Neo4j MCP Server for Cursor

by ezedinff

3 stars
252 downloads
Not rated
GitHub

About

Connects to Neo4j graph databases with ability to use GDS functions ( when available), a read only mode , and set the sample size for schema detection

Details

Author
ezedinff
GitHub stars
3
Downloads
252
Categories
Database, Knowledge Base, Other, Developer Tools, AI

- Connect to local or remote Neo4j databases
- Execute Cypher queries with optional parameters
- Transform Neo4j data types to JavaScript objects
- Retrieve database version, node counts, and labels
- Monitor connection state and diagnostics
- Manage credentials securely via environment variables

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Neo4j MCP Server for Cursor
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Clone the repository, install dependencies with Bun (v1.0.0+), and configure Neo4j credentials in a .env file. The server uses stdio transport and is launched by Cursor. Six tools are available: connect with explicit credentials, connect with environment variables, execute a Cypher query, get database information, get connection status, and disconnect. For testing, run bun run index.ts or ./run-mcp-server.sh.

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "neo4j mcp server for cursor": {
            "neo4j-mcp": {
                "command": "bun",
                "args": [
                    "run",
                    "index.ts"
                ]
            }
        }
    }
}

McpServers

{
    "neo4j-mcp": {
        "command": "bun",
        "args": [
            "run",
            "index.ts"
        ]
    }
}

Neo4j MCP Server for Cursor

A Cursor MCP (Model Context Protocol) server that enables seamless interaction with Neo4j databases directly from the Cursor IDE.

Features

- Connect to Neo4j databases
- Execute Cypher queries and retrieve results
- Transform Neo4j-specific data types to standard JavaScript objects
- Uses the official MCP SDK with stdio transport for seamless integration with Cursor
- Support for environment variables for secure credential management
- Retrieve detailed database information and metrics
- Monitor connection status and diagnostics

Prerequisites

- Bun (v1.0.0 or higher)
- Neo4j database (local or remote)

Running Neo4j with Docker Compose

This project includes a Docker Compose configuration to easily run Neo4j in a container:

1. Make sure you have Docker and Docker Compose installed.

2. Start Neo4j using Docker Compose:

   docker-compose up -d

3. Access the Neo4j Browser at http://localhost:7474
- Default username: neo4j
- Default password: your_password (as specified in the docker-compose.yml)

4. To stop Neo4j:

   docker-compose down

5. To stop Neo4j and remove all data:

   docker-compose down -v

> Note: The default password in docker-compose.yml is set to your_password. For production use, change this to a secure password and update your .env file accordingly.

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/neo4j-mcp.git
cd neo4j-mcp

2. Install dependencies:

   bun install

3. Configure environment variables:
Create a .env file in the root directory with your Neo4j credentials:

   NEO4J_URI=neo4j://localhost:7687
NEO4J_USERNAME=neo4j
NEO4J_PASSWORD=your_password
NEO4J_DATABASE=neo4j
NODE_ENV=development

Usage

Starting the Server

The server uses stdio transport for communication with Cursor, so it should be started by Cursor itself. However, you can test it manually:

```bash

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.