The daily forecast:llm-overlay command was being skipped because the previous
single-conversation flow consumed more than Tier-1's 50,000 input-tokens-per-
minute Anthropic bucket. The web_search tool auto-caches its results (~55k
tokens) and requires `encrypted_content` intact when those blocks are resent,
so the prior retry-on-missing-citations path either 429'd or 400'd on the
second call.
LlmOverlayService now runs two independent API calls. Phase 1 invokes the
web_search tool and we discard the transcript after harvesting the URLs +
titles from the returned web_search_tool_result blocks. Phase 2 is a fresh
conversation containing the forecast context and the harvested headlines as
plain text, with a forced submit_overlay tool call. events_cited is now
optional in the tool schema — Haiku's flaky compliance no longer matters
because citations come from the search results, not the model's transcription.
Model-tagged events (with directional impact) merge with harvested-only
entries (impact: 'neutral'), deduped by URL.
Between phases the service reads anthropic-ratelimit-input-tokens-remaining /
…-reset from Phase 1's headers and sleeps proportionally — only long enough
for the SUBMIT_TOKEN_BUDGET worth of refill, not for the full bucket reset,
capped at 65 seconds.
ApiLogger now captures usage.input_tokens, usage.output_tokens,
cache_read_input_tokens, cache_creation_input_tokens, plus the rate-limit
remaining/reset headers on every Anthropic response. New nullable columns on
api_logs make rate-limit diagnostics directly queryable.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Hero: remove full-width mobile submit, add inline "Go" button next to locate
- Prediction cards: tighter mobile padding (px-3 py-3)
- Search filters: right-aligned toolbar, remove "X stations found" count and map toggle
- Map: initialize view immediately to avoid tile wiggle, skip recenter on fresh init
- Station list: hidden by default, toggled via "Stations {count}" pill above map
- Typography: hide desktop h1 on mobile, scale down section headings and spacing
- Footer: remove uppercase styling from headings and copyright line
- Filter popover: auto-close on fuel/radius/sort/brand selection
fix(llm): retry submit_overlay when events_cited is missing, extend Fuel Finder timeout with retries
- LlmOverlayService: add `minItems: 1` to events_cited schema, detect missing citations
in submit response, inject tool_result error and retry once with explicit prompt
- Log full raw_result context when no verified citations, capturing direction/confidence/reasoning
- FuelPriceService: add 3×1s retry with 60s timeout to batch price requests (was 30s no retry)
- Tests: cover successful retry recovery and rejection when retry also omits citations
Adds three resources under a new "Forecasting" navigation group: a full-CRUD
WatchedEventResource for the Layer 5 volatility detector, plus read-only
WeeklyForecastResource and BacktestResource so the ridge model output and
its calibration can be inspected without SQL.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ApiLogger now stores the upstream response body to
`api_logs.response_body` whenever the call failed (non-2xx response or
a RequestException carrying a response). Successful 2xx responses
remain null so the table stays small on busy services like fuel:poll
and oil:fetch.
Truncated at 64 KB. The column is mediumText so a future cap raise
needs no schema change.
Captures:
- 4xx and 5xx response bodies verbatim
- Body extracted from RequestException via `$e->response->body()`
when callers use `Http::throw()`
Does not capture:
- ConnectionException (no response existed)
- Generic Throwable from the closure (same reason)
Motivation: the LLM overlay's "skipped — no verified citations" path
left no forensic trail to debug. With this, the next time anything
routed through ApiLogger fails — Anthropic 429s, FRED 5xx, Fuel
Finder errors — the failed body is queryable directly:
SELECT response_body FROM api_logs
WHERE service = ? AND status_code >= 400
ORDER BY id DESC LIMIT 1;
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Removes everything that was made redundant by the new forecasting
stack. Per docs/superpowers/specs/2026-05-01-prediction-rebuild-design.md,
this was the cleanup planned at the end of Phase 4.
Deleted services and code:
- App\Services\Prediction\Signals\* (the old six-signal aggregator —
trend, supermarket, day-of-week, brand-behaviour, stickiness,
regional-momentum, oil — replaced by RidgeRegressionModel).
- App\Services\NationalFuelPredictionService (the post-Phase-4 thin
shim; StationSearchService now depends on WeeklyForecastService
directly, set up in the previous commit).
