graphiti-mcp-server

Davidi18/graphiti-mcp-server

3.2

If you are the rightful owner of graphiti-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

Graphiti MCP Server is a standalone Model Context Protocol server designed for integrating Graphiti knowledge graphs with n8n workflows.

Graphiti MCP Server

A standalone Model Context Protocol server for Graphiti knowledge graphs, designed for n8n integration.

Features

  • 🧠 Knowledge Graph Memory - Powered by Graphiti and Neo4j
  • 🔄 Real-time Updates - Add episodes and search instantly
  • 🌐 REST API - Easy integration with n8n workflows
  • Fast Search - Semantic + keyword + graph search
  • 🕒 Temporal Awareness - Time-based queries and context
  • 🐳 Docker Ready - Deploy with Coolify

Quick Start

1. Environment Setup

Copy .env.example to .env and configure:

NEO4J_URI=bolt://your-neo4j:7687
NEO4J_USER=neo4j
NEO4J_PASSWORD=your_password
OPENAI_API_KEY=your_openai_key
2. Docker Deploy
bashdocker-compose up -d
3. Health Check
bashcurl http://localhost:8765/health
API Endpoints
Add Episode
bashPOST /add-episode
{
  "content": "User discussed project requirements",
  "source": {"type": "conversation", "user": "david"},
  "episode_type": "conversation"
}
Search Knowledge
bashPOST /search
{
  "query": "project requirements",
  "limit": 10
}
Health Check
bashGET /health
n8n Integration
Use HTTP Request nodes in n8n:

Store Memory: POST to /add-episode
Retrieve Context: POST to /search
Monitor Health: GET to /health

Deployment on Coolify

Create new application from GitHub repository
Set environment variables in Coolify
Deploy and access via your domain

Environment Variables
VariableDescriptionDefaultNEO4J_URINeo4j connection URIbolt://localhost:7687NEO4J_USERNeo4j usernameneo4jNEO4J_PASSWORDNeo4j passwordpassword123NEO4J_DATABASENeo4j database nameneo4jOPENAI_API_KEYOpenAI API keyRequiredMCP_SERVER_HOSTServer host0.0.0.0MCP_SERVER_PORTServer port8765
Architecture
n8n Workflows → HTTP REST → Graphiti MCP → Neo4j Database
Built with ❤️ for intelligent agent memory.