OpenCAP AI Chat Studio

A production-ready AI chat interface built on the SAP Cloud Application Programming Model (CAP) and SAPUI5, designed for enterprise deployment with SAP AI Core integration. This open-source project provides a complete foundation for building scalable, multi-model AI assistants on SAP Business Technology Platform.

Overview

OpenCAP AI Chat Studio is an enterprise-grade conversational AI platform that seamlessly integrates with SAP AI Core, enabling organizations to deploy and manage AI-powered chat interfaces with full control over model selection, configuration, and conversation management.

Key Features

AI Core Integration

• Native SAP AI Core Support: Direct integration with SAP AI Core deployment endpoints
• Flexible Authentication: JWT token-based authentication with configurable resource groups
• Multi-Model Architecture: Support for multiple AI models with per-conversation model selection
• Dynamic Configuration: Runtime configuration of AI endpoints without code changes

Enterprise Features

• Conversation Management: Persistent conversation history with full CRUD operations
• Model Configuration: Centralized management of AI model deployments and system prompts
• User Interface: Professional SAPUI5-based interface with responsive design
• Theme Support: Light, dark, and system-synchronized themes
• Message Operations: Copy, edit, and delete message capabilities

Technical Architecture

• SAP CAP Backend: Robust service layer following SAP Clean Core principles
• SAPUI5 Frontend: Modern, responsive UI built with SAP's enterprise UI framework
• RESTful API: OData v4 services for seamless integration
• Database Flexibility:
  • Development: SQLite for local development
  • Production: SAP HANA Cloud for enterprise deployment
  • Automatic database switching based on environment
• Streaming Display: Progressive message rendering for enhanced user experience

Prerequisites

Required

• Node.js: v18 or higher (v20 LTS recommended)
• SAP CDS Development Kit: npm install -g @sap/cds-dk
• Visual Studio Build Tools (Windows): Required for SQLite native module compilation

For Production Deployment

• SAP BTP Account: Cloud Foundry environment
• SAP HANA Cloud: Database service (hdi-shared plan minimum)
• Cloud Foundry CLI: For deployment (cf command)
• MTA Build Tool: For multi-target application builds (mbt command)

Optional

• SAP AI Core Access: For production AI model integration
• SAP Destination Service: For secure credential management

Installation

1. Clone the Repository

git clone https://github.com/OpenCAPai/opencapai-chat-studio.git
cd opencapai-chat-studio

2. Install Dependencies

npm install

Note for Windows users: If you encounter native module compilation errors, install Visual Studio Build Tools first.

3. Configure AI Core (Optional)

Create a .env file in the project root:

AICORE_ENDPOINT="https://api.ai.prod-eu20.<REGION>.azure.ml.hana.ondemand.com/v2/inference/deployments/<YOUR_DEPLOYMENT_ID>/v2/completion"
AICORE_RESOURCE_GROUP="default"
AICORE_TOKEN="your-jwt-token-here"

Alternatively, configure AI Core credentials through the application's Settings UI after startup.

4. Start the Application

npm run watch

The application will be available at:

• UI: http://localhost:4004/webapp/index.html
• API: http://localhost:4004/odata/v4/chat

Configuration

AI Core Setup

1. Access Settings: Click the settings icon (⚙️) in the application header
2. Configure AI Core:
   • Endpoint URL: Your SAP AI Core deployment endpoint
   • Resource Group: The resource group name (typically "default")
   • Authentication Token: Your JWT token for AI Core authentication
3. Save Configuration: Click "Save AI Core Configuration"

Model Management

Configure AI models in the Settings dialog:

• Model ID: Unique identifier for the model
• Display Name: User-friendly name shown in the UI
• Deployment URL: SAP AI Core deployment endpoint
• System Prompt: Base instructions defining the AI's behavior and personality

