n8n-nodes-mcp-client

What is n8n-nodes-mcp-client?

n8n-nodes-mcp-client is a community node for n8n that lets you interact with Model Context Protocol (MCP) servers directly in your n8n workflows. It runs on the n8n automation platform and is intended for developers and automation builders who want to connect AI models to external tools and data sources via the MCP standard.

How to use n8n-nodes-mcp-client?

Install the node via the n8n community nodes documentation, then create credentials for either STDIO (command‑line) or SSE transport. Set environment variables either in the credentials UI or via Docker with the MCP_ prefix. After enabling community nodes as tools (N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true), you can use the node as a tool inside AI Agent workflows or run its six operations (Execute Tool, Get Prompt, List Prompts, List Resources, List Tools, Read Resource).

Key features of n8n-nodes-mcp-client

• Connect to MCP servers via STDIO or SSE transport
• Execute tools, list resources, and read resource content
• Retrieve and use prompt templates from MCP servers
• Use the node as a tool inside n8n AI Agents
• Pass environment variables to servers via UI or Docker
• Compatible with n8n 1.0+ and MCP Protocol 1.0+

Use cases of n8n-nodes-mcp-client

• Execute a Brave Search tool from an MCP server to fetch latest news
• Build a multi‑server AI agent that combines weather, search, and news data
• Connect to a local MCP server with SSE for real‑time tool execution
• List all available tools from a remote MCP server before calling them
• Retrieve and fill prompt templates from an MCP server for AI workflows

FAQ from n8n-nodes-mcp-client

What n8n version is required?

The node requires n8n version 1.0.0 or later and is compatible with MCP Protocol version 1.0.0 or later.

Does it support SSE (Server‑Sent Events) transport?

Yes. The node supports both STDIO and SSE transports. SSE credentials require the SSE URL (default http://localhost:3001/sse) and optional custom endpoints and headers.

How do I enable the node as a tool in an AI Agent?

Set the environment variable N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true and restart n8n. This can be done via shell, Docker, or a .env file. After that the MCP Client node appears as a tool in AI Agent nodes.

How can I pass environment variables to an MCP server?

You can add them directly in the credentials UI (stored securely) or set them in Docker with the MCP_ prefix. For example, MCP_BRAVE_API_KEY=your‑key will be passed to the server when it runs.

What operations are available?

The node offers six operations: Execute Tool, Get Prompt, List Prompts, List Resources, List Tools, and Read Resource. Each operation is selected from a dropdown in the node settings.