概要
What is N8N MCP Server?
This MCP server acts as a bridge between AI assistants (like Claude) and the N8N automation platform. It provides a simplified interface to access N8N functionality through MCP endpoints for listing, creating, updating, and deleting workflows, executions, credentials, and tags.
How to use N8N MCP Server?
Clone the repository, install dependencies, configure environment variables (N8N_API_BASE_URL, N8N_API_KEY, PORT), and start the server with npm start. To use with Claude, run install_local_mcp_server path="./n8n-mcp-server".
Key features of N8N MCP Server
- Manage workflows: list, get, create, update, delete, activate, deactivate
- Handle executions: list, get, delete, execute workflows with optional data
- Manage credentials: list, get, create, update, delete
- Manage tags: list, create, delete
- Provides a simplified MCP interface over the N8N API
Use cases of N8N MCP Server
- Allow an AI assistant to list and manipulate N8N workflows via natural language
- Execute N8N workflows with custom data from a conversation
- Automate credential and tag management through chat interfaces
FAQ from N8N MCP Server
What are the requirements to run this server?
Node.js 16+, an N8N instance with API access, and a valid N8N API key.
How do I configure the connection to N8N?
Set the N8N_API_BASE_URL environment variable (default http://localhost:5678/api/v1) and your N8N_API_KEY in a .env file.
What data does this server access?
It accesses workflows, executions, credentials, and tags via the N8N API. All data lives on your N8N instance.
Is authentication handled?
Yes, the server uses the N8N API key provided in the environment variables for authentication.
What transport does this server use?
It exposes MCP endpoints over HTTP (default port 3000). The README warns against exposing it to the public internet without additional authentication.
