mcp-neo4j-cypher

Local
Community
888
Signed
GitHub Repo

Overview

The mcp-neo4j-cypher server is a Model Context Protocol (MCP) server that enables AI assistants and agents to interact directly with Neo4j graph databases using Cypher, Neo4j’s query language. It allows AI-driven workflows to query, explore, and analyze graph data — including nodes, relationships, and paths — without embedding Neo4j drivers or Cypher execution logic into the client.

This server is especially useful for knowledge graphs, dependency analysis, recommendation systems, and any workflow that benefits from graph-native querying and reasoning.

Transport

stdio

Tools

  • get_neo4j_schema
  • read_neo4j_cypher
  • write_neo4j_cypher

Key Capabilities

  • Graph querying with Cypher — Run expressive graph queries using Neo4j’s native query language.
  • Relationship-centric analysis — Explore how entities connect, interact, and evolve across the graph.
  • Schema and metadata access — Inspect labels, relationship types, and property definitions.
  • Knowledge graph workflows — Support AI-assisted reasoning over complex graph structures.
  • Insight generation — Enable assistants to summarize patterns, paths, and communities in graph data.

How It Works

The mcp-neo4j-cypher server runs as a local or containerized MCP service and connects to a Neo4j database using configured connection details such as URI, authentication credentials, and database name. AI clients communicate with the server over the MCP protocol to request graph context as part of broader reasoning workflows.

The server mediates execution of Cypher queries, handling authentication, query submission, and result normalization before returning structured outputs that AI assistants can reason over directly. This abstraction allows AI workflows to focus on graph insight and interpretation, rather than database connectivity or query execution mechanics.

By exposing Neo4j’s Cypher interface through MCP, the server enables AI-driven workflows such as graph exploration, relationship analysis, and knowledge-driven reasoning — all through natural language and automated reasoning within a single environment.