MCP Memory Service

Active
GitHub Python Apache-2.0

Description

Open-source persistent memory service for AI agents, supporting LangGraph, CrewAI, and AutoGen with REST API, knowledge graph, and autonomous memory consolidation.

Key Features

  • Open-source persistent memory service for AI agents — provides long-term memory storage and retrieval via REST API for any MCP-compatible framework
  • Knowledge graph-backed memory with autonomous consolidation — agents organize and merge memories into structured knowledge representations over time
  • Multi-framework compatibility — works with LangGraph, CrewAI, AutoGen, and other popular agent frameworks out of the box
  • Model Context Protocol (MCP) integration — standardized tool interface that any MCP-enabled agent can connect to for memory operations
  • Python-based implementation with Apache-2.0 license — self-hosted deployment with full control over data privacy and memory retention policies

Use Cases

💡 Adding persistent memory to stateless agent frameworks — give LangGraph, CrewAI, or AutoGen agents the ability to remember across sessions
💡 Building memory-augmented chatbots that retain user preferences, conversation history, and learned patterns over extended interactions
💡 Multi-agent memory sharing in collaborative systems — agents share a common knowledge graph for coordinated decision-making
💡 Enterprise AI assistants requiring auditable, self-hosted memory infrastructure with knowledge graph-based context management

Tags

agent-memory mcp knowledge-graph persistent python

Categories

🧩 Agent Memory

Quick Start

Clone the repo from `github.com/doobidoo/mcp-memory-service`, install Python dependencies, and start the service. Connect your MCP-compatible agent (LangGraph, CrewAI, AutoGen) to the REST API endpoint. The service provides memory store, retrieve, and consolidate operations that agents can call as tools via MCP protocol.
Visit GitHub

Related Projects