Memory
Implements long-term memory capabilities for AI assistants by leveraging PostgreSQL and pgvector for efficient vector similarity search.
About
This server offers long-term memory functionality for AI assistants, utilizing mem0 principles. It's built with PostgreSQL and pgvector to provide efficient vector similarity search. Features include automatic embedding generation, a RESTful API for memory operations, semantic search, support for various memory types, tag-based retrieval, confidence scoring, and Server-Sent Events (SSE) for real-time updates, compatible with the Cursor MCP protocol.
Key Features
- RESTful API for memory operations
- 2 GitHub stars
- Automatic embedding generation using BERT
- Server-Sent Events (SSE) for real-time updates
- PostgreSQL with pgvector for vector similarity search
- Semantic search capabilities
Use Cases
- Providing semantic search capabilities for AI to find relevant memories.
- Enabling AI assistants to retain and recall information over extended periods.
- Integrating long-term memory into Cursor IDE through MCP protocol.
Sponsored