Official

Neo4j MCP Server

Graph database management and natural language querying for your AI workspace.

Data last scanned today · Reviewed yesterday

60★ 981PythonMIT

Overview

The Neo4j MCP Server suite provides a collection of tools for managing Neo4j databases through natural language. These modules include capabilities for executing Cypher queries, managing Neo4j Aura cloud instances, performing graph data modeling, and maintaining a persistent knowledge graph memory across your AI sessions.

Our verdict

This server carries a high risk rating with a security score of 27/100. The primary concern is that it lacks an authentication method, meaning anyone with access to the server configuration can execute commands, compounded by the fact that it defaults to full write access without a restricted read-only mode. It is intended for experimental use by developers familiar with the risks; it is not recommended for environments where sensitive data integrity or strict access control is required.

Supports STDIO, SSE, and HTTP transport modes
Zero direct dependencies minimizes supply-chain surface area
Integrates directly with clients like Claude Desktop and Cursor

No authentication method implemented
Lacks a read-only mode for restricted database access
No official product support or service level agreements

Compatibility

Client	Local	Docker	Remote	Read-only
ChatGPT
Claude Desktop
Cursor
VS Code
Windsurf

Frequently asked questions

›What is required to use the mcp-neo4j-cypher server?

This server requires the APOC plugin to be installed and enabled on your Neo4j instance to perform schema inspection.

›Which transport modes are supported by these MCP servers?

All servers support STDIO for local use, as well as SSE and HTTP transport modes for web-based and microservice deployments.

›How can I configure the HTTP server settings?

You can configure HTTP mode using command-line flags such as --transport, --host, --port, and --path, or by setting corresponding environment variables.

›Are these servers officially supported by the Neo4j product team?

No, these servers are developed by the Neo4j Field GenAI team as part of Neo4j Labs and do not provide service level agreements or guarantees regarding backwards compatibility.

›Can these servers be deployed to cloud environments?

Yes, all servers are containerized and support HTTP transport for cloud deployment on platforms like AWS ECS Fargate and Azure Container Apps.