Crisis Forecaster
Crisis Forecaster is a Siri-native early-warning agent for sickle cell disease.
Core trick
Fuses HealthKit vitals and WeatherKit readings into a prompt that scores 24-72 hour crisis risk and drafts a handoff packet on elevated scores.
Description
Crisis Forecaster is a Siri-native early-warning agent for sickle cell disease. Vaso-occlusive crises are preceded by signals — rising resting heart rate, falling HRV, blood-oxygen dips, fragmented sleep, and weather shifts — but no patient-facing tool fuses them.
Crisis Forecaster reads HealthKit vitals + WeatherKit + the patient's own check-ins, and Claude Opus 4.8 scores crisis risk 24–72h out and explains why in plain language.
On elevated risk it auto-drafts an Emergency Passport — the ER handoff packet — so the handoff is done before the visit.
iOS 27 App Intents make it ambient: the patient asks Siri or glances at a lock-screen widget and never opens the app.
Built from 37 years of lived experience with the disease.
Prediction is the feature; the Passport is the moat.
README excerpt
# Crisis Forecaster **A Siri-native, vaso-occlusive-crisis (VOC) early-warning agent for sickle cell disease.** Built at the **Anthropic Hackathon 2026** on the iOS 27 dev beta. Vaso-occlusive crises rarely come from nowhere — they're preceded by signals: rising resting heart rate, falling HRV, SpO2 dips, sleep fragmentation, and weather shifts (barometric pressure drops, low humidity, cold). Crisis Forecaster fuses HealthKit vitals + WeatherKit and asks **Claude Opus 4.8** to score crisis risk 24–72h out, explain *why* in plain language, and — on elevated risk — auto-draft an **Emergency Passport** (the ER handoff packet) so the handoff is already done before the visit happens. > **Not medical advice.** Crisis Forecaster is an informational early-warning and > ER-handoff aid. It surfaces the patient's *own* self-management plan and organizes > their data for clinicians — it does not diagnose, prescribe, or replace a care team. > In an emergency, call 911 or your hematologist. ## Dual-model architecture (the iOS 27 spin) Two models, clean division of labor: - **Apple Foundation Models (on-device, iOS 27 `SystemLanguageModel`) — the triage nurse.** When the patient logs a check-in, the on-device model privately + instantly reads their free text, normalizes it into a clean clinical line, scores concern, and decides whether to escalate. No network, no data leaving the device. (`Services/OnDeviceTriage.swift`, guided generation via `@Generable`; availability-gated w