n8n-mcp-node-reneworks

This is an n8n community node that lets you use an MCP-enabled Chat API as a Language Model in n8n.

Installation

For Local n8n

1. Go to your n8n root directory (e.g. ~/.n8n).
2. Create a custom directory if it doesn't exist: mkdir custom.
3. Clone or copy this repository into custom/n8n-mcp-node-reneworks.
4. Inside custom/n8n-mcp-node-reneworks, run npm install and npm run build.
5. In ~/.n8n, run npm install ./custom/n8n-mcp-node-reneworks.
6. Start n8n: n8n start.

Installation via NPM / Git

If you have this node in a Git repository or as an NPM package:

1. From Git:

   npm install git+https://github.com/YourUsername/n8n-mcp-node-reneworks.git

   Run this command in your ~/.n8n directory.

2. From a Tarball (.tgz): If you have the .tgz file (e.g., n8n-mcp-node-reneworks-0.1.0.tgz):

   npm install /path/to/n8n-mcp-node-reneworks-0.1.0.tgz

   Run this command in your ~/.n8n directory.

3. Publishing to NPM: If you want to publish this to the npm registry:

   npm login
   npm publish --access public

   Then anyone can install it via:

   npm install n8n-mcp-node-reneworks

You need to build a custom Docker image or mount the volume. Use the n8n-nodes-starter guide for details on mounting.

Usage

1. Open your n8n workflow.
2. Add an AI Agent node.
3. Add the MCP Chat Model node.
4. Connect the MCP Chat Model output to the Model input of the AI Agent.
5. Configure the MCP Chat Model:
   • Base URL: The address of your MCP Chat API (default: http://localhost:1234).
   • Model Name: The model identifier (e.g., ibm/granite-4-micro).
   • Integrations: Paste the JSON configuration for your MCP servers/plugins.

Example Integrations JSON

[
  {
    "type": "ephemeral_mcp",
    "server_label": "huggingface",
    "server_url": "https://huggingface.co/mcp",
    "allowed_tools": [
      "model_search"
    ]
  },
  {
    "type": "plugin",
    "id": "mcp/playwright",
    "allowed_tools": [
      "browser_navigate"
    ]
  }
]