Merge pull request #101 from ojusave/feature/zoom-rtms-integration

Add Zoom RTMS + SmythOS SRE Integration Example

Author: alaa-eddine-k

examples/11-zoom-rtms-integration/README.md

@@ -0,0 +1,394 @@
+# Zoom RTMS + SmythOS SRE Integration
+
+This example demonstrates how to integrate **Zoom's Real-Time Media Streaming (RTMS)** with **SmythOS SRE** to create intelligent meeting analysis agents that process live meeting data in real-time.
+
+## Features
+
+- **Real-time Transcript Processing**: Captures and analyzes Zoom meeting transcripts as they happen
+- **AI-Powered Analysis**: Uses SRE agents to extract insights, action items, and key decisions
+- **Vector Database Integration**: Stores transcript segments for semantic search (Pinecone)
+- **Cloud Storage**: Saves meeting summaries and data to AWS S3
+- **Meeting Intelligence**: Generates comprehensive meeting summaries and insights
+- **Scalable Architecture**: Built on SmythOS SRE for enterprise-grade performance
+
+## Three Implementation Options
+
+This example provides three implementation approaches:
+
+### 1. JavaScript Version (`index.js`) - **Recommended**
+- **No TypeScript Issues**: Pure JavaScript implementation that runs immediately
+- **SRE Integration Ready**: Placeholder structure for easy SRE SDK integration
+- **Complete RTMS Integration**: Full Zoom WebSocket and webhook handling
+- **Zero Configuration**: Works with just `npm start`
+- **Production Ready**: Can be easily extended with real SRE functionality
+
+### 2. TypeScript Version (`index.ts`)
+- **Complete AI Agent Integration**: Uses SmythOS SRE agents for advanced transcript analysis
+- **VectorDB Support**: Semantic search and indexing with Pinecone
+- **Cloud Storage**: AWS S3 integration for persistent data storage
+- **Advanced Analytics**: Sophisticated sentiment analysis, action item extraction, and decision tracking
+- **Requires**: SmythOS SRE SDK and additional API keys (may have dependency issues)
+
+### 3. Simple Example (`simple-example.ts`)
+- **Basic Transcript Processing**: Simple keyword-based analysis without external AI dependencies
+- **Lightweight**: Only requires Node.js, Express, and WebSocket libraries
+- **Easy to Understand**: Clear demonstration of Zoom RTMS integration patterns
+- **No External APIs**: Works without OpenAI, Anthropic, Pinecone, or AWS credentials
+- **Perfect for**: Learning, testing, and basic transcript capture
+
+## Architecture
+
+```mermaid
+graph TB
+    A[Zoom Meeting] --> B[RTMS Webhook]
+    B --> C[Express Server]
+    C --> D[SRE Agent]
+    D --> E[LLM Analysis]
+    D --> F[VectorDB Storage]
+    D --> G[Cloud Storage]
+    E --> H[Meeting Insights]
+    F --> I[Transcript Search]
+    G --> J[Meeting Summaries]
+```
+
+## Quick Start
+
+### Prerequisites
+
+1. **Zoom App**: Create a Zoom app with RTMS permissions in the [Zoom Marketplace](https://marketplace.zoom.us/)
+2. **Node.js**: Version 18 or higher
+3. **API Keys**: OpenAI or Anthropic for AI analysis
+4. **Optional**: Pinecone for vector search, AWS S3 for storage
+
+### Installation
+
+1. **Clone and navigate to the example**:
+   ```bash
+   cd sre/examples/11-zoom-rtms-integration
+   ```
+
+2. **Install dependencies**:
+   ```bash
+   npm install
+   ```
+
+3. **Configure environment**:
+   ```bash
+   cp env.example .env
+   # Edit .env with your credentials
+   ```
+
+4. **Start the server**:
+   ```bash
+   # JavaScript version (recommended - no TypeScript issues)
+   npm start
+   
+   # TypeScript version with full SRE integration
+   npm run ts
+   
+   # Simple example without SRE dependencies
+   npm run simple
+   ```
+
