Graphiti
Graphiti Model Context Protocol (MCP) Server - An MCP server for knowledge graph management via Graphiti
What is Graphiti?
Graphiti is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to graphiti model context protocol (mcp) server - an mcp server for knowledge graph management via graphiti
Graphiti Model Context Protocol (MCP) Server - An MCP server for knowledge graph management via Graphiti
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Graphiti Model Context Protocol (MCP) Server - An MCP server
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx mcp-graphitiConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Graphiti
The Graphiti MCP Server provides AI assistants with a full knowledge graph management interface backed by Neo4j, enabling entity extraction, relationship storage, and graph querying through the Graphiti library. It is designed for teams and agents that need persistent, structured memory across sessions — storing facts, entities, and their relationships in a queryable graph rather than flat text. The server runs containerized via Docker and exposes an SSE endpoint that MCP-compatible clients connect to for real-time graph operations.
Prerequisites
- Python 3.10+ and pipx installed
- Docker and Docker Compose installed (for running Neo4j and Graphiti containers)
- Neo4j 5.x instance (launched via the provided Docker Compose config)
- pipx-installed graphiti CLI (from the GitHub repository)
- An MCP-compatible client such as Claude Desktop or Claude Code
Install the Graphiti CLI via pipx
Install the graphiti command-line tool directly from GitHub using pipx, which isolates the tool in its own environment.
pipx install 'git+https://github.com/rawr-ai/mcp-graphiti.git'Initialize a knowledge graph project
Create a new Graphiti project in your working directory. This generates a mcp-config.yaml configuration file you can customize.
graphiti init my-knowledge-graphGenerate Docker Compose and launch services
Generate the docker-compose.yml for your project and start Neo4j and the Graphiti MCP server in the background. Neo4j will be available at http://localhost:7474.
graphiti compose
graphiti up -dSet required environment variables
Set your Neo4j password before starting services. During development you can set GRAPHITI_ENV=dev to bypass strict password validation.
export NEO4J_PASSWORD=your_strong_password
export GRAPHITI_ENV=dev # development onlyConnect your MCP client to the SSE endpoint
The root Graphiti server listens on port 8000 at /sse; project containers start at port 8001 and increment. Configure your MCP client to connect to the appropriate SSE URL.
Verify connectivity via the health endpoint
Check that the Graphiti server is running and accepting connections by hitting the status endpoint.
curl http://localhost:8000/graphiti/statusGraphiti Examples
Client configuration
MCP client config connecting to the Graphiti SSE endpoint running locally on port 8000.
{
"mcpServers": {
"mcp-graphiti": {
"url": "http://localhost:8000/sse",
"type": "sse"
}
}
}Prompts to try
Example prompts that leverage the Graphiti knowledge graph for entity storage and retrieval.
- "Extract entities and relationships from this document and store them in the knowledge graph"
- "What relationships exist between Acme Corp and its subsidiary companies in the graph?"
- "Add a relationship between Alice and the 'Project Phoenix' entity with role 'Lead Engineer'"
- "Query the graph for all entities related to 'machine learning' within two hops"
- "Show me the knowledge graph status and how many nodes and edges are currently stored"Troubleshooting Graphiti
Neo4j connection refused or authentication failure
Ensure the NEO4J_PASSWORD environment variable is set before running 'graphiti up'. During development, set GRAPHITI_ENV=dev to relax password requirements. Verify Neo4j is running at http://localhost:7474.
SSE endpoint returns 404 or connection timeout
Check that the containers started successfully with 'docker-compose ps'. The root server is on port 8000; project containers start at 8001. Run 'graphiti reload <container>' if a container is unhealthy.
graphiti command not found after pipx install
Run 'pipx ensurepath' and restart your shell session so the pipx bin directory is in your PATH. Then verify with 'graphiti --help'.
Frequently Asked Questions about Graphiti
What is Graphiti?
Graphiti is a Model Context Protocol (MCP) server that graphiti model context protocol (mcp) server - an mcp server for knowledge graph management via graphiti It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Graphiti?
Follow the installation instructions on the Graphiti GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Graphiti?
Graphiti works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Graphiti free to use?
Yes, Graphiti is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Graphiti Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to Graphiti? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Graphiti in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Graphiti?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.