@neurodevworks/chatbot-core

Core AI client services and utilities for chatbot functionality in Angular applications.

🚀 Features

• Multi-AI Provider Support: OpenAI, Claude, Gemini, and more
• Streaming Responses: Real-time token streaming with RxJS observables
• TypeScript First: Full type safety and IntelliSense support
• Reactive Architecture: Built with RxJS for reactive programming
• Error Handling: Comprehensive error handling and retry logic
• Configurable: Flexible configuration options for different AI providers

📦 Installation

npm install @neurodevworks/chatbot-core

🛠️ Quick Start

Basic Usage

import { AIClientService } from '@neurodevworks/chatbot-core';

// Initialize the service
const aiClient = new AIClientService();

// Configure with OpenAI
aiClient.initialize({
  provider: 'openai',
  apiKey: 'your-openai-api-key',
  model: 'gpt-3.5-turbo'
});

// Send a message
const response = await aiClient.sendMessage('Hello, AI!');
console.log(response);

Streaming Responses

// Send streaming message
await aiClient.sendStreamingMessage('Tell me a story');

// Subscribe to token stream
aiClient.tokenStream$.subscribe(token => {
  console.log('New token:', token);
});

// Subscribe to completion
aiClient.responseComplete$.subscribe(fullResponse => {
  console.log('Complete response:', fullResponse);
});

Angular Integration

import { Injectable } from '@angular/core';
import { AIClientService, AIClientOptions } from '@neurodevworks/chatbot-core';

@Injectable({
  providedIn: 'root'
})
export class ChatService {
  constructor(private aiClient: AIClientService) {}

  async initializeChat(apiKey: string) {
    const options: AIClientOptions = {
      provider: 'openai',
      apiKey,
      model: 'gpt-3.5-turbo'
    };
    
    this.aiClient.initialize(options);
  }

  async sendMessage(message: string) {
    return await this.aiClient.sendMessage(message, {
      temperature: 0.7,
      maxTokens: 1000
    });
  }
}

🎯 API Reference

AIClientService

Methods

• initialize(options: AIClientOptions): void - Initialize the AI client
• sendMessage(message: string, options?: MessageOptions): Promise<string> - Send a message
• sendStreamingMessage(message: string, options?: MessageOptions): Promise<void> - Send streaming message
• clearMessages(): void - Clear message history
• isInitialized: boolean - Check if client is initialized

Observables

• tokenStream$: Observable<string> - Stream of tokens during response generation
• responseComplete$: Observable<string> - Complete response when generation finishes

Types

interface AIClientOptions {
  provider: 'openai' | 'claude' | 'gemini';
  apiKey: string;
  model: string;
  baseUrl?: string;
}

interface MessageOptions {
  temperature?: number;
  maxTokens?: number;
  systemPrompt?: string;
}

🔧 Configuration

OpenAI Configuration

const options: AIClientOptions = {
  provider: 'openai',
  apiKey: 'sk-...',
  model: 'gpt-3.5-turbo' // or 'gpt-4'
};

Claude Configuration

const options: AIClientOptions = {
  provider: 'claude',
  apiKey: 'sk-ant-...',
  model: 'claude-3-sonnet-20240229'
};

Gemini Configuration

const options: AIClientOptions = {
  provider: 'gemini',
  apiKey: 'AI...',
  model: 'gemini-pro'
};

🤝 Contributing

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

📄 License

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

🔗 Related Packages

• @neurodevworks/angular-chatbot - Angular UI components
• Angular Chatbot AI - Complete monorepo

👨‍💻 Author

Manoj Kumar

• Email: [email protected]
• GitHub: @neurodevworks
• NPM: @neurodevworks

Made with ❤️ by NeuroDevWorks