+### Environment Configuration
+
+Copy `env.example` to `.env` and configure:
+
+```env
+# Required - Zoom RTMS Configuration
+ZOOM_CLIENT_ID=your_zoom_client_id
+ZOOM_CLIENT_SECRET=your_zoom_client_secret
+ZOOM_SECRET_TOKEN=your_zoom_secret_token
+
+# Required - AI Model (choose one)
+OPENAI_API_KEY=your_openai_api_key
+# OR
+ANTHROPIC_API_KEY=your_anthropic_api_key
+
+# Optional - VectorDB for transcript search
+PINECONE_API_KEY=your_pinecone_api_key
+PINECONE_INDEX_NAME=zoom-meetings
+
+# Optional - Storage for meeting data
+AWS_ACCESS_KEY_ID=your_aws_access_key
+AWS_SECRET_ACCESS_KEY=your_aws_secret_key
+AWS_S3_BUCKET=zoom-meeting-data
+```
+
+## Zoom App Setup
+
+### 1. Create Zoom App
+
+1. Go to [Zoom Marketplace](https://marketplace.zoom.us/)
+2. Create a new **General App** → **User-Managed**
+3. Configure basic information
+
+### 2. Configure RTMS Permissions
+
+**Scopes Required**:
+- `meeting:read`
+- `meeting:write`
+- `rtms:read`
+- `rtms:write`
+
+**Event Subscriptions**:
+- `meeting.rtms_started`
+- `meeting.rtms_stopped`
+- `endpoint.url_validation`
+
+### 3. Webhook Configuration
+
+Set your webhook endpoint URL to:
+```bash
+https://your-domain.com/webhook
+```
+
+For local development, use [ngrok](https://ngrok.com/):
+```bash
+ngrok http 3000
+# Use the HTTPS URL: https://abc123.ngrok.io/webhook
+```
+
+## SRE Agent Capabilities
+
+The integration creates intelligent agents with these skills:
+
+### 1. Real-time Transcript Analysis
+- Extracts key topics and themes
+- Identifies action items and decisions
+- Analyzes participant sentiment
+- Tracks important questions
+
+### 2. Meeting Summarization
+- Generates comprehensive meeting summaries
+- Identifies next steps and follow-ups
+- Highlights key decisions and outcomes
+- Provides participant insights
+
+### 3. Vector Search Integration
+- Indexes transcript segments for semantic search
+- Enables finding specific topics across meetings
+- Supports historical meeting analysis
+
+### 4. Intelligent Storage
+- Saves meeting data to cloud storage
+- Organizes by meeting ID and timestamp
+- Preserves analysis results and summaries
+
+## Usage Examples
+
+### Basic Meeting Analysis
+
+When a Zoom meeting starts with RTMS enabled:
+
+1. **Webhook Triggered**: Server receives `meeting.rtms_started`
+2. **Agent Created**: SRE agent is instantiated for the meeting
+3. **WebSocket Connection**: Connects to Zoom's media stream
+4. **Real-time Processing**: Transcripts are analyzed as they arrive
+5. **Insights Generated**: Key information is extracted and stored
+
+### Example Output
+
+```json
+{
+  "speaker": "John Doe",
+  "timestamp": "2024-01-15T10:30:00Z",
+  "transcript": "Let's schedule a follow-up meeting for next Tuesday to review the project proposal",
+  "analysis": {
+    "topics": ["meeting scheduling", "project review"],
+    "actionItems": ["Schedule follow-up meeting for Tuesday"],
+    "decisions": [],
+    "questions": [],
+    "sentiment": "positive"
+  }
+}
+```
+
+### Meeting Summary
+
+```markdown
+# Meeting Summary - Project Review Call
+
+## Overview
+Team discussion about Q1 project proposal and next steps.
+
+## Key Decisions
+- Approved budget increase for additional resources
+- Selected vendor for cloud infrastructure
+
+## Action Items
+- [ ] John: Schedule follow-up meeting for Tuesday
+- [ ] Sarah: Prepare vendor contract by Friday
+- [ ] Team: Review technical specifications
+
+## Next Steps
+- Technical review session scheduled for next week
+- Final approval expected by month-end
+```
+
+## API Endpoints
+
+### Health Check
+```http
+GET /health
+```
+Returns server status and active meeting count.
