# ⚠️ ARCHIVED — this repo's surviving pieces live in HealthClaw Guardrails
**Archived 2026-07-19.** SmartHealthConnect was the patient-facing surface of
the HealthClaw platform. Its good parts were salvaged into the engine repo and
this repo is now read-only history:
- **The 6 patient skills** (`skills/*/SKILL.md`) live on as **CareAgents
advisor definitions** — engine-backed by construction, so every clinical
claim routes through the guardrails
([HealthClawGuardrails#175](https://github.com/aks129/HealthClawGuardrails/issues/175)).
- **The care-gaps MCP-App view** was rebuilt on the engine's own `$care-gaps`
operation at `/r6/fhir/mcp-apps/care-gaps/`
([HealthClawGuardrails#176](https://github.com/aks129/HealthClawGuardrails/issues/176)
has the full 7-view decision table).
- **The engine/surface contract** (`.health-context.yaml`) survives as the
pattern — with [CareAgents](https://github.com/aks129/HealthClawGuardrails/tree/main/careagents)
as the consumer surface.
**Why archived** (recorded honestly in
[#12](https://github.com/aks129/SmartHealthConnect/issues/12)): ~39 of the ~40
MCP tools here bypassed the HealthClaw engine, contradicting the one claim the
platform is built on — guardrails live server-side, in one place. The ideas
were right; maintaining two data planes with one maintainer was not.
The Claude-plugin skills installed from this repo
(`smarthealthconnect@healthclaw-marketplace`) remain installable — archived
repos stay clonable — but are frozen at v1.2.0. New capability lands in
CareAgents.
---
# SmartHealthConnect (Liara AI Health)
The patient-facing surface of the HealthClaw platform.
**v1.2.0** — now reads wearable signals (Garmin, Oura, Polar, Suunto, Whoop, Fitbit, Strava) sourced from [HealthClaw Guardrails](https://github.com/aks129/HealthClawGuardrails) v1.3.0. Still routes through the Compiled Truth engine from v1.1.0.
## New in v1.2.0 — Wearable signals
When HealthClaw's `OPEN_WEARABLES_URL` is configured, wearable data arrives as FHIR Observations with LOINC codes and device Provenance. The existing `healthy-habits`, `diet-exercise`, and `medication-refills` skills read them through the same `fhir_search` they already use — no new tools in this repo. Each skill's `SKILL.md` now lists the LOINC codes worth knowing.
## Engine / surface contract
SmartHealthConnect **does not** own data, policy, or the canonical record.
- **Engine** — [HealthClaw Guardrails](https://github.com/aks129/HealthClawGuardrails) owns the FHIR store, PHI redaction, audit trail, step-up auth, tenant isolation, and the **Compiled Truth** primitive (current state + Provenance timeline for every resource).
- **Surface** — this repo owns the 6 patient skills (`healthy-habits`, `care-completion`, `medication-refills`, `diet-exercise`, `kids-health`, `research-monitor`), the React client, and the conversational patterns.
Every skill that makes a resource-specific claim to the patient must call the `get_compiled_truth` MCP tool first — it proxies to HealthClaw's `fhir_compiled_truth` and returns the current redacted resource + an append-only evidence timeline. See each skill's `SKILL.md` for the rule. Configure with `HEALTHCLAW_MCP_URL=http://localhost:3001/mcp/rpc` (or your deployed URL).
`.health-context.yaml` at the repo root declares `role: surface` and names HealthClaw as the engine. Mirrored in the engine repo.
## Overview
SmartHealthConnect enables patients to connect their healthcare providers and view their complete health records in one secure platform. The application uses SMART on FHIR standards for interoperability and delegates guardrail enforcement to HealthClaw.
## Features
- **SMART on FHIR Integration** - Connect to Epic, Cerner, and other FHIR-enabled healthcare systems
- **Unified Health Dashboard** - View conditions, medications, lab results, and vital signs in one place
- **AI Health Insights** - Get personalized health recommendations and care gap identification
- **Trend Visualization** - Track vital signs over time with interactive charts
- **White-Label Ready** - Enterprise solution with customizable branding
## Tech Stack
- **Frontend**: React 18, TypeScript, Vite, Tailwind CSS, shadcn/ui
- **Backend**: Express.js, TypeScript
- **Database**: PostgreSQL with Drizzle ORM
- **Authentication**: Passport.js with session management
- **FHIR**: SMART on FHIR R4 protocol
## Getting Started
### Prerequisites
- Node.js 18+
- PostgreSQL database
### Installation
```bash
# Install dependencies
npm install
# Set up environment variables
cp .env.example .env
# Edit .env with your DATABASE_URL and other config
# Push database schema
npm run db:push
# Start development server
npm run dev
```
### Environment Variables
- `DATABASE_URL` - PostgreSQL connection string
- `OPENAI_API_KEY` - OpenAI API key for AI features (optional)
## Development
```bash
# Run development server
npm run dev
# Type check
npm run check
# Build for production
npm run build
# Start production server
npm start
```
## Production Deployment
The application can be deployed to any Node.js hosting platform:
```bash
# Build production assets
npm run build
# Set environment variables
export DATABASE_URL="postgresql://..."
export OPENAI_API_KEY="sk-..." # Optional
export NODE_ENV="production"
# Start server
npm start
```
The server runs on port 5000 by default and serves both the API and static frontend assets.
## Project Structure
```
├── client/ # React frontend
│ └── src/
│ ├── components/ # UI components by feature
│ ├── pages/ # Route components
│ ├── hooks/ # Custom React hooks
│ └── lib/ # Utilities and config
├── server/ # Express backend
│ ├── routes.ts # API routes
│ ├── fhir-client.ts # SMART on FHIR integration
│ └── ai-service.ts # OpenAI integration
├── shared/ # Shared types and schemas
└── migrations/ # Database migrations
```
## Contributing
We're actively looking for:
- **Beta Testers** - Help us refine the user experience
- **Collaborators** - Developers, designers, and healthcare professionals
- **Health System Partners** - Enterprise deployments and integrations
## License
Proprietary - All rights reserved.
## Contact
For partnership inquiries or to join the beta program, please reach out through the platform.
