n8n-mcp FAQ
These answers explain how the hosted n8n-mcp gateway connects AI clients to n8n workflows, how credentials are protected and when a gateway is useful compared with a local MCP server.
What is n8n-mcp?
n8n-mcp is a hosted Model Context Protocol gateway for connecting self-hosted n8n workflows to AI clients.
Do I need to expose my n8n API key to AI clients?
No. n8n API keys are stored by the gateway and encrypted at rest; AI clients authenticate with a platform API key.
Which AI clients are supported?
Any MCP-compatible client can connect, including Claude, ChatGPT, Cursor, Windsurf, VS Code and Zed.
How is n8n-mcp different from a local MCP server?
A local MCP server runs beside one client, while n8n-mcp provides a hosted gateway URL, per-user authentication, encrypted n8n credential storage and centralized request controls.