iflow-mcp_mjftw_neo4j-mcp-server

# Neo4j MCP Server This is a Memory Control Protocol (MCP) server implementation that uses Neo4j as the backend storage for knowledge graph management. It provides a stdio-based interface for storing and retrieving knowledge in a graph database format. ## Prerequisites - Python 3.8+ - Neo4j Database (local or remote) - Poetry (Python package manager) - Docker and Docker Compose (for running Neo4j) - Go Task (optional, for task automation) ## Installation 1. Clone the repository: ```bash git clone <repository-url> cd neo4j_mcp_server ``` 2. Install Poetry if you haven't already: ```bash curl -sSL https://install.python-poetry.org | python3 - ``` 3. Install dependencies: ```bash poetry install ``` ## Configuration ### Claude Desktop Configuration For Ubuntu users running Claude Desktop, you can configure the MCP server by adding it to your Claude desktop configuration file at: ``` ~/.config/Claude/claude_desktop_config.json ``` Before configuring, you need to build the standalone executable: ```bash task build ``` This will create a binary at `dist/neo4j_mcp_server`. Make sure to update the path in your configuration to point to this built executable. An example configuration is provided in `example_mcp_config.json`. You can copy and modify this file: ```bash cp example_mcp_config.json ~/.config/Claude/claude_desktop_config.json ``` Then edit the `command` path in the configuration file to point to your built executable: ```json { "mcpServers": [ { "name": "neo4j-knowledge-graph", "command": ["/path/to/your/dist/neo4j_mcp_server"], ... } ] } ``` The configuration includes: - Server name and description - Command to start the server (path to the built executable) - Available tools and their parameters - Required fields and data types ## Running the Server ### Using Task (Recommended) If you have Go Task installed, you can use the provided Taskfile to manage the server: ```bash # Show available tasks task # Start every…