MessageAI

Intelligent team messaging with AI-powered features

A production-ready React Native messaging app with real-time communication and 6 AI-powered features for productivity. Built with Expo, Firebase, and Claude AI.

✨ Key Features

🔥 Core Messaging

Real-time messaging with instant delivery
Direct & group chats with participant management
Read receipts with detailed read status in group chats
Typing indicators with presence tracking
Online/offline status for all participants
Local notifications with conversation deep links
Offline support with message retry/delete

🤖 AI-Powered Productivity (6 Features)

📝 Thread Summarization
- Generate AI summaries of long conversations
- Configurable message count (10, 25, 50, 100)
- Extracts key points and main topics
🔍 Semantic Search
- Search by meaning, not just keywords
- Powered by embeddings (OpenAI + Pinecone)
- Fast vector similarity search
✅ Action Items Extraction
- Auto-detects tasks from conversations
- Assigns to participants
- Tracks completion status
- Priority levels (high/medium/low)
🎯 Priority Detection
- Automatically flags urgent messages
- Visual badges for high-priority items
- Background processing pipeline
📊 Decision Tracking
- Captures key decisions from discussions
- Confidence scoring for decisions
- Searchable decision history
📅 Smart Meeting Scheduler
- Analyzes scheduling discussions
- Suggests 3 optimal meeting times
- Provides reasoning for each suggestion
- Copy-to-clipboard for quick sharing

🛡️ Production-Ready Error Handling

3-tier error boundaries (app, screen, feature)
User-friendly error messages (technical → plain language)
Message retry/delete for failed sends
AsyncStorage persistence (last 100 errors, last 50 failed messages)
Developer logging with context and stack traces

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                    React Native App                          │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  Screens (Expo Router)                                 │ │
│  │  - Auth (login, register)                              │ │
│  │  - Conversations List                                  │ │
│  │  - Chat Screen (messages + AI features)               │ │
│  │  - New Chat (direct + group)                          │ │
│  └────────────────────────────────────────────────────────┘ │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  Components                                            │ │
│  │  - MessageList, MessageBubble, MessageInput           │ │
│  │  - AI Modals (Summary, Search, Actions, etc.)        │ │
│  │  - Error Boundaries & Fallbacks                       │ │
│  └────────────────────────────────────────────────────────┘ │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  Services                                              │ │
│  │  - authService, firestoreService                      │ │
│  │  - aiService (Cloud Functions wrapper)                │ │
│  │  - presenceService, notificationService               │ │
│  │  - errorLogger, failedMessagesService                 │ │
│  └────────────────────────────────────────────────────────┘ │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  State (Zustand + AsyncStorage)                       │ │
│  │  - authStore, chatStore                               │ │
│  └────────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘
                            ↕
┌─────────────────────────────────────────────────────────────┐
│                    Firebase Backend                          │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  Firebase Authentication                               │ │
│  └────────────────────────────────────────────────────────┘ │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  Cloud Firestore                                       │ │
│  │  - /users, /conversations, /messages                  │ │
│  │  - /typingUsers, /ai_summaries, /ai_action_items     │ │
│  │  - Real-time listeners, participant-based security    │ │
│  └────────────────────────────────────────────────────────┘ │
│  ┌────────────────────────────────────────────────────────┐ │
│  │  Cloud Functions (Node.js + TypeScript)               │ │
│  │  - AI features (summarize, search, actions, etc.)     │ │
│  │  - Rate limiting, caching, embeddings pipeline        │ │
│  │  - Security validation                                 │ │
│  └────────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘
                            ↕
┌─────────────────────────────────────────────────────────────┐
│                  External AI Services                        │
│  ┌──────────────────┐  ┌──────────────────┐                │
│  │  Anthropic       │  │  OpenAI          │                │
│  │  Claude 3.5      │  │  text-embedding  │                │
│  │  (Sonnet)        │  │  -3-small        │                │
│  │  - Summarization │  │  - Embeddings    │                │
│  │  - Extraction    │  │                  │                │
│  │  - Analysis      │  │                  │                │
│  └──────────────────┘  └──────────────────┘                │
│  ┌──────────────────┐                                       │
│  │  Pinecone        │                                       │
│  │  Vector DB       │                                       │
│  │  - Semantic      │                                       │
│  │    search        │                                       │
│  └──────────────────┘                                       │
└─────────────────────────────────────────────────────────────┘

See ARCHITECTURE.md for detailed documentation.

