deltas suppress as
`algorithm_changed` across the straddle window, the notice key rotates so
each account is told once, and stored days stamp 4. No v3 row is rewritten
and no v3 number changes meaning.
The settings write follows the same rule rather than a stricter one. It
refused two-area and non-physiological selections on the ground that they
could not produce a score; they can, so refusing the write would be the two
layers disagreeing about the one question they share. The empty selection
is still refused, under a single reason `no_pillars_selected`, and the seven
message catalogues trade the two old refusal sentences for one.
Also fixes the stale `./series.ts` reference in the config header — that
module was deleted as dead in v1.35; the columns the sentence describes are
still written by `./record.ts`.
Five senders (APNs, Web Push, webhook, ntfy, email) each carried their
own single-pass replace(/<[^>]*>/g, "") — the shape CodeQL flags as
incomplete multi-character sanitization, because a removal can re-form
the pattern it removed. Whether one particular regex happens to be
idempotent is an argument about that regex; five copies of the argument
is how the next edit quietly breaks one of them.
The shared stripHtml re-applies the strip until the string stops
changing, with an iteration cap and a bracket-dropping fallback so no
tag-shaped sequence can survive any input. Plain text passes through
unchanged, same as before. Unit tests pin the nested-tag bypass shapes
and the no-surviving-tag invariant.
Resolves CodeQL alerts 38-42.
The sent-latch was a module-level boolean, so every worker boot re-sent
the same notice while the host's configuration never changed (#851).
Anchor the notice in the notification_events ledger instead, through the
same claimNotificationEvent helper the reminder schedulers use. The
anchor is unwindowed: entering the unconfigured state sends once per
admin, staying in it is silent across any number of restarts, and the
anchor is released when the state exits — databases configured, or
IP_GEO_LOOKUP_DISABLED set — so a genuine re-entry notifies again.
Three consequences worth naming:
- A host with lookups disabled is no longer notified at all. With no
egress there is nothing the notice warns about, and the disabled state
counts as an exit, matching the notice's own remediation advice.
- The evaluation now keys on offlineGeoReady() rather than the ASN
reader, so a host with only the ASN database mounted (city lookups
still egress) hears about it, where before it stayed silent.
- The retention prune excludes the new ADMIN_CONFIG_NOTICE event type:
a state-scoped anchor has a reader for as long as the state holds, and
the 90-day horizon delete would have re-opened the repeat on a slower
clock.
Refs #851
The per-user retry ledger has recorded every AI delivery outcome for a
long time, but no surface ever read it for the operator: a central
provider could fail on every call for weeks and the admin console said
nothing. The AI settings section now carries a health card, backed by a
new admin-only endpoint that folds the ledger into one row per provider
type: users tracked, users currently failing, the worst uninterrupted
failure run, and the last success and failure instants. Operator-managed
types sort first because their failure takes the shared fallback down
for everyone whose chain relies on it. Counts and timestamps only; the
response never names a user.
The Korean questionnaire wording was translated freshly from English, so
a score built from it was comparable to nothing in the literature. Each
instrument now carries the text its own publisher put out, or no Korean
at all.
PHQ-9 and GAD-7 come verbatim from the instrument owner's distribution —
Pfizer's `PHQ9_Korean for Korea.pdf` (2010-06-15) and `GAD7_Korean for
Korea.pdf` (2010-05-18), the latter being the file Ahn/Kim/Choi 2019
names as what it validated. The sheets grant reproduction, translation,
display and distribution without permission. Verbatim means verbatim:
the spacing in "7 일 이상 방해 받았다" and the sheet's "안절부절 못하거나"
against the GAD-7 sheet's "안절부절못해서" are reproduced as printed.
WHO-5 has no official Korean arm. Neither the WHO's 2024 distribution
(26 languages) nor the Psychiatric Research Unit folder it inherited
from (31 languages) has ever carried one, and ePROVIDE lists the same
31. Korean therefore carries the published forward/back-translation
from Kim HJ et al., J Korean Geriatr Psychiatry 2010;14(2):90-96, which
is CC BY-NC and so redistributable; the attribution line names it. Its
validation sample was community-dwelling elderly rather than general
adults, which the definition says out loud.
The Sleep Condition Indicator reverts to English. It is validated in
English only, its cut-off was derived from those words, and no openly
redistributable Korean exists — so `ko` now matches de/es/fr/it/pl:
items, section stems and response anchors byte-identical to en.json,
under the localized "validated in English" note.