+
+### Webhook Endpoint
+```http
+POST /webhook
+```
+Receives Zoom RTMS events and processes meeting data.
+
+## Advanced Configuration
+
+### Custom SRE Configuration
+
+You can customize the SRE initialization for production:
+
+```typescript
+import { SRE } from '@smythos/sre';
+
+// Custom SRE setup for enterprise
+SRE.init({
+    Storage: { 
+        Connector: 'S3', 
+        Settings: { bucket: 'enterprise-meetings' } 
+    },
+    VectorDB: { 
+        Connector: 'Pinecone', 
+        Settings: { indexName: 'company-meetings' } 
+    },
+    Cache: { 
+        Connector: 'Redis', 
+        Settings: { url: 'redis://prod-cluster' } 
+    }
+});
+```
+
+### Custom Agent Behaviors
+
+Modify the agent behavior for specific use cases:
+
+```typescript
+const agent = new Agent({
+    name: 'Sales Meeting Assistant',
+    behavior: `You are a sales meeting specialist. Focus on:
+    - Lead qualification insights
+    - Deal progression indicators  
+    - Customer objections and responses
+    - Follow-up opportunities`,
+    model: 'gpt-4o'
+});
+```
+
+## Monitoring and Debugging
+
+### Logging
+
+Set `LOG_LEVEL=debug` for detailed logs:
+
+```bash
+LOG_LEVEL=debug npm start
+```
+
+### Health Monitoring
+
+Check active meetings and agents:
+
+```bash
+curl http://localhost:3000/health
+```
+
+Response:
+```json
+{
+  "status": "healthy",
+  "activeMeetings": 2,
+  "activeAgents": 2,
+  "timestamp": "2024-01-15T10:30:00Z"
+}
+```
+
+## Troubleshooting
+
+### Common Issues
+
+1. **No Transcript Data**:
+   - Verify RTMS is enabled in Zoom settings
+   - Check webhook URL is accessible
+   - Ensure proper scopes are configured
+
+2. **Agent Creation Fails**:
+   - Verify AI API keys (OpenAI/Anthropic)
+   - Check SRE initialization
+   - Review log output for errors
+
+3. **Storage Issues**:
+   - Verify AWS credentials and permissions
+   - Check S3 bucket exists and is accessible
+   - Ensure proper IAM policies
+
+4. **VectorDB Problems**:
+   - Verify Pinecone API key and index name
+   - Check index dimensions match embedding model
+   - Ensure sufficient Pinecone quota
+
+### Debug Commands
+
+```bash
+# Test webhook endpoint
+curl -X POST http://localhost:3000/webhook \
+  -H "Content-Type: application/json" \
+  -d '{"event": "endpoint.url_validation", "payload": {"plainToken": "test"}}'
+
+# Check server health
+curl http://localhost:3000/health
+```
+
+## Security Considerations
+
+- Store sensitive credentials in environment variables
+- Use HTTPS for webhook endpoints in production
+- Implement proper authentication for API endpoints
+- Regular rotation of API keys and tokens
+- Monitor access logs and usage patterns
+
+## Scaling for Production
+
+### Performance Optimization
+
+- Use Redis for caching meeting data
+- Implement connection pooling for databases
+- Use load balancers for multiple instances
+- Monitor memory usage and optimize accordingly
+
+### Enterprise Features
+
+- Multi-tenant isolation with SRE scoping
+- Custom storage backends (enterprise databases)
+- Advanced security with HashiCorp Vault
+- Custom LLM models and fine-tuning
+
+## Contributing
+
+This example is part of the SmythOS SRE ecosystem. Contributions welcome!
+
+1. Fork the repository
+2. Create a feature branch
+3. Add your improvements
+4. Submit a pull request
+
+## License
+
+MIT License - see the main SRE repository for details.
+
+## Support
+
+- [SmythOS Documentation](https://smythos.github.io/sre/)
+- [Discord Community](https://discord.gg/smythos)
+- [GitHub Issues](https://github.com/SmythOS/sre/issues)
+
+---
+
+## Built using SmythOS SRE - The Operating System for Agentic AI