mcp-neo4j

Overview

This repository contains multiple MCP servers from Neo4j Labs that enable AI assistants to interact with Neo4j graph databases. The servers provide different capabilities: mcp-neo4j-cypher converts natural language to Cypher queries, mcp-neo4j-memory creates personal knowledge graphs, mcp-neo4j-cloud-aura-api manages cloud instances, and mcp-neo4j-data-modeling handles graph data modeling. All servers support multiple transport modes including STDIO, SSE, and HTTP, making them versatile for different deployment scenarios.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• get_database_schema

  Retrieve database schema for a configured Neo4j database

  Try:What's the schema of this graph?Try:Show me the structure of the database
• execute_cypher_query

  Execute read and write Cypher queries on a Neo4j database

  Try:Find all customers who bought more than 10 productsTry:Create a new node for a product
• store_entity

  Store an entity in your personal knowledge graph

  Try:Remember that I worked on MCP Neo4j servers todayTry:Store that Neo4j is a graph database
• retrieve_entities

  Retrieve entities and relationships from your knowledge graph

  Try:What projects have I been working on?Try:Show me all my contacts related to graph databases
• create_instance

  Create a new Neo4j Aura cloud instance

  Try:Create a new instance named mcp-test with 4GB memoryTry:Make an Aura Professional instance with Graph Data Science
• list_instances

  List your Neo4j Aura cloud instances

  Try:Show all my Aura instancesTry:List instances with name containing 'test'
• scale_instance

  Scale up or down a Neo4j Aura cloud instance

  Try:Scale my production instance to 16GBTry:Downsize my test instance to 2GB
• enable_instance_features

  Enable features on a Neo4j Aura cloud instance

  Try:Enable Graph Data Science on my instanceTry:Activate Bloom for my database
• find_instances

  Find Neo4j Aura cloud instances by name or criteria

  Try:Find instances with name 'production'Try:Show me all Aura Professional instances
• destroy_instance

  Delete a Neo4j Aura cloud instance

  Try:Delete the old test instanceTry:Remove my staging database
• create_data_model

  Create a new Neo4j graph data model

  Try:Create a model for a social media networkTry:Design a data model for an e-commerce site
• validate_data_model

  Validate a Neo4j graph data model

  Try:Check if my e-commerce model has issuesTry:Validate my social network design

Comparable tools

Installation

pip install mcp-neo4j-cypher mcp-neo4j-memory mcp-neo4j-cloud-aura-api mcp-neo4j-data-modeling

{
  "mcpServers": {
    "neo4j": {
      "command": "mcp-neo4j-cypher",
      "args": []
    }
  }
}

docker run -p 8080:8080 neo4j/mcp-neo4j-cypher --transport http --port 8080

FAQ

What is the difference between these MCP servers and the official Neo4j MCP server?

These are experimental servers from Neo4j Labs with cutting-edge features, while the official server at github.com/neo4j/mcp provides stable, product-supported functionality.

Do I need the APOC plugin for all servers?

No, only the mcp-neo4j-cypher server requires APOC for schema inspection. Other servers have different requirements.

Compare mcp-neo4j with