AI Engineer Neo4j Memory MCP Demo

@a-s-g93

About AI Engineer Neo4j Memory MCP Demo

A short demo of the Neo4j Memory MCP server.

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "neo4j-memory-store": {
      "command": "uvx",
      "args": [
        "[email protected]"
      ],
      "env": {
        "NEO4J_URL": "bolt://localhost:7687",
        "NEO4J_USERNAME": "neo4j",
        "NEO4J_PASSWORD": "password",
        "NEO4J_DATABASE": "neo4j"
      }
    }
  }
}

Tools

No tools detected

Fetch the live tool list by running this server in a temporary sandbox using the button above.

Overview

What is AI Engineer Neo4j Memory MCP Demo?

A demo project that shows how to configure Claude Desktop with agentic memory using the Neo4j Memory MCP Server. It stores and retrieves conversational context in a Neo4j graph database, allowing memory to persist across sessions and be shared with other MCP-compatible clients.

How to use AI Engineer Neo4j Memory MCP Demo?

Install uv (Python package manager), create a Neo4j instance (Aura cloud or Desktop), then edit Claude Desktop’s MCP config file with the neo4j-memory-store server command, credentials, and the provided system prompt. After setup, start conversations in Claude Desktop – each interaction is automatically logged in Neo4j.

Key features of AI Engineer Neo4j Memory MCP Demo

Persistent memory across Claude Desktop conversations
Memory stored in a Neo4j graph database
System prompt for automatic memory retrieval and update
Compatible with other MCP clients (Cursor, Windsurf)
Uses uvx to run the official mcp-neo4j-memory package

Use cases of AI Engineer Neo4j Memory MCP Demo

Demonstrating how to give Claude Desktop a persistent memory
Experimenting with agentic memory patterns in AI assistants
Testing multi-client memory sharing with a shared Neo4j instance

FAQ from AI Engineer Neo4j Memory MCP Demo

What dependencies are required?

You need uv (Python package manager), a running Neo4j database (Aura or Desktop), and Claude Desktop. The MCP server is installed via uvx.

How do I configure Claude Desktop?

Add the neo4j-memory-store entry to your MCP config file (Settings → Developer → Edit Config) with the command uvx, arguments [email protected], and environment variables for your Neo4j credentials.

Where is memory data stored?

In the Neo4j database instance you configure (cloud Aura or local Desktop). The server creates entities, relations, and observations in that database.

Can other clients access the same memory?

Yes – if other MCP clients (like Cursor or Windsurf) are configured with the same Neo4j instance, they can read and write the same memory graph.

What transport does the server use?

The server uses standard MCP over stdio, launched via uvx. No network transport configuration is described; authentication is via Neo4j credentials in environment variables.

Frequently asked questions

What dependencies are required?

You need `uv` (Python package manager), a running Neo4j database (Aura or Desktop), and Claude Desktop. The MCP server is installed via `uvx`.

How do I configure Claude Desktop?

Add the `neo4j-memory-store` entry to your MCP config file (Settings → Developer → Edit Config) with the command `uvx`, arguments `[email protected]`, and environment variables for your Neo4j credentials.

Where is memory data stored?

In the Neo4j database instance you configure (cloud Aura or local Desktop). The server creates entities, relations, and observations in that database.

Can other clients access the same memory?

Yes – if other MCP clients (like Cursor or Windsurf) are configured with the same Neo4j instance, they can read and write the same memory graph.

What transport does the server use?

The server uses standard MCP over stdio, launched via `uvx`. No network transport configuration is described; authentication is via Neo4j credentials in environment variables.

Comments

Basic information

More Memory & Knowledge MCP servers

Orcha

westonhancock

Orcha is a unified context layer for AI tools and agents. It stores and indexes and exposes organizational, team, or individual knowledge. Files, structured databases, and connected sources live in one workspace, and age

Obsidian MCP Server

StevenStavrakis

A simple MCP server for Obsidian

717

Memory

modelcontextprotocol

Model Context Protocol Servers

88.1K

Anytype MCP Server

anyproto

An MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.

467

minutes

silverstein

Every meeting, every idea, every voice note — searchable by your AI. Open-source, privacy-first conversation memory layer.

1.3K