MongoDB Adapter for ElizaOS

A robust MongoDB adapter for ElizaOS that provides persistence, vector search capabilities, and caching functionality.

Features

• Full MongoDB database support with connection pooling
• Vector search capabilities for efficient similarity searches
• Automatic fallback to standard search when vector search is unavailable
• Built-in caching system with TTL support
• Sharding support for better performance at scale
• Comprehensive memory and knowledge management
• Relationship tracking between users
• Goal tracking and management
• Participant and room management

Prerequisites

• MongoDB 6.0 or later (recommended for vector search support)
• NodeJS 16.0 or later
• ElizaOS installation

Installation

npm install @elizaos-plugins/adapter-mongodb

Configuration

Add the adapter to your ElizaOS configuration:

{
  "plugins": ["@elizaos-plugins/adapter-mongodb"],
  "settings": {
    "MONGODB_CONNECTION_STRING": "your_mongodb_connection_string",
    "MONGODB_DATABASE": "your_database_name"  // Optional, defaults to "elizaAgent"
  }
}

Required Environment Variables

• MONGODB_CONNECTION_STRING: Your MongoDB connection string
• MONGODB_DATABASE (optional): Database name to use

Connection Options

The adapter is configured with optimal connection settings:

• Maximum pool size: 100 connections
• Minimum pool size: 5 connections
• Connection timeout: 10 seconds
• Socket timeout: 45 seconds
• Retry support for both reads and writes
• Compression enabled (zlib)

Features in Detail

Vector Search

The adapter automatically detects and enables vector search capabilities if your MongoDB instance supports it. This provides efficient similarity searches for:

• Memory retrieval
• Knowledge base searches
• Semantic similarity matching

If vector search is unavailable, the adapter automatically falls back to standard search methods.

Caching

Built-in caching system with:

• 24-hour TTL by default
• Automatic cache invalidation
• Memory-efficient storage
• Cache hit/miss optimizations

Memory Management

Comprehensive memory management features:

• CRUD operations for memories
• Vector-based similarity search
• Batch processing support
• Automatic uniqueness checking

Knowledge Base

Robust knowledge base management:

• Support for shared and private knowledge
• Vector-based knowledge retrieval
• Metadata support
• Chunk management for large content

Development

Running Tests

The test suite uses Docker for running a test MongoDB instance:

cd src/__tests__
./run_tests.sh

This will:

1. Start a MongoDB container
2. Run the test suite
3. Clean up resources automatically

Contributing

1. Fork the repository
2. Create a feature branch
3. Add tests for new functionality
4. Submit a pull request

Error Handling

The adapter includes comprehensive error handling:

• Connection failure recovery
• Automatic reconnection
• Graceful degradation of vector search
• Detailed error logging

Performance Considerations

• Uses connection pooling for optimal performance
• Implements efficient batch processing
• Supports sharding for horizontal scaling
• Includes index optimization