MCP Memory Server
by sdimitrov
This server implements long-term memory capabilities for AI assistants using mem0 principles. It is powered by PostgreSQL with pgvector for efficient vector similarity search.
Last updated: N/A
What is MCP Memory Server?
The MCP Memory Server is a server that provides long-term memory capabilities for AI assistants. It leverages PostgreSQL with the pgvector extension to perform efficient vector similarity searches, enabling AI agents to store, retrieve, and reason about information over time.
How to use MCP Memory Server?
To use the server, first ensure you have PostgreSQL 14+ with the pgvector extension and Node.js 16+ installed. Then, install dependencies using npm install, configure environment variables in the .env file, initialize the database with npm run prisma:migrate, and start the server using npm start. The server can be integrated with Cursor by modifying the mcp.json configuration file. The API provides endpoints for creating, searching, and listing memories, as well as a health check endpoint.
Key features of MCP Memory Server
PostgreSQL with pgvector for vector similarity search
Automatic embedding generation using BERT
RESTful API for memory operations
Semantic search capabilities
Support for different types of memories (learnings, experiences, etc.)
Tag-based memory retrieval
Confidence scoring for memories
Server-Sent Events (SSE) for real-time updates
Cursor MCP protocol compatible
Use cases of MCP Memory Server
Enabling AI assistants to remember past interactions and learnings
Providing context for AI agents to make more informed decisions
Building AI systems that can learn and adapt over time
Creating personalized AI experiences based on user history
FAQ from MCP Memory Server
What is the purpose of the MCP Memory Server?
What is the purpose of the MCP Memory Server?
The MCP Memory Server provides long-term memory capabilities for AI assistants, allowing them to store, retrieve, and reason about information over time.
What technologies does the server use?
What technologies does the server use?
The server uses PostgreSQL with the pgvector extension for vector similarity search, BERT for automatic embedding generation, and a RESTful API for memory operations.
How do I install and configure the server?
How do I install and configure the server?
You need PostgreSQL 14+ with pgvector and Node.js 16+. Install dependencies, configure environment variables, initialize the database, and start the server using the provided npm scripts.
How do I integrate the server with Cursor?
How do I integrate the server with Cursor?
Modify your ~/.cursor/mcp.json file to include the server's command and arguments, pointing to the server's entry point.
What API endpoints are available?
What API endpoints are available?
The API provides endpoints for creating, searching, and listing memories, as well as a health check endpoint and an SSE endpoint for real-time updates.