MongoDB Memory Bank MCP Server

🚀 The world's first MongoDB-powered MCP server with hybrid search capabilities

Transform your AI coding workflow with lightning-fast memory management, semantic search, and MongoDB's cutting-edge $rankFusion technology.

⚡ Game-Changing Features

🔥 MongoDB $rankFusion Hybrid Search (8.1+)

• World's first MCP implementation of MongoDB's revolutionary $rankFusion
• Combines text + vector search in a single query using reciprocal rank fusion
• 10-100x faster than traditional file-based memory systems
• Semantic understanding with Voyage AI's state-of-the-art embeddings

🎯 Intelligent Memory Management

• Auto-tagging with AI-powered content analysis
• Related memory discovery finds connections you never knew existed
• Sub-second search across thousands of memories
• Rich metadata with word counts, timestamps, and analytics

🌟 Dual-Mode Architecture

• Atlas Mode: Full hybrid search with vector embeddings
• Community Mode: Lightning-fast text search and document storage
• Seamless fallback ensures compatibility across MongoDB versions

🚀 Quick Start

Installation

npm install -g mongodb-memory-bank-mcp

Atlas Setup (Recommended - Full Features)

# Interactive setup with MongoDB Atlas
npx mongodb-memory-bank-mcp setup:atlas

Local Setup (Community Edition)

# Quick local setup with MongoDB Community
npx mongodb-memory-bank-mcp setup:local

🛠 MCP Client Configuration

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "memory-bank-mongodb": {
      "command": "npx",
      "args": ["-y", "mongodb-memory-bank-mcp"],
      "env": {
        "MONGODB_URI": "your_mongodb_connection_string",
        "MONGODB_DATABASE": "memory_bank",
        "VOYAGE_API_KEY": "your_voyage_api_key"
      }
    }
  }
}

Cursor / Windsurf / VS Code

Configure in your MCP settings with the same environment variables.

🎯 MCP Tools

Core Memory Operations

• list_projects - List all projects
• list_project_files - List files in a project
• memory_bank_read - Read memory content
• memory_bank_write - Create new memory
• memory_bank_update - Update existing memory

Enhanced MongoDB Features

• memory_search - Hybrid text/semantic search
• memory_discover - Find related memories

💡 Usage Examples

Store with Auto-Tagging

Store this authentication strategy:
"JWT implementation with refresh tokens, Redis session store, rate limiting, and brute force protection."

Result: Automatically tagged with auth, jwt, security, redis, performance

Hybrid Search (Atlas)

Search for "database performance optimization" using semantic search

Result: Finds related memories about indexing, query optimization, caching strategies

Discover Related Memories

Find memories related to my auth-strategy.md file

Result: Discovers security patterns, session management, and API design memories

🏗 Architecture

MongoDB-First Design

• Single source of truth: All data in MongoDB
• ACID transactions: Data consistency guaranteed
• Horizontal scaling: Grows with your needs
• Rich indexing: Optimized for search performance

Performance Benchmarks

🔧 Configuration

Environment Variables

# Required
MONGODB_URI=mongodb://localhost:27017
MONGODB_DATABASE=memory_bank

# Atlas Features (Optional)
MONGODB_ATLAS=true
ENABLE_VECTOR_SEARCH=true
VOYAGE_API_KEY=your_voyage_api_key

MongoDB Atlas Setup

1. Create Atlas Cluster (Free tier available)
2. Enable Vector Search in cluster settings
3. Get Voyage AI API Key for embeddings
4. Configure connection string with credentials

MongoDB Community Setup

1. Install MongoDB Community locally
2. Start MongoDB service
3. Use local connection string
4. Enjoy core features without vector search

🌟 Why MongoDB Over Files?

Performance Revolution

• Instant search vs slow file scanning
• Concurrent access vs file locking
• Rich queries vs basic grep
• Scalable storage vs linear degradation

Advanced Capabilities

• $rankFusion hybrid search (MongoDB 8.1+)
• Vector embeddings with Voyage AI
• Aggregation pipelines for complex analytics
• Real-time indexing for optimal performance

Developer Experience

• Drop-in replacement for existing memory banks
• Backward compatibility with all original tools
• Enhanced features without breaking changes
• Production-ready with enterprise-grade reliability

🚀 Latest Technologies

MongoDB $rankFusion (8.1+)

• Reciprocal rank fusion algorithm
• Weighted search results for optimal relevance
• Multiple search methods combined intelligently
• Automatic fallback for older MongoDB versions

Voyage AI Integration

• voyage-3-large - Latest state-of-the-art model
• 32K token context vs OpenAI's 8K
• Multilingual support across 26 languages
• Quantization support for cost optimization

📊 Use Cases

AI Development

• Code patterns and architecture decisions
• Bug fixes and debugging strategies
• API designs and integration patterns
• Performance optimizations and best practices

Knowledge Management

• Meeting notes and team decisions
• Research findings and technical insights
• Learning notes and skill development
• Project documentation and requirements

Team Collaboration

• Shared knowledge base across team members
• Decision tracking and reasoning documentation
• Best practices and coding standards
• Troubleshooting guides and solutions

🔒 Security & Reliability

• Input validation and sanitization
• Path security validation
• MongoDB injection prevention
• Connection pooling for high availability
• Automatic indexing for performance
• ACID transactions for data integrity

📈 Roadmap

• Multi-tenant support for team deployments
• Real-time collaboration features
• Advanced analytics dashboard
• Custom embedding models support
• GraphQL API for advanced integrations

🤝 Contributing

We welcome contributions! This project follows MongoDB and MCP best practices.

1. Fork the repository
2. Create a feature branch
3. Add tests for new features
4. Ensure all tests pass
5. Submit a pull request

📄 License

MIT License - see LICENSE file for details.

🙏 Acknowledgments

Built with cutting-edge technologies:

• MongoDB - The developer data platform
• Voyage AI - State-of-the-art embeddings
• Model Context Protocol - AI tool integration standard

Transform your AI coding workflow today with MongoDB's power! 🚀