🚀 Tech Stack

Frontend

React Native 0.81 with Expo SDK 54
Expo Router (file-based navigation)
TypeScript for type safety
Zustand for state management
AsyncStorage for persistence

Backend

Firebase Authentication (email/password)
Cloud Firestore (NoSQL database)
Cloud Functions (Node.js 20 + TypeScript)
Firebase Admin SDK for server-side operations

AI Services

Anthropic Claude 3.5 Sonnet (summarization, extraction, analysis)
OpenAI text-embedding-3-small (semantic embeddings)
Pinecone (vector database for semantic search)

Testing & Quality

Jest (265 tests: 212 backend + 53 frontend)
ESLint with TypeScript rules
TypeScript strict mode
Firebase Emulator for integration tests

📦 Installation & Setup

Prerequisites

Node.js 20+ (use nvm use to auto-select)
Expo CLI
Firebase CLI (npm install -g firebase-tools)
iOS Simulator (Mac) or Android Emulator

1. Clone & Install

git clone <repository-url>
cd message-ai
npm install

2. Configure Environment Variables

Create .env in project root:

# Firebase Config (from Firebase Console)
EXPO_PUBLIC_FIREBASE_API_KEY=your_api_key
EXPO_PUBLIC_FIREBASE_AUTH_DOMAIN=your_project.firebaseapp.com
EXPO_PUBLIC_FIREBASE_PROJECT_ID=your_project_id
EXPO_PUBLIC_FIREBASE_STORAGE_BUCKET=your_project.appspot.com
EXPO_PUBLIC_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
EXPO_PUBLIC_FIREBASE_APP_ID=your_app_id

# AI Services (Optional - only needed for AI features)
ANTHROPIC_API_KEY=sk-ant-...
OPENAI_API_KEY=sk-...
PINECONE_API_KEY=...
PINECONE_ENVIRONMENT=us-west1-gcp
PINECONE_INDEX_NAME=messageai-embeddings

See PHASE_0_SETUP.md for detailed Firebase setup guide.

3. Deploy Backend

# Deploy Firestore rules
firebase deploy --only firestore:rules

# Deploy Cloud Functions (for AI features)
cd functions
npm install
npm run build
firebase deploy --only functions

4. Run the App

npx expo start

Then press:

i for iOS Simulator
a for Android Emulator
Scan QR code for physical device

🧪 Testing

Run All Tests

npm run validate  # Lint + type-check + all tests

Unit Tests (Frontend + Backend)

npm test          # Run all unit tests (265 tests)
cd functions && npm test  # Backend only (212 tests)

Integration Tests (Firebase Emulator)

# Terminal 1: Start emulator
firebase emulators:start

# Terminal 2: Run tests
npm run test:integration

What's Tested:

Rate limiting (per-user, per-conversation, global)
Caching (age-based, message-count invalidation)
Security (participant validation, access control)
Error handling (translation, logging, retry/delete)
Message array operations (pagination, deduplication)

Performance Testing

# Create 1500-message test conversation
node scripts/createPerformanceTestData.js

# Test burst performance (10 messages in 2 seconds)
node scripts/testMessageBurst.js perf_test_1500

# Cleanup test data
node scripts/cleanupPerformanceTestData.js

See PERFORMANCE_TEST_RESULTS.md for benchmarks.

📚 AI Features Documentation

1. Thread Summarization

Purpose: Generate concise summaries of long conversations.

How to Use:

Open any conversation
Tap the ✨ AI menu button
Select "Summarize Thread"
Choose message count (10/25/50/100)
View summary with key points

Implementation:

Cloud Function: functions/src/ai/threadSummary.ts
Client: components/SummaryModal.tsx
Caching: 1 hour (invalidates on new messages)

2. Semantic Search

Purpose: Search by meaning, not just keywords.

