Graphiti Cloud via Cloudflare Workers

[![GitHub Actions Workflow Status][workflow-src]][workflow-href] [![License][license-src]][license-href]

A Cloudflare Container Worker that serves as a proxy for the Graphiti MCP (Model Context Protocol) server, providing scalable, serverless access to AI agent memory capabilities through Neo4j-backed knowledge graphs.

Overview

Graphiti Cloud bridges the gap between AI applications and persistent memory by leveraging Cloudflare's new container service to host and proxy requests to a Graphiti MCP server. This enables AI agents to maintain context and memory across interactions using a powerful knowledge graph backend.

Features

• Serverless Architecture: Built on Cloudflare Workers with container support
• AI Memory Service: Persistent memory for AI agents via knowledge graphs
• Neo4j Integration: Robust graph database for complex relationship storage
• Auto-scaling: Up to 5 container instances with intelligent sleep management
• Global Edge Network: Deployed across Cloudflare's global infrastructure
• Secure: Environment-based configuration for sensitive credentials

Architecture

---
config:
  theme: neutral
  look: handDrawn
  layout: dagre
---
flowchart TB
 subgraph subGraph0["MCP-Enabled Clients"]
        Cursor["Cursor IDE"]
        Claude["Claude Desktop"]
        MCPClient["Other MCP Clients"]
  end
 subgraph subGraph1["Cloudflare Edge"]
        Worker["Graphiti Cloud Worker"]
        Container["Graphiti MCP Container"]
        CF["Cloudflare Infrastructure"]
  end
 subgraph subGraph2["External Services"]
        Neo4j[("Neo4j Knowledge Graph")]
        OpenAI["OpenAI API"]
  end
    Cursor --> Worker
    Claude --> Worker
    MCPClient --> Worker
    Worker --> Container
    Container --> Neo4j & OpenAI
    Worker -.-> CF
    Container -.-> CF
     Cursor:::Aqua
     Cursor:::Ash
     Claude:::Pine
     Claude:::Peach
     Claude:::Ash
     MCPClient:::Rose
     MCPClient:::Ash
     Worker:::Sky
     Container:::Sky
     CF:::Peach
     Neo4j:::Sky
     OpenAI:::Aqua
    classDef Pine stroke-width:1px, stroke-dasharray:none, stroke:#254336, fill:#27654A, color:#FFFFFF
    classDef Ash stroke-width:1px, stroke-dasharray:none, stroke:#999999, fill:#EEEEEE, color:#000000
    classDef Rose stroke-width:1px, stroke-dasharray:none, stroke:#FF5978, fill:#FFDFE5, color:#8E2236
    classDef Peach stroke-width:1px, stroke-dasharray:none, stroke:#FBB35A, fill:#FFEFDB, color:#8F632D
    classDef Sky stroke-width:1px, stroke-dasharray:none, stroke:#374D7C, fill:#E2EBFF, color:#374D7C
    classDef Aqua stroke-width:1px, stroke-dasharray:none, stroke:#46EDC8, fill:#DEFFF8, color:#378E7A

Quick Start

Prerequisites

• Node.js (v18 or later)
• pnpm package manager
• Wrangler CLI
• Neo4j database instance
• OpenAI API key

Neo4j Setup

You'll need a Neo4j database to store the knowledge graph. Here are your options:

1. Neo4j AuraDB (Recommended) - Fully managed cloud service

   • Sign up at Neo4j AuraDB
   • Create a free instance (up to 200k nodes and 400k relationships)
   • Get your connection URI, username, and password

2. Self-hosted Neo4j

   • Download Neo4j Community Edition
   • Follow the installation guide

3. Neo4j Desktop

   • Download Neo4j Desktop
   • Create a local database for development

Learn more about Neo4j: Neo4j is a gra

... View full README on GitHub