Memorizer

•
118
•

Enables AI agents to store, retrieve, and semantically search through memories using vector embeddings.

Memorizer is a .NET-based service leveraging PostgreSQL with the pgvector extension to provide efficient similarity search capabilities. Designed for AI agents, it offers a robust long-term memory system where agents can store structured memories with vector embeddings. It supports retrieving memories by ID, performing semantic searches based on vector similarity, filtering search results by tags, and creating relationships to form knowledge graphs. The service integrates with the Model Context Protocol (MCP) for seamless use with AI agents and includes a web-based UI for manual memory management and configuration.

Key Features

01Integration with Model Context Protocol (MCP) for AI agents
02Create relationships between memories to form knowledge graphs
0351 GitHub stars
04Web-based UI for manual memory management
05Semantic search through memories using vector similarity
06Store structured memories with vector embeddings

Use Cases

01Providing long-term, retrievable memory for AI agents
02Building and managing knowledge graphs for AI systems
03Enabling semantic search capabilities over agent-specific data