# Azure AI MCP Server

A mission-critical Model Context Protocol (MCP) server providing comprehensive Azure AI services integration with enterprise-grade reliability, observability, and chaos engineering capabilities.

## 🚀 Features

### Core Azure AI Services
- **Azure OpenAI**: Chat completions, embeddings, and text generation
- **Cognitive Services Text Analytics**: Sentiment analysis, entity recognition, key phrase extraction
- **Computer Vision**: Image analysis, object detection, OCR
- **Face API**: Face detection, recognition, and analysis
- **Azure Storage**: Blob storage integration for data persistence

### Mission-Critical Capabilities
- **High Availability**: Multi-region deployment with automatic failover
- **Observability**: Comprehensive logging, metrics, and distributed tracing
- **Security**: Azure AD integration, API key management, and encryption at rest/transit
- **Rate Limiting**: Intelligent throttling and backpressure handling
- **Retry Logic**: Exponential backoff with circuit breaker patterns
- **Chaos Engineering**: Built-in chaos testing with Azure Chaos Studio

### DevOps & CI/CD
- **Infrastructure as Code**: Terraform modules for all environments
- **Multi-Environment**: Integration, E2E, and Production pipelines
- **Container Support**: Docker containerization with health checks
- **Monitoring**: Azure Monitor, Application Insights integration
- **Security Scanning**: Automated vulnerability assessments

## 📋 Prerequisites

- Node.js 18+ 
- Azure subscription with appropriate permissions
- Azure CLI installed and configured
- Terraform 1.5+
- Docker (optional, for containerized deployment)

## 🔧 Installation

### 1. Clone and Setup
```bash
git clone https://github.com/caiotk/nexguideai-azure-ai-mcp-server.git
cd azure-ai-mcp-server
npm install
```

### 2. Environment Configuration
Copy the environment template and configure your Azure credentials:
```bash
cp .env.example .env
```

Required environment variables:
```env
# Azure OpenAI
AZURE_OPENAI_ENDPOINT=https://your-openai.openai.azure.com/
AZURE_OPENAI_API_KEY=your-api-key

# Azure Cognitive Services
AZURE_COGNITIVE_SERVICES_ENDPOINT=https://your-region.api.cognitive.microsoft.com/
AZURE_COGNITIVE_SERVICES_KEY=your-key

# Azure Storage
AZURE_STORAGE_CONNECTION_STRING=your-connection-string

# Azure AD (for production)
AZURE_TENANT_ID=your-tenant-id
AZURE_CLIENT_ID=your-client-id
AZURE_CLIENT_SECRET=your-client-secret

# Monitoring
AZURE_APPLICATION_INSIGHTS_CONNECTION_STRING=your-connection-string
LOG_LEVEL=info
```

### 3. Build and Run
```bash
npm run build
npm start
```

## 🏗️ Architecture

### System Overview
```
┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   MCP Client    │────│  Azure AI MCP   │────│  Azure Services │
│                 │    │     Server      │    │                 │
└─────────────────┘    └─────────────────┘    └─────────────────┘
                              │
                              ▼
                    ┌─────────────────┐
                    │   Observability │
                    │   & Monitoring  │
                    └─────────────────┘
```

### Component Architecture
- **API Layer**: MCP protocol implementation with request validation
- **Service Layer**: Azure AI service integrations with retry logic
- **Infrastructure Layer**: Terraform modules for cloud resources
- **Observability Layer**: Logging, metrics, and distributed tracing

## 🔄 CI/CD Pipeline

### Environments
1. **Integration (INT)**: Feature testing and integration validation
2. **End-to-End (E2E)**: Full system testing with chaos engineering
3. **Production (PROD)**: Live environment with blue-green deployment

### Pipeline Stages
```yaml
Build → Test → Security Scan → Deploy INT → E2E Tests → Chaos Tests → Deploy PROD
```

### Deployment Strategy
- **Blue-Green Deployment**: Zero-downtime deployments
- **Canary Releases**: Gradual traffic shifting for risk mitigation
- **Automated Rollback**: Automatic rollback on health check failures

## 🧪 Testing Strategy

### Test Pyramid
- **Unit Tests**: Individual component testing (Jest)
- **Integration Tests**: Service integration validation
- **E2E Tests**: Full workflow testing
- **Chaos Tests**: Resilience and failure scenario testing

### Chaos Engineering
Integrated with Azure Chaos Studio for:
- **Service Disruption**: Simulated Azure service outages
- **Network Latency**: Increased response times
- **Resource Exhaustion**: CPU/Memory pressure testing
- **Dependency Failures**: External service failures

## 📊 Monitoring & Observability

### Metrics
- **Performance**: Response times, throughput, error rates
- **Business**: API usage, feature adoption, cost optimization
- **Infrastructure**: Resource utilization, availability

### Logging
- **Structured Logging**: JSON format with correlation IDs
- **Log Levels**: ERROR, WARN, INFO, DEBUG
- **Centralized**: Azure Log Analytics integration

### Alerting
- **SLA Monitoring**: 99.9% availability target
- **Error Rate Thresholds**: >1% error rate alerts
- **Performance Degradation**: Response time anomalies

## 🔒 Security

### Authentication & Authorization
- **Azure AD Integration**: Enterprise identity management
- **API Key Management**: Secure key rotation and storage
- **RBAC**: Role-based access control

### Data Protection
- **Encryption at Rest**: Azure Storage encryption
- **Encryption in Transit**: TLS 1.3 for all communications
- **Data Residency**: Configurable data location compliance

### Security Scanning
- **Dependency Scanning**: Automated vulnerability detection
- **SAST**: Static application security testing
- **Container Scanning**: Docker image vulnerability assessment

## 🚀 Deployment

### Local Development
```bash
npm run dev
```

### Docker Deployment
```bash
npm run docker:build
npm run docker:run
```

### Terraform Deployment
```bash
cd terraform/environments/prod
terraform init
terraform plan
terraform apply
```

## 📈 Performance

### Benchmarks
- **Latency**: P95 < 500ms for chat completions
- **Throughput**: 1000+ requests/minute sustained
- **Availability**: 99.9% uptime SLA

### Optimization
- **Connection Pooling**: Efficient Azure service connections
- **Caching**: Intelligent response caching strategies
- **Rate Limiting**: Adaptive throttling based on service limits

## 🤝 Contributing

1. Fork the repository
2. Create a feature branch (`git checkout -b feature/amazing-feature`)
3. Commit your changes (`git commit -m 'Add amazing feature'`)
4. Push to the branch (`git push origin feature/amazing-feature`)
5. Open a Pull Request

### Development Guidelines
- Follow TypeScript strict mode
- Maintain 90%+ test coverage
- Use conventional commits
- Update documentation for new features

## 📄 License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

## 🆘 Support

- **Documentation**: [docs/](./docs/)
- **Issues**: [GitHub Issues](https://github.com/caiotk/nexguideai-azure-ai-mcp-server/issues)
- **Discussions**: [GitHub Discussions](https://github.com/caiotk/nexguideai-azure-ai-mcp-server/discussions)

## 🗺️ Roadmap

- [ ] Multi-model support (GPT-4, Claude, Gemini)
- [ ] Advanced caching strategies
- [ ] GraphQL API support
- [ ] Kubernetes deployment manifests
- [ ] Advanced chaos engineering scenarios
- [ ] Cost optimization recommendations

---

**Built with ❤️ by NexGuide AI**