`ALL_APP_LOCALES` grows to seven, which is what stops the check-in
wizard claiming English wording over Korean items on every Korean
check-in.
One gap is deliberate and not yet resolved. The official Korean PHQ-9
and GAD-7 sheets disagree on anchor 2 — "7 일 이상 방해 받았다" against
"2 주 중 절반 이상 방해 받았다" — but both instruments read one shared
`mentalHealth.options` set, which holds for the other six locales
because their two sheets agree word for word. Korean takes the PHQ-9
wording for both here. GAD-7's anchors are therefore validated in
substance but not verbatim; splitting them into a `gad7Options`
namespace is the fix, and is left for the maintainer to weigh against
the 24 duplicated keys it would add across the other locales.
The functional-impairment item and its four anchors come from the same
PHQ-9 sheet. They are not scored and not covered by the
`validatedItemLocales` contract.
The SSRF pin inventory asserted a literal unconditional pin in the
openai-client; the previous commit made that pin conditional for the
two person-typed base-URL tags. The guard now asserts the exact
ternary — allowlist helper for gateway/admin-key, literal true as the
floor — so neither the floor nor the condition can drift silently.
The global admin provider's base URL is operator-typed in the admin
settings, but the client pinned requirePublicHost unconditionally for
its tag — so an operator pointing it at a private OpenAI-compatible
proxy (an OAuth sidecar on the instance's own Docker network) got a
safeFetch refusal on every call, indistinguishable from an outage. The
admin-key tag now consults the same ALLOW_LOCAL_AI_PRIVATE_HOSTS
policy as the user-configured gateway: a public URL never consults the
allowlist, a private host without an entry is refused exactly as
before, and both ends of the grant are the operator's own. The codex
tag stays hard-pinned; its base URL is a repository constant. Every
non-operator admin-key construction pins the public api.openai.com
constant, so the personal-key posture is unchanged.
The test asserting the old blanket pin is split into the three cases
that actually differ: allowlisted private host passes, unlisted
private host still refuses, codex refuses even a blanket allowlist.
The tier for a chart series keyed off the requested window width. The
All range always requests ~3650 days, so every account got MONTH
buckets regardless of how much history exists — a record spanning four
months came back as four monthly means, and the chart captioned them
from the real span as a weekly average. The reader now re-reads at the
tier the returned rows' span calls for; a coverage miss on the finer
tier keeps the coarser rows.
The ladder comment claimed to mirror the client's pickBucket
thresholds and misquoted them. The two ladders are deliberately not
identical; a test now pins the direction that keeps the caption honest
(server never coarser than the client captions) for every span up to
ten years.
Registering `ko` in the `Locale` union left seven locale-keyed tables
without an entry, so the build could not type-check and the safety
screens, the citation-coverage grader and the reference-range parser
read nothing at all for a Korean reader.
Three failures were not missing keys but wrong assumptions:
- The shared fold decomposes with NFD and strips U+0300-U+036F. Latin
accents become plain letters, but Hangul decomposes into jamo, which
are not in that range and survived — so folded Korean text was a jamo
sequence no precomposed table entry could ever match. It recomposes to
NFC now; every Latin, Polish and German fold is byte-identical.
- `normaliseLabKey` kept `[a-z0-9]`, which reduced every Korean analyte
name to the empty string, so the derived index held none of them and a
Korean cholesterol result reached a FHIR export uncoded.
- The reference-range parser had the hyphen and the dashes but not the
tilde a Korean lab actually prints (`3.5~5.0`).
The Korean safety banks carry no `\b` next to a Hangul token — `\b` is
an ASCII `\w` boundary and Hangul is not `\w`, so `/\b증량/` matches
nothing — and the clause order is the mirror of the Latin banks, since
Korean closes with the verb. The dose patterns require a target-marking
particle for the same reason the Latin banks require "to"/"by": without
it the permitted restatement "7.5mg을 복용 중이고, 체중은 내려가고
있어요" tripped the lowering pattern.
The Coach and Insights prompts route `ko` down the reviewed English body
plus a Korean reply directive; there is no `safety-contracts.ko.yaml`, so
the native builder has no Korean ground rules to compose from.
Register `ko` in the locale list, language names, and labels, and ship
messages/ko.json so the whole UI — dashboard, settings, share views,
clinician PDF, notifications, and AI prompts — renders in Korean.
Extend the locale-parity, English-leak, plural-completeness, and copy
guard tests to cover the new locale, and update README/CONTRIBUTING to
list Korean among the supported languages.