How to Use:

Open any conversation
Tap the 🔍 search icon
Enter natural language query
View semantically relevant results

Example Queries:

"When is the deadline?" → Finds deadline mentions
"Who's working on the API?" → Finds API discussions
"Budget approval status" → Finds budget-related messages

Implementation:

Embeddings: OpenAI text-embedding-3-small
Vector DB: Pinecone (1536 dimensions)
Background pipeline: Auto-embeds new messages
Cloud Function: functions/src/ai/semanticSearch.ts

3. Action Items Extraction

Purpose: Automatically detect and track tasks from conversations.

How to Use:

Open any conversation
Tap ✨ AI menu → "Extract Action Items"
View detected tasks with:
- Task description
- Assigned person
- Priority (high/medium/low)
- Due date (if mentioned)
- Completion status

Features:

Auto-assignment based on mentions
Priority detection (urgent keywords)
Toggle completion status
Persistent in Firestore

Implementation:

Cloud Function: functions/src/ai/actionItems.ts
Caching: 15 minutes
Client: components/ActionItemsModal.tsx

4. Priority Detection

Purpose: Auto-flag urgent messages for quick triage.

How to Use:

Happens automatically in background
High-priority messages show 🔴 badge
Detects: "urgent", "asap", "critical", deadlines

Implementation:

Background pipeline: Batch analysis every 15 minutes
Quick heuristic + Claude analysis
Firestore: messages/{id}.priority field
Visual: Priority badges in message bubbles

5. Decision Tracking

Purpose: Capture and track key decisions from discussions.

How to Use:

Open any conversation
Tap ✨ AI menu → "Track Decisions"
View captured decisions with:
- Decision statement
- Confidence level
- Context/reasoning
- Timestamp

Implementation:

Cloud Function: functions/src/ai/decisionTracking.ts
Caching: 1 hour
Client: components/DecisionsModal.tsx

6. Smart Meeting Scheduler

Purpose: Analyze scheduling discussions and suggest meeting times.

How to Use:

Have a conversation about scheduling
Tap ✨ AI menu → "Suggest Meeting Times"
View 3 suggested times with reasoning
Tap "Copy All" to paste into calendar app

Features:

Analyzes last 50 messages
Considers mentioned availability
Provides confidence scores
Accounts for constraints

Implementation:

Cloud Function: functions/src/ai/proactiveMeeting.ts
Uses Claude Tool Use API
Client: components/MeetingSchedulerModal.tsx

🛠️ Development

Project Structure

message-ai/
├── app/                      # Expo Router screens
│   ├── (auth)/              # Login, register
│   ├── (tabs)/              # Conversations list, new chat
│   ├── chat/[id].tsx        # Main chat screen
│   └── _layout.tsx          # Root layout (error boundary)
├── components/               # UI components
│   ├── MessageList.tsx      # Inverted FlatList for messages
│   ├── MessageBubble.tsx    # Individual message UI
│   ├── AI modals/           # 6 AI feature modals
│   ├── ErrorBoundary.tsx    # Error handling
│   └── ErrorFallback.tsx    # Error UI components
├── services/                 # Business logic
│   ├── aiService.ts         # Cloud Functions wrapper
│   ├── authService.ts       # Firebase Auth
│   ├── firestoreService.ts  # Firestore operations
│   ├── errorLogger.ts       # AsyncStorage error logging
│   └── failedMessagesService.ts  # Message retry
├── store/                    # Zustand state
│   ├── authStore.ts         # User session
│   └── chatStore.ts         # Conversations list
├── utils/                    # Helpers
│   ├── errorTranslator.ts   # User-friendly errors
│   ├── timeFormat.ts        # Smart date formatting
│   └── constants.ts         # App constants
├── functions/                # Cloud Functions
│   └── src/
│       ├── ai/              # 6 AI feature functions
│       ├── embeddings/      # Background pipeline
│       └── index.ts         # Function exports
└── docs/                     # Documentation
    ├── ARCHITECTURE.md      # Detailed architecture
    ├── phase2-ai-spike/     # AI implementation docs
    └── phase3-full-implementation/  # Final sprint docs

Code Quality

npm run lint          # ESLint
npm run type-check    # TypeScript
npm run validate      # All checks + tests

