LLM Provider - SAP AI Core Integration for VSCode

Integrate SAP AI Core models seamlessly with VSCode's Language Model API for enhanced AI-powered development.

🚀 Features

Core Capabilities

🤖 AI Model Integration: Access SAP AI Core foundation models directly within VSCode
💬 Chat Provider: Native integration with VSCode's chat interface
🔄 Real-time Streaming: Support for streaming responses from AI models
🔐 Secure Credential Management: Encrypted storage of SAP AI Core credentials
📊 Token Estimation: Accurate token counting for cost management
🎯 Multi-Scenario Support: Foundation models and orchestration scenarios

Advanced Features

⚡ Version Compatibility: Automatic detection and adaptation to different VSCode API versions
🔧 Diagnostic Tools: Built-in API compatibility checking and troubleshooting
📝 Comprehensive Logging: Configurable logging levels for debugging and monitoring
🔄 Model Cache Management: Efficient caching with manual refresh capabilities
🎛️ Configuration Management: Flexible scenario switching and model selection

📦 Installation

From VSCode Marketplace

Open VSCode
Go to Extensions (Ctrl+Shift+X / Cmd+Shift+X)
Search for "LLM Provider"
Click Install

Manual Installation

Download the latest .vsix file from Releases
Open VSCode
Run Extensions: Install from VSIX... from Command Palette
Select the downloaded .vsix file

⚙️ Configuration

1. SAP AI Core Credentials Setup

The extension requires SAP AI Core credentials to function. Set up your credentials using the Command Palette:

Open Command Palette (Ctrl+Shift+P / Cmd+Shift+P)
Run SAP AI Core: Manage Credentials
Select Set/Update Credentials
Enter your credential JSON:

{
  "clientid": "your-client-id",
  "clientsecret": "your-client-secret",
  "url": "https://your-auth-url",
  "aiApiUrl": "https://your-ai-api-url",
  "resourceGroup": "your-resource-group"
}

2. Scenario Configuration

Configure the AI scenario in your VSCode settings:

{
  "sapAiCore.scenarioId": "foundation-models"
}

Available scenarios:

foundation-models - Direct access to foundation models (default)
orchestration - Advanced orchestration capabilities (future)

3. Logging Configuration

Control logging verbosity for debugging:

{
  "sapAiCore.logLevel": "info"
}

Log levels: debug | info | warn | error

🎯 Usage

Basic Chat Integration

Open VSCode Chat: Use Ctrl+Alt+I / Cmd+Alt+I
Select SAP AI Core: Choose from the model dropdown
Start Chatting: Ask questions or request code assistance

Command Palette Actions

Command	Description
`SAP AI Core: Manage Credentials`	Set up or update your SAP AI Core credentials
`SAP AI Core: Refresh Models`	Clear model cache and reload available models
`SAP AI Core: Check API Compatibility`	View compatibility diagnostics
`SAP AI Core: Test API Compatibility`	Run compatibility detection tests

Credential Management

View Status: Check which credentials are configured
Update Credentials: Modify existing credential configuration
Remove Credentials: Securely delete stored credentials
Legacy Cleanup: Remove old API key formats

🔧 Development

Prerequisites

Node.js 18+
VSCode 1.103.0+
TypeScript 5.8+

Setup

# Clone the repository
git clone https://github.com/anjanpoonacha/llmprovider.git
cd llmprovider

# Install dependencies
npm install

# Compile TypeScript
npm run compile

# Run tests
npm test

# Package extension
npm run package

Project Structure

src/
├── commands/           # VSCode commands
├── config/            # Configuration and constants
├── core/              # Core extension logic
├── providers/         # Language model providers
├── services/          # Business logic services
├── strategies/        # AI scenario strategies
├── types/             # TypeScript definitions
└── utils/             # Utility functions

Architecture Highlights

🏗️ Modular Design: Clean separation of concerns with dependency injection
🔄 Adapter Pattern: Support for multiple VSCode API versions
🎯 Strategy Pattern: Flexible AI scenario handling
📊 Universal Provider: Unified interface for different model types
🔍 Compatibility System: Automatic API version detection and adaptation

🐛 Troubleshooting

Common Issues

Extension not loading models:

Check credentials: SAP AI Core: Manage Credentials → View Credential Status
Verify API compatibility: SAP AI Core: Check API Compatibility
Refresh models: SAP AI Core: Refresh Models

Authentication errors:

Verify your SAP AI Core credentials are correct
Check network connectivity to SAP AI Core endpoints
Ensure resource group has proper permissions

Performance issues:

Adjust log level to warn or error in production
Clear model cache if experiencing stale data
Check VSCode version compatibility

Debug Mode

Enable detailed logging for troubleshooting:

{
  "sapAiCore.logLevel": "debug"
}

View logs in: Output → SAP AI Core Language Model Provider

📋 Requirements

System Requirements

VSCode: Version 1.103.0 or higher
Node.js: Version 18+ (for development)
Operating System: Windows, macOS, or Linux

SAP AI Core Requirements

Valid SAP AI Core subscription
Configured service instance with appropriate permissions
Network access to SAP AI Core endpoints

🤝 Contributing

We welcome contributions! Please see our Contributing Guidelines for details.

Development Workflow

Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Commit your changes: git commit -m 'Add amazing feature'
Push to the branch: git push origin feature/amazing-feature
Open a Pull Request

Code Standards

Follow TypeScript strict mode guidelines
Use ESLint configuration provided
Write tests for new features
Update documentation as needed

📄 License

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

LLM Provider

anjanpoonacha