The upload execution policy stopped an agent at upload: App Review
submission and rollout each needed their own explicit authorization, so a
request to ship a release ended with an artifact sitting in App Store
Connect and a question back to the user. That contradicted the harness
standard in documentation/standards/react-native.md, which has always
authorized the full pipeline, and it is the wrong default: every step from
here is recoverable. A submission can be cancelled while it waits for
review, a phased release paused, a staged rollout halted.
A release request now authorizes build, verification, upload, App Review
submission, and rollout for the platforms it names.
What still needs its own authorization is anything that is not part of
shipping the requested release: widening TestFlight distribution to new
groups, and re-uploading after an ambiguous upload result. The existing
automatic-TestFlight-distribution check before an iOS upload is unchanged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Install expo-observe (56.0.29, matching the SDK 56 line) and configure
its endpoint under expo.extra.eas.observe in app.json, alongside the
existing EAS projectId. Leave dispatchInDebug and sampleRate at their
defaults.
Log an app_started custom event once per cold start from the root
layout's existing "run once" startup effect, and install a global JS
error handler that logs xprem_js_crash and then always delegates to
whatever handler was previously registered, so existing Crashlytics
reporting is unaffected.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Bump expo, react, react-native, and all expo-* / react-native-* peers to
the SDK 56 line via expo install --fix, adding expo-splash-screen and
expo-system-ui for config plugins the SDK split out of core. Bump
@bacons/apple-targets, react-native-view-shot, and other native modules
to versions compatible with React Native 0.85 and the new architecture.
Remove @react-navigation/* direct dependencies and route their hooks
through expo-router, which is no longer compatible with react-navigation
as of SDK 56. Fix RN 0.85's removed StyleSheet.absoluteFillObject (use
absoluteFill) and react-native-view-shot 5.x's ViewShotRef type across
call sites. Downgrade eslint-plugin-react-hooks v7's new React Compiler
era rules to warnings pending a dedicated refactor, and repair Jest
fallout from the jest-expo/reanimated/worklets/expo-media-library bumps.
Regenerate the committed ios/ and android/ native projects with
`expo prebuild --clean` (required for @bacons/apple-targets 5.x) against
the production release channel, and document that requirement.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The entrypoint guard compared process.argv[1] against
new URL(import.meta.url).pathname. A URL pathname percent-encodes spaces, so
in any checkout whose absolute path contains one the comparison never matched
and main() never ran.
Every ota:preflight:*, ota:publish:*, and ota:doctor:* script then exited 0
having done absolutely nothing, which is indistinguishable from success. A
publish that silently no-ops is worse than one that fails: it reports a
shipped update that never left the machine.
Compare resolved filesystem paths instead.
Verified from a checkout whose path contains a space: before the change
`node scripts/ota-publish.mjs production --preflight` exited 0 with no
output; after it runs a real export for both platforms and prints the
preflight confirmation. Argument validation still rejects unknown targets and
unsupported EOAS arguments.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Server registration is complete: app id 9a66f3c2-fda7-416f-87e2-6db833b32e9d,
real code-signing certificate (public half), staging/production channels
mapped server-side. Drops the app-id placeholder guard (cert guard stays) and
re-includes certificate.pem past the root *.pem ignore rule, which had been
silently excluding it.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Add docs/mobile-ota.md covering how OTA works here, the branch and channel
model, the appVersion runtime version policy and its store-build rule, the
eoas publish commands per channel, both rollback paths, the pending native
prebuild, and the exact placeholders the server deployment still has to fill.
Index it and cross-link it from the release guide.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Install expo-updates and the eoas CLI, and configure the client against
https://ota.apps.jbrunton.co.uk.
app.json stays the static source of truth for version, buildNumber, and
versionCode, which EAS rewrites during a release build. A new app.config.js
spreads it and layers on the updates block, because XPREM selects a release
channel from a build-time request header and that value differs per EAS
build profile.
- updates.url points at the XPREM manifest endpoint, /manifest
- runtimeVersion uses the appVersion policy, so the runtime version is the
store-visible expo.version and a native change requires a store build
- expo-channel-name comes from RELEASE_CHANNEL, set per EAS build profile:
development, staging (preview profile), and production
- expo-app-id and an empty xprem-branch header are declared at build time,
since expo-updates only sends headers that existed when the build was made
- code signing uses keyid main and rsa-v1_5-sha256 against a certificate at
certs/certificate.pem, skipped when DISABLE_CODE_SIGNING is set so a dev
client can run against the packager
The XPREM app id and the code-signing certificate are still placeholders
pending the update server deployment. The config throws a descriptive error
whenever RELEASE_CHANNEL is set while either is unresolved, so a
misconfigured build fails at config resolution rather than shipping a client
that can never validate a manifest.
