n8n-mcp

Free

A MCP for Claude Desktop / Claude Code / Windsurf / Cursor to build n8n workflows for you

Model APIsFreeFree tier

Type

Open Source

Visit Website GitHub (19.3k stars)

About n8n-mcp

n8n MCP is a connector that allows AI tools like Claude, ChatGPT, Cursor, and 20+ others to directly interact with n8n workflow automation. It enables AI to build, validate, test, debug, deploy, and manage n8n workflows — all without copy-paste or manual import. The platform offers two deployment modes: a hosted service with one-click OAuth for quick setup, and an open-source self-hosted option (MIT licensed) for full control. Key capabilities include access to 2,000+ n8n nodes and 2,700+ templates, surgical diff editing, validation autofix, execution debugging, version rollback, and self-correction where AI catches its own mistakes. Privacy is maintained by keeping workflows and credentials in the user's n8n instance, with self-hosted AI screening for shared patterns.

Key Features

One-click OAuth connection from 20+ AI tools (Claude, ChatGPT, Cursor, etc.)

2,000+ n8n nodes and 2,700+ templates accessible to AI

Surgical diff editing for precise workflow modifications

Validation autofix that catches and corrects errors before deployment

Execution debugging with real execution logs

Version history and one-click rollback to previous snapshots

Self-correction: AI validates its own work and fixes mistakes

Live workflow context understanding — AI reads existing workflows for targeted improvements

Documentation synced with latest n8n releases within 48 hours

Private by design: credentials and data stay in user's n8n instance; self-hosted AI screening for pattern contributions

Pros & Cons

Pros

Free tier available (100 tool calls per day) for trying out
Open-source MIT licensed self-hosted version for full control
Works with 20+ popular AI tools out of the box
Direct deployment without copy-paste or import errors
Active community with tutorials and rapid star growth
Privacy-focused: workflow data stays in user's instance

Cons

Hosted free tier limited to 100 tool calls per day; higher usage requires paid upgrade
Requires an existing n8n instance to connect to
AI tools (e.g., Claude, ChatGPT) may require separate subscriptions

Best For

Build n8n workflows using natural language in Claude Desktop, ChatGPT, or WindsurfAutomate testing and debugging of n8n workflows with AIMaintain production workflows with rollback capabilities when failures occurDevelop AI agents that interact with n8n for complex automation tasks

Alternatives to n8n-mcp

AutoGPT

AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

aider

aider is AI pair programming in your terminal

axolotl

Go ahead and axolotl questions

Chroma

Open-source embedding database

awesome-claude-code

A curated list of awesome skills, hooks, slash-commands, agent orchestrators, applications, and plugins for Claude Code by Anthropic

agents-course

This repository contains the Hugging Face Agents Course.

FAQ

How does n8n MCP work?

It connects AI tools such as Claude, ChatGPT, and Cursor to your n8n instance. The AI can then build, validate, test, debug, deploy, and manage workflows directly, using over 2,000 nodes and 2,700 templates.

Is n8n MCP free?

Yes, there is a generous free tier offering 100 tool calls per day. The self-hosted open-source version is free forever under MIT license.

What AI tools are supported?

n8n MCP works with Claude, ChatGPT, Cursor, and 20+ other AI tools via one-click OAuth or API key.

Can I run n8n MCP on my own infrastructure?

Yes, you can self-host using npx (npx n8n-mcp) or Docker (docker run -i ghcr.io/czlonkowski/n8n-mcp). It connects to localhost n8n instances and has no quotas.

How does n8n MCP handle privacy?

Your workflows, executions, and credentials remain in your n8n instance. When the platform learns from usage, it stores only workflow patterns (not your data), and every contribution is screened by self-hosted AI models before ingestion.