An intelligent WhatsApp chatbot for managing personal notes using AI-powered content processing. Ikigai serves as your second brain, automatically organizing and retrieving information through natural conversation.
Ikigai Chat transforms WhatsApp into a powerful note-taking system that understands context, extracts meaningful information, and stores everything in a structured Notion database. Simply send messages to the bot and it will intelligently categorize your thoughts, recipes, links, tasks, and ideas.
Key Features:
- Natural language processing for automatic content classification
- Intelligent tag assignment with support for multiple tags per note
- Advanced search capabilities with fuzzy matching and relevance scoring
- Seamless Notion integration for structured data storage
- Conversational interface with context awareness
- Support for ambiguous content with clarification requests
The system follows a clean layered architecture:
WhatsApp User → Baileys WebSocket → Express Backend → OpenAI API → Notion Database
Core Components:
- Communication Layer: WhatsApp integration via Baileys, message parsing and response formatting
- Processing Layer: Intent classification using OpenAI, content extraction and database query processing
- Data Layer: Notion API client, database schema management and intelligent search algorithms
Built on the solid foundation of the Baileys Starter boilerplate, Ikigai Chat extends it with advanced AI capabilities and database integration.
Backend: Node.js with TypeScript and Express WhatsApp Integration: Baileys library for WhatsApp Web API AI Processing: OpenAI GPT for content analysis and intent classification Database: Notion API for structured note storage Logging: Pino for structured logging with metadata support Testing: Jest for unit and integration testing
Clone the repository and install dependencies:
git clone <repository-url>
cd ikigai-chat
npm installCreate a .env file based on .env.example:
# Server Configuration
PORT=3000
BOT_NAME="Ikigai Bot"
LOG_LEVEL=info
SESSION_NAME=ikigai_session
# AI Configuration
OPENAI_API_KEY=your_openai_api_key
AI_ENABLED=true
AI_SYSTEM_PROMPT="You are Ikigai, an intelligent assistant for note management."
# Notion Integration
NOTION_API_KEY=your_notion_integration_token
NOTION_DATABASE_ID=your_notion_database_idStart the bot in development mode:
npm run devFor production deployment:
npm run build
npm startThe bot will generate a QR code for WhatsApp Web authentication. Scan it with your phone to connect.
Saving Notes: Send any message and Ikigai will automatically extract a title, organize the content, and assign appropriate tags. For example, sending "Pasta recipe: boil pasta, sauté garlic, add tomatoes" creates a structured recipe note.
Searching Notes: Ask natural questions like "What recipes do I have?" or "Show me notes about projects" to retrieve relevant information with intelligent ranking.
Clarification: When content is ambiguous, Ikigai asks clarifying questions to ensure proper categorization.
Tag Management: The system supports multiple tags per note and can suggest tags based on similar existing content.
src/
├── ai/ # OpenAI integration and prompt management
├── config/ # Environment configuration and settings
├── handlers/ # WhatsApp message processing and routing
├── logger/ # Structured logging with Pino
├── services/ # Core business logic (Notion, intent classification)
├── routes/ # Express API routes
├── server/ # Express server setup
├── socket/ # Baileys WebSocket connection management
├── store/ # In-memory state management
├── types/ # TypeScript type definitions
└── index.ts # Application entry point
- WhatsApp Integration: Full Baileys implementation with session management
- OpenAI Integration: GPT-powered content analysis and intent classification
- Notion Integration: Structured note storage with multiple tag support
- Intelligent Search: Multi-level search with exact, synonym, and fuzzy matching
- Conversational Context: State management for natural dialogue flow
- Enhanced Conversational Responses: More natural and contextual bot interactions
- Advanced Content Processing: Better handling of complex note structures
- User Authentication: Multi-user support with secure session management
- Database Customization: Dynamic schema creation and field configuration
- Voice Message Support: Audio transcription and processing for WhatsApp voice notes
- Calendar Integration: Automatic event extraction and scheduling capabilities
- Multi-Language Support: Intelligent language detection and localized responses
- Response Personalization: User-specific communication styles and preferences
- Advanced Analytics: Usage patterns and note organization insights
- Webhook Notifications: Real-time updates and external integrations
We welcome contributions to Ikigai Chat. Please read our contributing guidelines and ensure all tests pass before submitting pull requests.
This project is licensed under the MIT License. See the LICENSE file for details.
Built upon the excellent Baileys Starter boilerplate by Volt-Chat, which provided the foundational WhatsApp integration and project structure.