Prosocial

An experiment in AI-guided relationship intelligence.

N.B. This is incomplete for now.

Be the friend you want to have

We believe technology should help us be more human, not less. Prosocial uses AI and interaction data to surface insights that help you become a better friend—transforming the overwhelming complexity of maintaining relationships into an elegant, actionable experience.

This project is evolving. What started as a personal tool for staying connected is growing into a comprehensive platform that addresses real social challenges: the friend who drifts away, the birthday you forgot, the conversation you've been meaning to have.

Vision

Modern life makes it hard to be a good friend. We're scattered across cities, buried in notifications, and genuinely struggle to maintain the connections that matter most. Prosocial approaches this as a solvable problem—using the data we already generate (messages, calendars, interactions) to provide gentle, intelligent guidance on being more present for the people in our lives.

Features

Intelligent Relationship Tracking

• People Management - Store contacts with notes, tags, birthdays, and anniversaries—the foundation for AI-powered insights
• Today View - A curated daily feed of who needs your attention: upcoming birthdays, overdue check-ins, and calendar meetings
• Friendship Streaks - Gamified habit tracking that makes consistency visible and rewarding
• Quick Log - One-tap interaction logging that feeds the AI understanding of your social patterns
• Smart Reminders - Contextual push notifications that arrive when they're actually useful

Data-Informed Insights

Import your existing communication history to build a richer picture of your relationships:

• WhatsApp - Chat exports reveal conversation patterns and topics
• Discord - DM history surfaces digital friendships
• Email - Gmail, mbox, or eml imports for professional and personal contacts

AI Coaching

The core of what makes Prosocial different—AI that learns your social patterns and provides actionable guidance:

• Pre-meeting Briefs - Context and talking points based on relationship history
• Pattern Recognition - Surface trends in your social habits you might not notice yourself
• Awkwardness Detection - Identify relationships that need attention before they fade
• Conversation Starters - Personalized suggestions based on shared interests and history

Privacy by Design

Sensitive social data demands serious protection:

• Client-side Encryption - Notes are encrypted before they ever leave your device
• Row-level Security - Supabase RLS ensures complete data isolation between users
• Zero Tracking - Your relationship data is never sold or used for advertising
• Full Data Portability - Export everything, anytime—your data belongs to you

Tech Stack

Built for cross-platform reach with a privacy-conscious architecture:

Layer	Technology	Why
Frontend	Expo (React Native) + Expo Router	Single codebase for iOS, Android, and web
Styling	NativeWind (Tailwind for RN)	Rapid UI iteration with consistent design tokens
Backend	Supabase (Postgres + Auth + Storage)	Open-source, self-hostable, built-in RLS
State	Zustand	Minimal boilerplate, excellent TypeScript support
AI	OpenAI API	Flexible, with graceful fallback to local tips

Getting Started

Prerequisites

• Node.js 18+
• npm or yarn
• Expo CLI (npm install -g expo-cli)
• Supabase account (free tier works)

1. Clone and Install

git clone https://github.com/leo-hammett/prosocial.git
cd prosocial
npm install

2. Set Up Supabase

1. Create a new project at supabase.com
2. Go to SQL Editor and run the migration:

-- Copy contents of supabase/migrations/001_initial_schema.sql

3. Get your API credentials from Settings → API

3. Configure Environment

cp .env.example .env

Edit .env with your credentials:

EXPO_PUBLIC_SUPABASE_URL=https://your-project.supabase.co
EXPO_PUBLIC_SUPABASE_ANON_KEY=your-anon-key
EXPO_PUBLIC_OPENAI_API_KEY=sk-... # Optional, for AI features

4. Run the App

# Web
npm run web

# iOS Simulator
npm run ios

# Android Emulator
npm run android

5. Try Demo Mode

If you want to test without setting up Supabase:

1. Open the app
2. Click "Get Started"
3. Click "Try Demo Mode"

This loads sample data so you can explore all features.