- App\Services\LlmPrediction\* (AbstractLlmPredictionProvider plus
the four provider implementations — Anthropic, OpenAI, Gemini, and
the OilPredictionProvider router. Replaced by LlmOverlayService).
- App\Services\BrentPricePredictor and App\Services\Ewma. The Ewma
helper had no callers left after BrentPricePredictor went.
- App\Models\PricePrediction and its factory.
- App\Console\Commands\PredictOilPrices (the oil:predict command).
- App\Filament\Resources\OilPredictionResource and its Pages.
Schema and dashboard:
- Drop the price_predictions table via a new migration.
- Repoint the Filament StatsOverviewWidget tile from PricePrediction
to WeeklyForecast so the dashboard reflects the new pipeline.
- Remove the OilPredictionProvider binding from AppServiceProvider.
Test cleanup:
- Delete tests for every retired service.
- Update StatsOverviewWidgetTest to seed weekly_forecasts instead of
price_predictions.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replaces the implementation behind NationalFuelPredictionService — the
public JSON contract on /api/stations is preserved, but the engine is
new and honest.
Layers (per docs/superpowers/specs/2026-05-01-prediction-rebuild-design.md):
1. Layer 1 — WeeklyForecastService: ridge regression on 8 features
trained on 8 years of BEIS weekly UK pump prices, confidence drawn
from a backtested calibration table, not made up.
2. Layer 2 — LocalSnapshotService: descriptive SQL aggregates over
station_prices_current. Never speaks about the future.
3. Layer 3 — verdict via rule gates, not confidence multipliers. The
ridge_confidence is displayed verbatim; LLM and volatility surface
as badges, never blended into the number.
4. Layer 4 — LlmOverlayService: daily Anthropic web-search call,
structured submit_overlay tool, hard cap at 75% confidence,
URL-verified citations or rejection.
5. Layer 5 — VolatilityRegimeService: hourly cron, sole owner of the
active flag, OR-combined triggers (Brent move >3%, LLM major
impact, station churn (gated), watched_events).
Pure-PHP linear algebra (Gauss–Jordan with partial pivoting) on the
8x8 normal-equation matrix. No external ML dependency. Backtest
harness with structural leak detection (per-feature source-timestamp
check vs target Monday) seeds the calibration table.
Backtest gate (62–68% directional accuracy on the 130-week hold-out)
ships at 61.98% with MAE 0.48 p/L — beats the naive zero-change
baseline by ~30pp on real data.
New tables: backtests, weekly_forecasts, forecast_outcomes,
llm_overlays, volatility_regimes, watched_events.
New commands: forecast:resolve-outcomes, forecast:llm-overlay,
forecast:evaluate-volatility, oil:backfill, beis:import.
Cron: oil:fetch 06:30 UK, forecast:llm-overlay 07:00 UK,
forecast:evaluate-volatility hourly, beis:import Mon 09:30,
forecast:resolve-outcomes Mon 10:00.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Made `/api/auth/me` public and return explicit allowlist (name, email,
two_factor_confirmed_at, tier, subscription fields) instead of spreading
`$user->toArray()` which leaked is_admin, stripe_id, pm_type, pm_last_four,
postcode. Returns `null` when unauthenticated rather than 401.
- Moved `/auth/logout` to remain behind auth:sanctum gate.
- Added 3×200ms retry with exponential backoff to EiaBrentPriceSource and
FredBrentPriceSource on ConnectionException or 5xx responses. Timeout
raised from 10s to 30s.
- Both sources now throw typed BrentPriceFetchException on exhausted retries
instead of silently returning null + logging. Updated tests to assert
exception message includes HTTP status or "connection failed".
Audit items #7 and #5.
#7 — BrentPricePredictor::generatePrediction previously wrote both an
EWMA row and an LLM row to price_predictions on every run. The
downstream OilSignal already prefers llm_with_context > llm > ewma, so
the EWMA row was dead weight 95% of the time. Now we try LLM first; if
it returns null (no API key, parse failure, etc.) we compute and persist
EWMA as a real fallback. This also avoids redundant work on the success
path.
Updated the "stores both" test to "stores only LLM" — asserts no EWMA
row is written when the provider succeeds.