Git Workflow

Commit messages follow conventional commits
All features tested before commit
Validation passes before push
No force pushes to main

📊 Performance

Benchmarks (1500-message conversation)

Metric	Target	Actual	Status
Message delivery	<200ms	~150ms	✅
Scroll performance	60 FPS	Smooth	✅
Initial render	<500ms	~300ms	✅
Burst handling (10msg/2s)	No jank	Smooth	✅

Optimizations

Inverted FlatList for natural chat scrolling
Pagination (100 messages initially, load more on scroll)
Deduplication for real-time + paginated data
Efficient array operations (sub-millisecond prepend/append)

See PERFORMANCE_TEST_RESULTS.md for details.

🐛 Troubleshooting

Common Issues

"Firebase config missing"

Ensure .env file exists in project root
Variables must start with EXPO_PUBLIC_
Restart dev server after adding .env

"Cloud Functions not working"

Deploy functions: cd functions && firebase deploy --only functions
Check API keys in Firebase Console → Functions → Configuration
View logs: firebase functions:log

"Tests failing"

Clear cache: npm test -- --clearCache
For integration tests: Ensure emulator is running
Check Node version: node -v (should be 20+)

See DEVELOPER_SETUP.md for comprehensive troubleshooting.

📖 Documentation

ARCHITECTURE.md - System design, data flow, security
Phase 2 AI Spike - AI implementation details
Phase 3 Full Implementation - Final sprint PRD
Testing Guide - Integration test documentation
Performance Results - Benchmark data

🚧 Known Limitations (MVP)

Local notifications only (FCM not configured)
No message editing/deletion (intentional for MVP)
No media uploads (text-only for now)
No voice/video calls
English only (no i18n)

See POST_MVP.md for roadmap.

🤝 Contributing

This is an MVP project. For production use, consider:

Implementing FCM for background notifications
Adding message editing/deletion
Media upload support (images, files)
End-to-end encryption
Internationalization (i18n)
Analytics and monitoring

📄 License

MIT License - See LICENSE for details.

🙏 Acknowledgments

Built with:

Expo - React Native framework
Firebase - Backend infrastructure
Anthropic Claude - AI summarization and analysis
OpenAI - Embeddings for semantic search
Pinecone - Vector database

Made with ❤️ for the Gauntlet.ai Founder Fellowship

Name		Name	Last commit message	Last commit date
.vscode		.vscode
app		app
assets/images		assets/images
components		components
docs		docs
functions		functions
hooks		hooks
scripts		scripts
services		services
store		store
styles		styles
types		types
utils		utils
.cursorignore		.cursorignore
.cursorrules		.cursorrules
.firebaserc		.firebaserc
.gitignore		.gitignore
.nvmrc		.nvmrc
README.md		README.md
app.json		app.json
eslint.config.js		eslint.config.js
firebase.config.ts		firebase.config.ts
firebase.json		firebase.json
firestore.indexes.json		firestore.indexes.json
firestore.rules		firestore.rules
jest.config.js		jest.config.js
package-lock.json		package-lock.json
package.json		package.json
tsconfig.json		tsconfig.json

Latest commit

History

Folders and files

Repository files navigation

MessageAI

✨ Key Features

🔥 Core Messaging

🤖 AI-Powered Productivity (6 Features)

🛡️ Production-Ready Error Handling

🏗️ Architecture

🚀 Tech Stack

Frontend

Backend

AI Services

Testing & Quality

📦 Installation & Setup

Prerequisites

1. Clone & Install

2. Configure Environment Variables

3. Deploy Backend

4. Run the App

🧪 Testing

Run All Tests

Unit Tests (Frontend + Backend)

Integration Tests (Firebase Emulator)

Performance Testing

📚 AI Features Documentation

1. Thread Summarization

2. Semantic Search

3. Action Items Extraction

4. Priority Detection

5. Decision Tracking

6. Smart Meeting Scheduler

🛠️ Development

Project Structure

Code Quality

Git Workflow

📊 Performance

Benchmarks (1500-message conversation)

Optimizations

🐛 Troubleshooting

Common Issues

📖 Documentation

🚧 Known Limitations (MVP)

🤝 Contributing

📄 License

🙏 Acknowledgments

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages