Skip to main content
Glama
NEO4J_SETUP.md1.94 kB
# Neo4j Setup Guide ## The Problem Your server is trying to connect to Neo4j at `localhost:7687`, but there's no Neo4j instance running. You need to configure a Neo4j database connection in your `.env` file. ## Quick Solution: Use Neo4j Aura (Free Cloud) ### Step 1: Get a Free Neo4j Aura Account 1. Go to **https://neo4j.com/cloud/aura/** 2. Click **"Start Free"** or **"Sign Up"** 3. Create an account (free tier available) 4. Create a new database instance ### Step 2: Get Your Connection Details After creating your database, you'll see connection details like: - **URI**: `neo4j+s://xxxxx.databases.neo4j.io` - **Username**: Usually `neo4j` - **Password**: The password you set when creating the database ### Step 3: Update Your .env File Open your `.env` file and add these lines: ```dotenv NEO4J_URI=neo4j+s://xxxxx.databases.neo4j.io NEO4J_USER=neo4j NEO4J_PASSWORD=your_actual_password_here ``` **Important**: - Replace `xxxxx.databases.neo4j.io` with your actual Aura URI - Replace `your_actual_password_here` with your actual password - The URI format `neo4j+s://` (with `+s`) is required for Aura ### Step 4: Whitelist Your IP (if needed) If you get connection errors: 1. Go to your Neo4j Aura dashboard 2. Find "IP Whitelist" or "Network Access" 3. Add your current IP address ### Step 5: Run the Server Again ```powershell .\run-server.ps1 ``` Or: ```powershell python graphiti_mcp_server.py --transport sse --port 8000 ``` ## Alternative: Local Neo4j If you have Neo4j running locally: ```dotenv NEO4J_URI=bolt://localhost:7687 NEO4J_USER=neo4j NEO4J_PASSWORD=your_local_password ``` ## Verify Your Setup Run the setup script to check your configuration: ```powershell .\setup-neo4j.ps1 ``` ## Need Help? - **Neo4j Aura Docs**: https://neo4j.com/docs/aura/ - **Connection Issues**: Check that your IP is whitelisted in Aura - **URI Format**: Make sure you're using `neo4j+s://` for Aura, not `bolt://`

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/apexneural-hansika/graphiti_mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server