Neo4j MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| NEO4J_URI | Yes | Neo4j connection URI | bolt://localhost:7687 |
| NEO4J_DATABASE | No | Neo4j database name | neo4j |
| NEO4J_PASSWORD | Yes | Neo4j password | password |
| NEO4J_USERNAME | Yes | Neo4j username | neo4j |
| NEO4J_NAMESPACE | No | Tool namespace prefix | |
| NEO4J_TRANSPORT | No | Transport protocol (stdio, http, or sse) | stdio |
| NEO4J_READ_TIMEOUT | No | Timeout in seconds for read queries | 30 |
| NEO4J_MCP_SERVER_HOST | No | Host to bind to | 127.0.0.1 |
| NEO4J_MCP_SERVER_PATH | No | Path for accessing MCP server | /api/mcp/ |
| NEO4J_MCP_SERVER_PORT | No | Port for HTTP/SSE transport | 8000 |
| NEO4J_RESPONSE_TOKEN_LIMIT | No | Maximum tokens for read query responses | |
| NEO4J_MCP_SERVER_ALLOWED_HOSTS | No | Comma-separated list of allowed hosts (DNS rebinding protection) | localhost,127.0.0.1 |
| NEO4J_MCP_SERVER_ALLOW_ORIGINS | No | Comma-separated list of allowed CORS origins |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_neo4j_schemaA | List all nodes, their attributes and their relationships to other nodes in the neo4j database. This requires that the APOC plugin is installed and enabled. |
| read_neo4j_cypherB | Execute a read Cypher query on the neo4j database. |
| write_neo4j_cypherA | Execute a write Cypher query on the neo4j database. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: write_neo4j_cypher for mutations, read_neo4j_cypher for queries, and get_neo4j_schema for metadata. There is no overlap or ambiguity between them.
All tools follow a consistent verb_neo4j_noun pattern (write/read/get). The naming is predictable and the verb clearly indicates the operation type.
With three tools, the server is well-scoped for its purpose: read, write, and schema introspection. Each tool covers an essential need without redundancy.
The surface covers the core operations for a Neo4j MCP server: fetching schema, executing read queries, and executing write queries. Since write Cypher can handle create, update, and delete, there are no major gaps.