mcpField
A
AI Engineer Neo4j Memory MCP Demo

AI Engineer Neo4j Memory MCP Demo

Local
github.com/a-s-g93Memory & Knowledge5Updated 73w ago
Repository

About AI Engineer Neo4j Memory MCP Demo

This is a short demo of the Neo4j Memory MCP server, showing how to configure Claude Desktop with graph-based conversation memory. Memories are stored in Neo4j and can be accessed by other clients using the same database.

Key features

  1. 01Demonstrates Claude Desktop configuration
  2. 02Logs interactions to a Neo4j database
  3. 03Uses memory across conversations
  4. 04Allows other clients to access the same database

Use cases

  1. 01Configure persistent memory for Claude Desktop
  2. 02Refer to information recorded in earlier conversations
  3. 03Let multiple clients access the same memories

MCP support

The demo shows how to configure the Neo4j Memory MCP server in Claude Desktop so interactions are recorded in Neo4j and used in later conversations.

#ai-engineer#neo4j#memory

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "ai-engineer-neo4j-memory-demo": {
      "command": "uvx",
      "args": [
        "[email protected]"
      ],
      "env": {
        "NEO4J_URL": "<NEO4J_URL>",
        "NEO4J_USERNAME": "<NEO4J_USERNAME>",
        "NEO4J_PASSWORD": "<NEO4J_PASSWORD>",
        "NEO4J_DATABASE": "<NEO4J_DATABASE>"
      }
    }
  }
}
Run commanduvx [email protected]
Environment variables to set
NEO4J_URL
NEO4J_USERNAME
NEO4J_PASSWORDSecret
NEO4J_DATABASE

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypi[email protected]

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

AI Engineer Neo4j Memory MCP Demo

This is a short demo of the Neo4j Memory MCP server.

This file descibes how to configure Claude Desktop with agentic memory courtesy of the Neo4j Memory MCP Server.

Set Up

Install uv Python Package Manager

The Neo4j Memory MCP Server is written in Python and managed by uv. Follow the directions on this page to properly install.

Neo4j Database

Create a Neo4j database instance by either

Claude Desktop

Download Claude Desktop

  • navigate to Settings -> Developer -> Edit Config
  • Add the Neo4j Memory MCP server to the config file

You may replace the config values with your own database credentials

{
  "mcpServers": {
    "neo4j-memory-store": {
      "command": "uvx",
      "args": [ "[email protected]" ],
      "env": {
        "NEO4J_URL": "bolt://localhost:7687",
        "NEO4J_USERNAME": "neo4j",
        "NEO4J_PASSWORD": "password",
        "NEO4J_DATABASE": "neo4j"
      }
    }
  }
}
  • Navigate to Settings -> General -> Claude Settings -> Configure
  • In the personal preferences section - paste this text:
Follow these steps for each interaction:

1. Memory Retrieval:
   - Always begin your chat by saying only "Remembering..." and retrieve all relevant information from your knowledge graph
   - Always refer to your knowledge graph as your "memory"

2. Memory Update:
   - If any new information was gathered during the interaction, update your memory as follows:
     a) Create entities for recurring organizations, people, and significant events
     b) Connect them to the current entities using relations
     b) Store facts about them as observations

This text is a modified version of the system prompt Anthropic recommends for their memory MCP server.

Demo

Begin a conversation with Claude Desktop.

Each interaction should be automatically logged in the Neo4j database you've configured.

These memories may be used between different conversations with Claude and even accessed by other clients such as Cursor or Windsurf, if they are configured with the same Neo4j instance.

Example Knowledge Graph

This knowledge graph contains the semantic memories of our conversation.

knowledge-graph

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.