The native projects are Expo CNG output and are left untouched: the
expo-updates config plugin embeds the real app id and certificate contents,
so they are regenerated as a reviewed source change once both values land.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The API has no CORS handling, so local admin development serves API calls
same-origin and forwards them to the local API process.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The dare proof migration file landed on staging without a journal entry, so the
drizzle migrator would never apply it. Register it in the journal and move the
solo sealed answers migration to 0022 to keep filenames and journal order aligned.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Replace the inline ShareableCard in SharePreviewModal with a designed
QuestionShareCard: Sauci logo + wordmark header, pack badge, and a
branded sauci.app footer pill on the captured story image. The card
deliberately never carries the couple join code - it is a private
pairing credential, not a referral link (a regression test enforces
this). Text-share fallbacks now carry a full https://sauci.app link,
and all three share paths fire a new question_shared analytics event
with the destination.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The sender can require photo or audio proof when sending any dare,
catalogue or custom. The recipient captures and uploads the proof as a
new couple-scoped 'dare_proof' media kind and attaches it on submit;
the server enforces ownership, kind, couple, and MIME-type match.
The proof renders inline on the dare card for the sender to review
before confirming completion.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Unpaired users could only reach a waiting-room dead end and could not answer
questions until a partner joined. Responses.couple_id is now nullable so a
solo user answers immediately and banks "sealed answers" (couple_id IS NULL).
Pairing claims both members' sealed answers and computes matches for every
question both have now answered. Invite copy and the unpaired push nudge
escalate with the sealed answer count.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Give support read-only visibility into purchase history, streak locality
fields, notification preferences, the dares loop, couples quiz sessions,
invite-code lookup, and recent live-draw activity, all backed by resources
already allow-listed in the admin repository.
Both share modals opened the instagram-stories:// scheme without ever
placing the captured PNG on the pasteboard, so Instagram opened an empty
story. Replace the manual scheme launch with react-native-share
shareSingle(INSTAGRAM_STORIES), which writes the image to UIPasteboard
under com.instagram.sharedSticker.backgroundImage on iOS and uses the
ADD_TO_STORY intent on Android. Gate the flow on the required Facebook
App ID (EXPO_PUBLIC_FACEBOOK_APP_ID) and Instagram being installed, with
fallback to the system share sheet, and remove the dead base64 reads.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Add a Quiz Questions content page (list/filter/create/edit/activate-
deactivate, no delete since quiz_answers reference questions by id),
route it under manage_questions, and add a sidebar entry.
Fix DaresPage so intensity (1-5) is editable and shown in the table;
previously every admin-created dare silently defaulted to intensity 1.
Add authored min/max intensity fields to DarePacksPage: no trigger or
worker in apps/api computes these columns, so they are admin-set
thresholds gating pack visibility, not derived aggregates. avg_intensity
is displayed read-only since nothing writes to it and it is not used
for filtering.
Extract dare intensity and quiz option validation into small lib
modules with focused unit tests.
Lets new users skip the onboarding avatar step and instead defers the ask
to a lightweight, dismissible prompt shown once after the couple's first
match. Also personalizes the home screen's first pack recommendation using
the usage_reason already collected during onboarding, which now round-trips
through GET /v1/me.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Resolve the earlier open question by mapping the profile-settings
domain's notification toggles to the existing notification_preferences
table (primary key user_id, no sensitive columns). Read-only, gated on
view_users, with a test covering read access and rejected writes.
Extend the admin allow-list so upcoming admin screens can reach quiz
content and activity, sent dares, and dare chat messages. Also fix a
schema drift where question_packs.scheduled_release_at and
release_notified existed in the database but were missing from the
drizzle TS schema, which could have let a future generate propose
dropping them.
Note: no profile_settings table exists in the schema. The
profile-settings domain reads and writes the existing profiles and
notification_preferences tables, so that entry was intentionally
omitted pending a product decision on which table it should map to.
Expand the seeded pool from 12 to 36 questions (migration 0020) and
select 10 per session preferring questions the couple has never seen,
then least recently used, with random tiebreaks, so repeat quizzes stop
serving the same set.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
jjbruntonandClaude Opus 5