Architecture

The codebase follows a clean separation of concerns—UI, business logic, and data access are decoupled to enable rapid iteration on any layer.

prosocial/
├── app/                    # Expo Router screens (file-based routing)
│   ├── (tabs)/            # Main tab navigation
│   │   ├── index.tsx      # Today screen - your daily social brief
│   │   ├── people.tsx     # People list with search/filter
│   │   ├── streaks.tsx    # Gamified streak dashboard
│   │   └── settings.tsx   # Privacy controls & preferences
│   ├── (auth)/            # Auth flow
│   │   ├── login.tsx
│   │   └── signup.tsx
│   └── person/            # Person detail screens
│       ├── [id].tsx       # Full relationship view
│       ├── new.tsx        # Add new person
│       └── edit/[id].tsx  # Edit existing
├── components/            # Reusable UI components
│   ├── ui/               # Design system primitives
│   ├── PersonCard.tsx    # Person list item
│   ├── StreakCard.tsx    # Streak visualization
│   └── TodayCard.tsx     # Daily action items
├── lib/                   # Core business logic
│   ├── supabase.ts       # Database client
│   ├── crypto.ts         # Client-side encryption
│   ├── streaks.ts        # Streak calculation engine
│   ├── coaching.ts       # AI coaching logic
│   ├── notifications.ts  # Push notification handling
│   ├── calendar.ts       # Calendar sync
│   ├── offline.ts        # Offline-first support
│   └── imports/          # Data import parsers
│       ├── whatsapp.ts   # WhatsApp chat parser
│       ├── discord.ts    # Discord DM parser
│       └── email.ts      # Email (Gmail/mbox/eml)
├── stores/               # Zustand state management
│   ├── authStore.ts      # Auth state + session
│   ├── peopleStore.ts    # People CRUD + cache
│   ├── interactionsStore.ts  # Interaction logging
│   └── streaksStore.ts   # Streak state + calculations
└── supabase/
    └── migrations/       # Database schema (version controlled)

Configuration

Push Notifications

1. Create an Expo account at expo.dev
2. Get your project ID
3. Add to .env:

EXPO_PUBLIC_PROJECT_ID=your-expo-project-id

4. Build with EAS:

npx eas build --platform ios
npx eas build --platform android

Google Calendar Integration

1. Create a project in Google Cloud Console
2. Enable Calendar API
3. Create OAuth credentials
4. Add to .env:

EXPO_PUBLIC_GOOGLE_CLIENT_ID=your-client-id.apps.googleusercontent.com

OpenAI (for AI Coaching)

1. Get an API key from platform.openai.com
2. Add to .env:

EXPO_PUBLIC_OPENAI_API_KEY=sk-your-key

Note: AI features work without this - they fall back to local tips.

Deployment

Web (Vercel/Netlify)

npx expo export --platform web
# Deploy the dist/ folder

iOS/Android (EAS Build)

# Install EAS CLI
npm install -g eas-cli

# Configure
eas build:configure

# Build for stores
eas build --platform ios
eas build --platform android

# Submit to stores
eas submit --platform ios
eas submit --platform android

Development

Running Tests

npm test

Linting

npm run lint

Type Checking

npm run typecheck

Privacy & Security

Social data is inherently sensitive. Prosocial takes a principled approach to data handling—collecting only what's needed for functionality, encrypting what's personal, and giving you full control.

Encryption Model

Data Type	Encrypted	Reason
Notes about people	✅ Client-side	Personal observations stay private
Interaction notes	✅ Client-side	Conversation details are sensitive
Names, dates, tags	❌	Required for search, sort, reminders
Relationship types	❌	Needed for filtering and AI context
Timestamps	❌	Required for streak calculations

User Control

• Full Export - Download all your data anytime (Settings → Export My Data)
• Complete Deletion - Remove your account and all associated data permanently
• Granular Permissions - Revoke calendar/Google access independently
• Self-hosting Option - Supabase backend can be self-hosted for complete data sovereignty

Roadmap

Prosocial is actively evolving. Current focus areas:

• Relationship Health Scores - ML-derived metrics for relationship strength
• Group Dynamics - Track friend groups and social circles
• Calendar Intelligence - Suggest optimal times to reach out based on both parties' schedules
• Memory Prompts - "Last time you saw Alex, you talked about..." context before meetings
• Cross-device Sync - Seamless experience across phone, tablet, and desktop

Contributing

This is an open experiment—contributions are welcome.

1. Fork the repo
2. Create a feature branch (git checkout -b feature/your-idea)
3. Make your changes
4. Submit a PR with a clear description of what problem you're solving

Particularly interested in contributions around:

• Accessibility improvements
• Additional data import sources
• Privacy-preserving AI techniques
• UI/UX refinements

License

MIT License - see LICENSE file

Connect

• Issues: GitHub Issues
• Email: support@prosocial.app
• Twitter: @prosocialapp

---

Prosocial is built on a simple premise: technology that understands our relationships can help us nurture them. This is an ongoing experiment in making that vision real—one friendship at a time.