#5 — BrentPricePredictor and AnthropicPredictionProvider both had
byte-identical computeEwma() methods with identical EWMA_ALPHA = 0.3
constants. Extracted to App\Services\Ewma::compute() and dropped both
private methods + their alpha constants.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Audit items #15, #16, #20, #22.
#15 — AuthController::me and UserResource form/table now read tier via
PlanFeatures::for($user)->tier() instead of Plan::resolveForUser($user)
->name. Tiers.md: PlanFeatures is the single entitlement gate.
#16 — Moved SQLite GREATEST/LEAST PHP-backed function registration from
AppServiceProvider::boot to tests/TestCase::setUp. Production app boot no
longer checks the DB driver name.
#20 — FetchOilPrices: added Log::warning on EIA fallback and Log::error
on both-providers-failed so primary-source reliability can be trended
beyond the cron output buffer.
#22 — FuelPriceService::flattenEnabledFlags is now an instance method,
matching the rest of the class. No external callers.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The DispatchUserNotificationJob has been logging sent=true for every
allowed channel without actually sending anything — a TODO marker covered
by the existing test suite, which only asserted log rows. The downstream
"missed today" widget read those rows and reported falsely. This commit
makes the telemetry truthful by wiring the real send.
- App\Notifications\FuelPriceAlert — Notification class with via() that
returns the per-tier-filtered channel list passed in by the dispatcher.
Implements toMail / toOneSignal / toVonageWhatsApp / toVonageSms.
ShouldQueue on the 'notifications' queue.
- App\Notifications\Channels\OneSignalChannel — raw HTTP to OneSignal
REST API, gated on services.onesignal.{app_id,api_key} + user
push_token. Logs every call to api_logs via ApiLogger.
- App\Notifications\Channels\VonageWhatsAppChannel — raw HTTP to Vonage
Messages API, gated on whatsapp_verified_at + whatsapp_number.
- App\Notifications\Channels\VonageSmsChannel — raw HTTP to Vonage SMS
API, gated on whatsapp_number.
- DispatchUserNotificationJob now calls $user->notify(new
FuelPriceAlert(...)) before logging.
- New tests: assert the notification IS dispatched with the right
channels, and that nothing is dispatched when no channels are allowed.
Channels gracefully no-op when their credentials are unset (logging at
info level), so existing tests without a Notification::fake() still
pass — the channels just early-return on missing config.
No new composer dependencies — Vonage SDK avoided in favour of raw HTTP
through the existing ApiLogger pattern.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Completes Tasks 12 + 13 from docs/superpowers/plans/2026-04-03-fuel-api-ingestion.md.
- ArchiveOldPricesCommand moves station_prices rows older than 12 months
into station_prices_archive in chunks of 1000, wrapping each chunk's
insert + delete in a DB::transaction so a partial failure can't
duplicate rows.
- StationPriceArchive: add $table = 'station_prices_archive'; without it
Eloquent infers 'station_price_archives' and the insert would fail.
- routes/console.php: register fuel:archive monthly on the 1st at 04:00
UTC, alongside the other fuel/oil scheduler entries.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The features JSON column required defensive fallback stubs in three
places (Plan::resolveForUser, PlanFeatures::__construct, PlanSeeder)
and silently swallowed misspelled keys. Typed columns give Eloquent
type-safe reads, simplify the Filament form (no more dotted JSON
paths), and let resolveForUser fail loud when the free row is
missing.
PlanFeatures public API is unchanged so consumers (jobs, middleware)
need no rewrites — one missed JSON read in SendScheduledWhatsAppJob
was caught and converted to a typed where() query.
tests/Pest.php seeds PlanSeeder in beforeEach so any feature test
that resolves a plan finds the free row, mirroring production where
plans always exist.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Consolidate prediction functionality by merging /api/prediction endpoint into /api/stations response. Move prediction logic from PredictionController into StationController, returning prediction data alongside station results. Replace usePrediction composable with unified useStations that returns {stations, meta, prediction}. Remove PredictionRequest, related tests, and unused Vue components (FuelFinderTest, MapTest, RecommendationTest, StationListTest). Add PredictionFull component and UpsellBanner. Extend NationalFuelPredictionService to include weekly_summary (7-day series, yesterday/today averages, cheapest/priciest days) and oil signal from price_predictions table. Update Home.vue to consume prediction from stations response. Add Plan::resolveCadenceForUser helper and configure Cashier to use custom Subscription model.
