n8n

The n8n-MCP server provides your AI assistants, like Claude Desktop, with direct access to n8n’s node library.

You can use it to have an AI generate, validate, and even deploy n8n workflows.

Features

• 🔍 Smart Node Search: Find n8n nodes by their name, function, or category.
• ✅ Configuration Validation: Check your node setups for errors before you deploy them.
• ⚡ Partial Workflow Updates: Modify existing workflows with diff operations, saving tokens and time.
• 📄 Essential Properties: Request just the key properties for a node, not the entire complex object.
• 🚀 Multiple Deployment Options: Run it instantly with NPX, use an isolated Docker container, or deploy to the cloud.
• 🔧 Direct n8n Management: Optionally connect your n8n API key to create, update, and manage workflows from your AI assistant.

How to Use It

Option 1: Run with npx

This approach runs the latest version of n8n-MCP directly without a permanent installation.

Open your Claude Desktop configuration file. The location depends on your operating system:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

Add the following JSON to the mcpServers object. This basic setup enables documentation and validation tools.

{
  "mcpServers": {
    "n8n-mcp": {
      "command": "npx",
      "args": ["n8n-mcp"],
      "env": {
        "MCP_MODE": "stdio",
        "LOG_LEVEL": "error",
        "DISABLE_CONSOLE_OUTPUT": "true"
      }
    }
  }
}

To enable tools that manage your n8n instance directly (like creating or updating workflows), add your n8n API URL and key.

{
  "mcpServers": {
    "n8n-mcp": {
      "command": "npx",
      "args": ["n8n-mcp"],
      "env": {
        "MCP_MODE": "stdio",
        "LOG_LEVEL": "error",
        "DISABLE_CONSOLE_OUTPUT": "true",
        "N8N_API_URL": "https://your-n8n-instance.com",
        "N8N_API_KEY": "your-api-key"
      }
    }
  }
}

Restart the Claude Desktop application for the changes to take effect.

Option 2: Run with Docker

This method uses a pre-built Docker image. It does not require Node.js, only Docker.

Pull the latest image from the GitHub Container Registry.

docker pull ghcr.io/czlonkowski/n8n-mcp:latest

Next, edit your claude_desktop_config.json file. The -i and --init flags are important for proper communication and container lifecycle management.

For documentation and validation tools only:

{
  "mcpServers": {
    "n8n-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--init",
        "-e", "MCP_MODE=stdio",
        "-e", "LOG_LEVEL=error",
        "-e", "DISABLE_CONSOLE_OUTPUT=true",
        "ghcr.io/czlonkowski/n8n-mcp:latest"
      ]
    }
  }
}

To add direct n8n management capabilities, include your API credentials as environment variables.

{
  "mcpServers": {
    "n8n-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--init",
        "-e", "MCP_MODE=stdio",
        "-e", "LOG_LEVEL=error",
        "-e", "DISABLE_CONSOLE_OUTPUT=true",
        "-e", "N8N_API_URL": "https://your-n8n-instance.com",
        "-e", "N8N_API_KEY": "your-api-key",
        "ghcr.io/czlonkowski/n8n-mcp:latest"
      ]
    }
  }
}

Save the file and restart Claude Desktop.

Available Tools

• tools_documentation: Provides information on how to use all other tools. Start here.
• list_nodes: Lists available n8n nodes. You can filter by category.
• get_node_essentials: Fetches only the most important properties for a node.
• search_nodes: Searches for nodes based on a keyword or description.
• validate_node_operation: Validates a node’s configuration for a specific operation.
• validate_workflow: Performs a full validation of a workflow, including connections.
• n8n_create_workflow: Creates a new workflow in your n8n instance (requires API key).
• n8n_update_partial_workflow: Applies small changes to an existing workflow (requires API key).

FAQs

Q: Do I need to provide my n8n API key?
A: No, the API key is optional. You only need it if you want the AI to directly create, read, update, or delete workflows in your n8n instance. All documentation, search, and validation tools work without it.

Q: Can this tool accidentally modify my live workflows?
A: Yes. If you provide your n8n API credentials, the AI will have the ability to change your workflows. You should always test on copies or in a development environment first. Never let an AI edit production workflows directly without review.

Q: What is the main difference between the npx and Docker installation?
A: npx is a command-line tool that runs a package from the npm registry without installing it locally, but it requires you to have Node.js installed. Docker runs the server in a self-contained, isolated environment, which is cleaner but requires you to have Docker installed and running.

Q: Can I use this with an n8n instance that is also running in Docker?
A: Yes. When configuring the N8N_API_URL in the MCP server’s Docker command, use http://host.docker.internal:5678 to allow the MCP container to communicate with the n8n container on the same machine.