Introduction to OpenMemory

OpenMemory is a production-ready long-term memory system designed specifically for AI agents and conversational systems. It implements the HMD v2 (Holistic Memory Descriptor v2) specification with advanced features like multi-sector embeddings, time-based decay, and graph-based waypoints.

Why OpenMemory?

Most "memory" solutions are just vector databases wrapped in a Python script. OpenMemory is different. It is a Cognitive Engine.

The Problem: Stateless AI

Modern AI assistants have amnesia. They forget everything the moment the session ends.

• No Continuity: Every chat starts from zero.
• No Learning: They repeat the same mistakes.
• High Cost: Re-sending context tokens is expensive.

The Solution: A Living Brain

OpenMemory gives your agent a persistent, evolving brain that lives on your local machine.

1. Local-First: Runs inside your app (Node.js/Python). No external API calls required.
2. Temporal: Understands time. Knows that "current CEO" changes over time.
3. Multi-Sector: Stores memories in 5 distinct sectors (Episodic, Semantic, Procedural, Emotional, Reflective).
4. Adaptive Decay: Important memories stick; trivial ones fade.

Key Features

• ⚡ Standalone Mode: npm install openmemory-js. No Docker. No Server.
• 🕸️ Temporal Knowledge Graph: Tracks facts like (User, location, New York, 2023-2024).
• 🧠 5-Sector Embeddings: Context-aware retrieval based on why you are asking.
• 🔌 MCP Support: Plug-and-play with Claude Desktop and Cursor.

How it Works

graph TD
    A[User Query] --> B{Cognitive Router}
    B -->|Factual| C[Semantic Sector]
    B -->|Events| D[Episodic Sector]
    B -->|Feelings| E[Emotional Sector]
    C & D & E --> F[Ranking Engine]
    F -->|Decay + Salience| G[Final Context]
    G --> H[LLM Response]

Get Started

Ready to give your agent a brain?

• Quick Start: Up and running in 5 minutes.
• Standalone Mode: Learn about the local-first architecture.
• Examples: Copy-paste code snippets.