`GET /api/insights/derived` accepts `windowDays`, bounded to 1..90, and
threads it to the dispatcher. Omitting it changes nothing: every engine
keeps its own default, so a caller that asks for nothing reads exactly
what it read before.
Ninety is where the tier this route promises runs out. The rollup router
opens its WEEK floor at 91 days, and the baseline engines cannot compose
a spread from WEEK buckets, so a 91-day request falls through to a raw
`measurement.findMany` with no row cap — for a densely sampled type that
is a full history scan, and the analytics-read budget allows 120 of them
a minute per account. The trailing `series` is capped at 30 points
regardless, so past the ceiling a wider window would only broaden the
mean behind the value. Out of range is a 422 rather than a clamp.
An unhonoured window is visible rather than silent. `provenance.windowDays`
is the window the engine actually used and `coverage.historyDays` the days
that actually backed it, so a three-week record answering a thirty-day
request reports both numbers instead of implying coverage it does not
have. `HEALTH_SCORE` composes fixed per-pillar windows and reports the
widest; a new suite freezes that it is the only metric that does, so an
arm added later cannot quietly drop the parameter.
Recovery keeps its canonical resolution under the wider window. A worn
band stamps the wake morning and the computed proxy the night before, and
only `resolveCanonicalRecovery` pairs them; a test seeds forty such
nights, asks for sixty days, and holds the answer to forty nights with a
flat trend — a bypass leaves eighty rows and a false climb.
Contract: model `WellnessScoreValue` field for field and publish it. The
envelope's `value` has to stay an open record because one route serves
eighteen differently-shaped metrics, so nothing can `$ref` the shape and
it went out undocumented — `series`, `daysInWindow` and `asOf` were on
the wire with no way to find them short of inspecting a live response. It
is registered in the forced-components slot instead, held against the
runtime interface by a test, and the fields it does not cover are named
in the schema comment so the remaining gap is known rather than invisible.
Both routes were exempt from the contract as `shareLink` — "authenticated
by the URL credential, for a recipient with no account". That reason
holds for the page a person opens. It does not hold for two routes whose
whole purpose is to hand a practice a file it files into another system,
which is exactly the audience a contract has. `/c/{token}/d/{id}` was
published all along, one route away in the same tree.
Each entry states the gate it shares with the page, the frozen selection
it can never exceed, the flat 404 that covers every miss class including
a documents-only link, and the 20/h per-link bucket. The FHIR entry says
plainly that it is a download and not a REST face, and names the two
resource families the bundle deliberately omits.
The create-request description claimed "there is no FHIR or other
machine-readable face behind a share token", which stopped being true
when these routes landed. It now describes what a link actually serves
and records that the insurance leaf is refused outright.
The coverage guard gains the other half of its staleness check. It caught
an exemption for a route that had disappeared, and never the reverse — an
exemption left behind after the route was published, which is how these
two would have stayed on the list while the contract already carried
them. Break-proof: re-add either entry and it goes red naming it.
A leaf on a share link can be absent from the payload for two reasons
that look identical downstream: the owner shared it and recorded
nothing, or they shared it and the domain is switched off on their
account. The aggregator ANDs the selection and the module map and
returns the same null either way, so the recipient could not tell an
empty section from one that never had a chance to carry anything.
`loadShareViewData` now resolves the whole module map — it needed the
`doctorReport` key from it anyway — and returns `unavailableLeaves`: the
leaves the link DOES carry whose owning module is off. Only selected
leaves appear, because a withheld leaf's module state is not the
recipient's business in any direction.
The map is deliberately NOT handed to the aggregator. The third argument
is the frozen selection and there is no fourth, which is what keeps this
surface from growing an options object that widens what it asks for; the
resolver memoises its reads per request, so the aggregator resolving its
own map again costs no round-trip.
The module-gate suite moves off `isModuleEnabled` with it, and gains
three cases for the new verdict, including the one that proves an
unshared leaf is never named.
Three surfaces built the label key by interpolation — the doctor-report
PDF's visit table, the daily digest's upcoming-visit line and the
clinician share view. `EncounterKind` is `ROUTINE` and the bundle leaf is
`routine`, so all three rendered `encounters.kind.ROUTINE` verbatim: into
a document a practice files, onto a lock screen, and onto the page.
Route them through `encounterKindLabelKey`, a literal switch beside the
existing `encounterKindLabel`, so the mapping is written down once.
