Neo4j FastMCP Server with Descope Authentication
@artur-nohup
关于 Neo4j FastMCP Server with Descope Authentication
Neo4j MCP Server built with FastMCP TypeScript framework, featuring OAuth authentication and Docker support
基本信息
配置
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"neo4j-fastmcp-server": {
"command": "npx",
"args": [
"@modelcontextprotocol/inspector"
]
}
}
}工具
未检测到工具
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
概览
What is Neo4j FastMCP Server with Descope Authentication?
A production-ready remote Model Context Protocol (MCP) server with enterprise-grade authentication powered by Descope. It integrates with Neo4j knowledge graph for persistent memory and is fully compatible with Claude.ai and other MCP clients.
How to use Neo4j FastMCP Server with Descope Authentication?
Clone the repository, install dependencies (npm install), configure the .env file with Descope, Neo4j, and OAuth credentials, then start with npm run dev (port 8081 for MCP, 8082 for OAuth). For Claude.ai, add a custom integration pointing to the MCP server URL and complete the OAuth flow. Docker deployments use docker run with environment variables or docker compose up -d.
Key features of Neo4j FastMCP Server with Descope Authentication
- Enterprise authentication: OAuth 2.1 with PKCE + API keys
- Full OAuth flow integration for Claude.ai custom integrations
- Persistent Neo4j knowledge graph with full-text search
- Fine-grained permissions with MCP-specific scopes and RBAC
- Automatic token refresh and session validation
- Docker support with health checks and monitoring
Use cases of Neo4j FastMCP Server with Descope Authentication
- Add persistent memory to Claude.ai via custom MCP integration
- Build enterprise knowledge management with company OAuth and shared graphs
- Enable AI agents to store, search, and retrieve context across sessions
- Create contextual AI applications with secure, remote knowledge storage
FAQ from Neo4j FastMCP Server with Descope Authentication
What authentication methods are supported?
OAuth 2.1 with PKCE (for interactive clients like Claude.ai) and API key authentication (for programmatic access). Both require Descope as the identity platform.
How do I integrate with Claude.ai?
In Claude.ai settings, add a custom MCP integration with the server URL (e.g., http://localhost:8081/stream). Complete the OAuth flow when prompted to connect.
What environment variables are required?
You must set DESCOPE_PROJECT_ID, DESCOPE_MANAGEMENT_KEY, NEO4J_URI, NEO4J_USERNAME, NEO4J_PASSWORD, OAUTH_CLIENT_ID, OAUTH_CLIENT_SECRET, OAUTH_REDIRECT_URI, and SESSION_SECRET.
How can I deploy with Docker?
Use the pre-built Docker image arturrenzenbrink/neo4j-fastmcp-server:latest with environment variables for Descope, Neo4j, and session secret, or use the included docker-compose.yml.
What testing tools are available?
You can use the MCP Inspector (npx @modelcontextprotocol/inspector) or test with cURL by sending requests to http://localhost:8081/stream using either an OAuth bearer token or an API key header.
开发工具 分类下的更多 MCP 服务器
Hello World MCP Server (Reference Extension)
anthropicsDesktop Extensions: One-click local MCP server installation in desktop apps
Serena
oraiosA powerful MCP toolkit for coding, providing semantic retrieval and editing capabilities - the IDE for your agent
nuxt-mcp / vite-plugin-mcp
antfuMCP server helping models to understand your Vite/Nuxt app better.
Test
x1xhlolFULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia & v0. (And other Open Sou
FastAPI-MCP
tadata-orgExpose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth!
评论