Neo4j
FreeNeo4j 官方贡献,操作 Neo4j 图数据库(模式+读写 Cypher),并提供图数据库支持的记忆功能。
About Neo4j
Neo4j Labs MCP (Model Context Protocol) Servers provide a suite of tools that allow large language models and AI assistants to interact with Neo4j graph databases and Aura cloud services through natural language. The collection includes four servers: mcp-neo4j-cypher for executing read/write Cypher queries with automatic schema inspection (requires APOC plugin), mcp-neo4j-memory for storing and retrieving personal knowledge graph memory across sessions and clients, mcp-neo4j-cloud-aura-api for managing Neo4j Aura instances (create, destroy, scale, enable features), and mcp-neo4j-data-modeling for interactive graph data modeling and visualization with Arrows.app import/export. All servers support STDIO, SSE, and HTTP transport modes and can be used with Claude Desktop, VS Code, Cursor, Windsurf, Gemini CLI, and other MCP clients.
Key Features
Pros & Cons
- Enables non-technical users to interact with Neo4j via natural language
- Memory server allows information to be retained across different sessions and LLM clients
- Open source with active development by Neo4j Field GenAI team
- Flexible transport modes accommodate local, web, and microservice deployments
- Strong integration with popular AI development tools (Claude, VS Code, etc.)
- Cypher server requires APOC plugin for schema inspection, adding a dependency
- Not officially supported by Neo4j product team; no SLA or backward compatibility guarantees
- Each server must be configured separately, increasing setup complexity
- Memory and data modeling servers require a running Neo4j instance