Memgraph MCP Server

What is Memgraph MCP Server about?

Enables seamless communication between Memgraph and LLMs by exposing Memgraph operations (e.g., Cypher queries, schema retrieval) through the Model Context Protocol.

How to use Memgraph MCP Server?

1. Install uv (a fast Python package manager). Create and activate a virtual environment with uv venv and .venv\Scripts\activate.
2. Add dependencies: uv add "mcp[cli]" httpx.
3. Start the server: uv run server.py.
4. Configure an MCP client (e.g., Claude for Desktop) by adding the server command and arguments to the Claude config file.
5. Run Memgraph (e.g., via Docker) with schema info enabled: docker run -p 7687:7687 memgraph/memgraph-mage --schema-info-enabled=True.
6. Interact with the database through the LLM client.

Key features of Memgraph MCP Server

• run_query(): Executes arbitrary Cypher queries against a running Memgraph instance.
• get_schema(): Retrieves full schema information when Memgraph is started with --schema-info-enabled=True.
• Minimal setup using Python's uv and a single server.py entry point.
• Designed for integration with LLM tools like Claude for Desktop.

Use cases of Memgraph MCP Server

• Chat‑driven graph analytics: Users can ask natural‑language questions and have the LLM translate them into Cypher queries.
• Rapid prototyping of AI agents that need graph data as context.
• Integrating Memgraph with existing LLM workflows without writing custom adapters.
• Educational demos showing how graph databases can be queried through conversational interfaces.

FAQ from the Memgraph MCP Server

• Q: Which Python package manager should I use? A: uv is recommended for its speed and simplicity, but any environment that can install mcp[cli] and httpx will work.
• Q: How do I point Claude to the server? A: Add an entry under mcpServers in claude_desktop_config.json with the command uv and arguments --directory <repo-path> run server.py.
• Q: Do I need to expose the Memgraph port? A: Yes, run Memgraph (or Memgraph MAGE) with the default Bolt port (7687) exposed so the server can connect.
• Q: What if I want schema information? A: Start Memgraph with --schema-info-enabled=True; then get_schema() will return full schema details.
• Q: Is there a JavaScript/TypeScript version? A: A TypeScript version is planned for a future release.

[!IMPORTANT]
This repository has been merged into the Memgraph AI Toolkit monorepo to avoid duplicating tools.
It will be deleted in one month—please follow the MCP integration there for all future development, and feel free to open issues or PRs in that repo.

🚀 Memgraph MCP Server

Memgraph MCP Server is a lightweight server implementation of the Model Context Protocol (MCP) designed to connect Memgraph with LLMs.

⚡ Quick start

📹 Memgraph MCP Server Quick Start video

1. Run Memgraph MCP Server

1. Install uv and create venv with uv venv. Activate virtual environment with .venv\Scripts\activate.
2. Install dependencies: uv add "mcp[cli]" httpx
3. Run Memgraph MCP server: uv run server.py.

2. Run MCP Client

1. Install Claude for Desktop.
2. Add the Memgraph server to Claude config:

MacOS/Linux

code ~/Library/Application\ Support/Claude/claude_desktop_config.json

Windows

code $env:AppData\Claude\claude_desktop_config.json

Example config:

{
    "mcpServers": {
      "mpc-memgraph": {
        "command": "/Users/katelatte/.local/bin/uv",
        "args": [
            "--directory",
            "/Users/katelatte/projects/mcp-memgraph",
            "run",
            "server.py"
        ]
     }
   }
}

[!NOTE]
You may need to put the full path to the uv executable in the command field. You can get this by running which uv on MacOS/Linux or where uv on Windows. Make sure you pass in the absolute path to your server.

3. Chat with the database

1. Run Memgraph MAGE:

   docker run -p 7687:7687 memgraph/memgraph-mage --schema-info-enabled=True
   

   The --schema-info-enabled configuration setting is set to True to allow LLM to run SHOW SCHEMA INFO query.
2. Open Claude Desktop and see the Memgraph tools and resources listed. Try it out! (You can load dummy data from Memgraph Lab Datasets)

🔧Tools

run_query()

Run a Cypher query against Memgraph.

🗃️ Resources

get_schema()

Get Memgraph schema information (prerequisite: --schema-info-enabled=True).

🗺️ Roadmap

The Memgraph MCP Server is just at its beginnings. We're actively working on expanding its capabilities and making it even easier to integrate Memgraph into modern AI workflows. In the near future, we'll be releasing a TypeScript version of the server to better support JavaScript-based environments. Additionally, we plan to migrate this project into our central AI Toolkit repository, where it will live alongside other tools and integrations for LangChain, LlamaIndex, and MCP. Our goal is to provide a unified, open-source toolkit that makes it seamless to build graph-powered applications and intelligent agents with Memgraph at the core.