Translates natural language to Cypher queries and executes them against a Neo4j database
Stores and retrieves entities and relationships in a personal knowledge graph using Neo4j
Manages Neo4j Aura cloud instances (create, find, scale, enable features, destroy)
Neo4j MCP provides a suite of servers that enable AI assistants to interact with Neo4j graph databases through natural language. Users can query databases with plain English, manage Neo4j Aura cloud instances, and build personal knowledge graphs that persist across conversations. This integration brings the power of graph databases directly into AI chat interfaces, making database operations more accessible and intuitive.
The Neo4j MCP integration allows you to interact with Neo4j graph databases directly from your AI assistant. This MCP provides three specialized servers that handle different aspects of Neo4j functionality:
To use the Neo4j MCP servers, you'll need to:
Clone the repository:
git clone https://github.com/neo4j-contrib/mcp-neo4j.git
cd mcp-neo4j
Install the server you want to use. Each server has its own installation instructions in its respective directory.
Configure your MCP client (Claude Desktop, VS Code, Cursor, etc.) to use the server by adding the appropriate configuration to your client's settings.
The Cypher server allows you to:
To use it, you'll need to provide connection details for your Neo4j database.
The Memory server enables:
This server requires a Neo4j instance (local or remote) to store the knowledge graph.
The Aura API server lets you:
You'll need to provide your Neo4j Aura API credentials to use this server.
Here are some examples of what you can ask the Neo4j MCP:
For detailed usage instructions for each server, refer to the README files in the respective server directories within the repository.