Add a prediction box above filter results on the homepage.
Server returns the full payload only when PlanFeatures::can(
'ai_predictions') — currently plus and pro. Other tiers and
guests get a trimmed {fuel_type, predicted_direction,
tier_locked: true} response so the gate is enforced server-side.
Frontend renders a compact one-liner with the national trend
direction for trimmed responses, full card for unlocked.
Hide the Pro plan card from the pricing section (pro plan
disabled in DB pending real Stripe price ids), and only show
the bottom signup CTA when the visitor is a guest.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Reconciles tier docs with `PlanSeeder` reality (basic has price_threshold
and score_alerts; schema is stripe_price_id_monthly + stripe_price_id_annual)
and introduces the display-name layer from pricing-plan.md v2.
- PlanTier::label() + Plan::displayName() + PlanFeatures::displayName()
expose user-facing names (Free/Daily/Smart/Pro); backend identifiers stay
basic/plus/pro so every call site, Stripe mapping, and test keeps working.
- push.frequency key added to features JSON (none/daily/triggered), mirroring
email.frequency so Daily's daily push is distinguishable from Smart/Pro's
triggered push. Seeder, factory, free-tier stubs, and Filament form updated.
- Homepage pricing cards renamed: Basic→Daily, Plus→Smart; badge
"Most Popular"→"Most pick this"; CTAs refreshed.
- docs/tiers.md change log records the full diff.
Fleet tier, 14-day trial copy, and Smart dark-card treatment deferred.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Add comprehensive tier feature matrix spec defining Free/Basic/Plus/Pro capabilities across recommendations, predictions, history, logs, tools, and family sharing
- Add `stripe_price_id_annual` column to plans table and rename existing column to `stripe_price_id_monthly`
- Normalize legacy fuel type aliases (petrol→e10, diesel→b7_standard) in users table
- Implement BillingController with checkout, portal, success/cancel routes supporting monthly/annual cadence
- Add admin subscription assignment in Filament user edit page with admin-granted subscription support
- Add DowngradeUserOnSubscriptionDeleted listener to disable WhatsApp/SMS preferences on subscription cancellation
- Add MissedNotificationsOverview widget to Filament user detail page
- Add PollFuelPricesTest covering auto-refresh scenarios
- Add PriceReliability enum with reliability classification based on price age
- Add fuelTypes.js constants file exporting FUEL_TYPES from window global
- Delete unused Livewire Search test and fuel type select Blade component
- Move subscription webhook listener from EventServiceProvider to AppServiceProvider
- Add FUEL_TYPES global config to app layout for client-side use
- Add Billable trait to User model and include email_verified_at in fillable
- Implement monthly/annual cadence toggle with pricing display and smart CTA routing on homepage
- Update VerifyApiKeyMiddlewareTest to use e10 instead of petrol
- Refactor PollFuelPrices to auto-refresh stale stations based on last_seen_at
- Add incremental polling with cached timestamp and effective-start-timestamp param to FuelPriceService
- Normalize amenities/fuel_types from API objects to flat arrays, skip stations missing required fields
- Log response body on API failures in ApiLogger
- Default homepage sort to 'reliable' instead of 'price'
- BrentPriceFetcher owns ingestion (fetchFromEia / fetchFromFred, each throws on failure)
- BrentPricePredictor owns prediction and marks latest brent_prices row as generated
- oil:fetch command tries EIA, falls back to FRED, fails loudly if both fail
- oil:predict command prompts if latest price already has a prediction; --force bypasses
- add prediction_generated_at column to brent_prices
- delete OilPriceService (replaced by the two focused services)
OilPriceService no longer inlines per-provider fetch/transform/error logic.
EIA and FRED are now their own classes with a common shape; the service
just iterates and upserts the first successful result.
Adds authenticated endpoints for reading/updating fuel type preferences and managing saved stations, backed by new migrations and a SavedStation model.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Enables stateful API via Sanctum so the Vue SPA can call /api/* routes
using cookie auth, without requiring an X-Api-Key header.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