The enum-derived i18n guard could not have caught this: it only knows
key spaces of the form `prefix + member`, and this one is not. It gains a
second arm for spaces whose mapping is a function, with `encounters.kind`
as its first entry. Break-proof: point one arm of the resolver back at
the uppercase key and six locale cases go red.
The authenticated shell wraps every route and read one destination
predicate off the `@/lib/record-settings` barrel. That barrel also
re-exports the managed settings-patch schemas, so Zod, the module
registry and three validation modules rode into the eager client
bundle of every authenticated page for a function that inspects a URL
slug. The shell imports `record-settings/classification` now; the
barrel loses the re-export nobody else used.
Measured over all 585 client-rendered routes: none grew, 100 fell, by
4.7 KB gz at the median and 8.1 KB at the most. /insights had been
sitting on its 445 KB ceiling for three releases and went over it on
this branch's CI run; it measures 437 now, /measurements 435, / 450,
/insights/mood 450.
The whole-build sum moved the other way, 3451 to 3476, because the
aggregate counts each emitted chunk once: a graph that leaves the
chunk group every route shares gets counted again wherever it lands,
here inside the deferred settings chunks that use it. Bytes on disk
rose, bytes a reader downloads only fell. The aggregate cap moves to
3500 with the same 24 KB headroom the four earlier re-measurements
kept, and the reasoning sits next to the number.
The two halves of the ECG fix were built apart. One taught the parser
Apple's full English verdict set; the other taught it the metadata keys and
verdict wording a German watch writes. Each was correct alone, and each
left "Sinusrhythmus" unresolved for a different reason.
Together they meet: the alias table translates the wording onto Apple's
own English, and the classifier decides what it means. One place still
owns the meaning, which is what stops a per-language verdict table from
drifting the first time Apple adds a case.
The test that expected null was pinning the gap between the branches
rather than a decision about German. It now expects NOT_DETECTED, and a
French verdict takes over the null case, because four of the six
languages genuinely have no observed key map yet.
The CSV is written in the device's language, and the keys are translated
along with the values: a German watch writes `Aufzeichnungsdatum` and
`Klassifizierung`, not `Recorded Date` and `Classification`. Indexing on
the English keys found none of them, so the parser never located the
waveform and rejected the recording outright — a lost import, not a
missing field.
Two alias maps close it, both taken from real exports rather than from a
dictionary: nine metadata keys and two verdicts, consistent across every
German file observed. Adding a language is a data addition to those maps
and nothing else.
The verdict map is an indirection onto the English wording rather than a
second verdict-to-enum table, so there stays exactly one place that
decides what a verdict means.
The other four shipped languages are absent on purpose. A wrong key
silently mis-files a value and a wrong clinical verdict is worse than
none, so an export this parser cannot place is refused — which an operator
can see — rather than read into the wrong column, which they cannot. Two
gaps are written down at the map: no observed file carries an
average-heart-rate row in any language, and the paired layout's column
header has only ever been seen in English.
The rate matcher demanded "Hz". Observed exports spell the unit out and
lower-case it — "511,422 hertz" — and the German ones separate it from the
number with a no-break space. The fixtures had only ever used "512 Hz", so
the mismatch never showed.
Rate and heart rate now split into a magnitude and a unit word, with the
magnitude read through the same decimal helper the waveform uses. "512 Hz",
"512 hertz" and "511,562 Hertz" all resolve; a missing unit, the wrong
unit, a grouped number and a rate past the bound are all still refused.
The unit word is matched loosely and the magnitude strictly. That is the
right way round: the unit is presentation and varies by region, while the
number is the measurement.
A watch set to a comma-decimal region writes its samples as "-180,596".
The single-column reader rejected any row containing a comma outright, so
every one of those files failed on its first sample row.
The two waveform layouts disagree about what a comma means, which is why
this cannot be decided per line. The parser now names the section it is in
— header, single-column, paired — and both crossings are commented at the
point they happen: the paired layout announces itself with a `Lead,Voltage`
column header, while the single-column one begins at the first blank line
AFTER Lead and Unit have been declared, because real exports also carry
blank lines mid-header.
Scalars in both waveforms now read either dialect through one helper, so
"-180.596" and "-180,596" give the same microvolts and neither is
privileged. A number carrying thousands grouping is refused rather than
guessed at, since resolving it needs the file's region and the file never
states it. The one genuinely ambiguous shape is a separator followed by
exactly three digits; it is read as a decimal, because observed waveforms
put two- and three-digit fractions in the same file and grouping is always
three, and because a machine writing one bare value per row has no reason
to group at all.
Marc-André Bombeck