Project Structure

opencapai-chat-studio/
├── app/                    # SAPUI5 Frontend Application
│   ├── webapp/
│   │   ├── controller/     # UI Controllers
│   │   ├── view/           # XML Views
│   │   └── Component.js    # Application Component
│   └── index.cds           # UI Annotations
├── db/                     # Database Schema
│   └── schema.cds          # Entity Definitions
├── srv/                    # CAP Service Layer
│   ├── chat-service.cds    # Service Definitions
│   └── chat-service.js     # Service Implementation
├── .env                    # Environment Configuration (not in repo)
├── package.json            # Dependencies and Scripts
└── README.md               # This file

Development

Local Development

npm run watch

Starts the application with hot-reload enabled for both backend and frontend changes.

Production Build

npm run build

Testing

npm test

Database Configuration

The application automatically adapts to the environment:

Local Development (SQLite)

# Default configuration - uses SQLite
npm run watch

Local Development with HANA Cloud

# Deploy schema to HANA
cds deploy --to hana --store-credentials

# Start with production profile
NODE_ENV=production npm start

Production (SAP HANA Cloud)

The application automatically uses HANA when deployed to SAP BTP Cloud Foundry with the HANA service bound.

Deployment

Quick Deployment with MTA

The recommended deployment method uses Multi-Target Application (MTA) for automated service provisioning:

# Install MTA Build Tool
npm install -g mbt

# Build MTA archive
mbt build

# Deploy to Cloud Foundry
cf deploy mta_archives/opencapai-chat-studio_*.mtar

This automatically:

• Creates and configures SAP HANA HDI container
• Deploys database schema
• Sets up XSUAA authentication
• Deploys the application
• Configures all service bindings

Manual Deployment

For step-by-step manual deployment:

1. Create Services:

   cf create-service hana hdi-shared opencapai-chat-studio-db
   cf create-service xsuaa application opencapai-chat-studio-uaa -c xs-security.json

2. Build and Deploy:

   npm run build
   cf push

Detailed Deployment Guide

For comprehensive deployment instructions including:

• HANA Cloud setup
• Environment configuration
• Security setup
• Monitoring and troubleshooting

API Documentation

OData Services

• GET /odata/v4/chat/Conversations - List all conversations
• GET /odata/v4/chat/Messages - List all messages
• POST /odata/v4/chat/sendMessage - Send a message and receive AI response
• GET /odata/v4/chat/getAICoreConfig - Retrieve current AI Core configuration
• POST /odata/v4/chat/saveAICoreConfig - Update AI Core configuration

For complete API documentation, access the service metadata at: http://localhost:4004/odata/v4/chat/$metadata

Contributing

We welcome contributions from the community !

Development Guidelines

• Follow SAP CAP best practices
• Maintain code quality with ESLint
• Write comprehensive tests for new features
• Update documentation for API changes

Security

• Authentication: Implement proper authentication mechanisms for production use
• Token Management: Store AI Core tokens securely using SAP Destination Service or credential stores
• Data Privacy: Ensure compliance with data protection regulations (GDPR, etc.)
• API Security: Use HTTPS in production and implement rate limiting

Troubleshooting

Common Issues

Module compilation errors on Windows

• Install Visual Studio Build Tools 2022

AI Core connection failures

• Verify endpoint URL and token validity
• Check network connectivity and firewall settings
• Ensure resource group name is correct

Database errors (Local)

• Delete node_modules and run npm install again
• Check SQLite installation
• Ensure Visual Studio Build Tools are installed (Windows)

Database errors (Production)

• Verify HANA service is bound: cf services
• Check database deployer logs: cf logs opencapai-chat-studio-db-deployer --recent

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Acknowledgments

Built with:

• SAP Cloud Application Programming Model
• SAPUI5
• SAP AI Core

Support

• Issues: Report bugs and request features via GitHub Issues
• Discussions: Join the community in GitHub Discussions
• Documentation: Comprehensive guides in the repository root

---

OpenCAP AI Chat Studio - Enterprise AI Chat Platform for SAP Ecosystems