n8n-MCP open source explodes: Use AI dialogue to directly build workflow, 19K Stars

Core conclusion

In May 2026, the open source project n8n-MCP accumulated 19,259 Stars on GitHub, becoming one of the fastest-growing AI automation tools this week. It is essentially a server side of the Model Context Protocol (MCP), providing AI assistants with read and write capabilities to all 1,650 of n8n’s workflow automation nodes.

To put it simply: You can now use natural language to say to AI, "Help me connect Slack notifications to Telegram, filter keywords and send emails." AI will automatically complete the construction of the n8n workflow, and the entire process does not require opening the n8n editor.

Key Points

• Release time: Neutral project in 2025, currently under active development, latest v2.50.0 (released yesterday)
• GitHub Stars: 19,259, Fork 3,220, 989 commits
• Supported clients: Claude Code, Cursor, Windsurf, VS Code Copilot, Codex
• Nodes covered: 1,650 (820 core + 830 community) with 265 AI tool variants
• Template library: 2,352 reusable workflow templates, AI metadata coverage 99.96%

Project background

n8n-MCP was created by developer czlonkowski. It was initially just a personal tool, but as the MCP protocol became popular in the field of AI Agents, it quickly became a key infrastructure linking AI and automated workflows. The project released v2.50.0 yesterday, adding local LLM template metadata support and template store refresh functions.

MCP (Model Context Protocol) is an open protocol proposed by Anthropic that allows AI models to call external tools and data sources through standard interfaces. n8n-MCP applies this protocol to n8n node management, allowing the AI ​​assistant to directly understand, search, and call 1,650 n8n nodes.

Core Competencies at a Glance

For practitioners of AI automation and content production, the greatest value of this project is: AI no longer just helps you write n8n configuration code, but directly controls n8n itself. You can ask Claude Code to search for n8n nodes, find templates, and verify workflows in the conversation, all without switching windows.

Hands-on guide: How to use n8n-MCP with Claude

Option 1: Hosted version (fastest, free of charge)

The project provides free hosting service dashboard.n8n-mcp.com, with 100 free calls per day. After registration, obtain the API Key and configure the MCP server in Claude Code.

Option 2: Self-hosting (suitable for production environment)

Use npx to start with one click:

npx n8n-mcp

Docker

docker run -i --rm -v ./data:/app/data ghcr.io/czlonkowski/n8n-mcp

IDE

n8n-MCP Claude Code、Cursor、Windsurf、VS Code Copilot Codex。 Cursor .cursor/mcp.json MCP ，。

AI

n8n-MCP 「」

1. ** AI ** — search_templates() 、、
2. **** — search_nodes() +
3. **** — minimal → full →

。 n8n ， AI 。

（）

⚠️ AI ！

n8n-MCP 。AI ，。

，Claude、Cursor、n8n、OpenAI、DeepSeek etc. are all tool entries maintained by the platform, and AI can automatically match hover-card display.

Internal link guidance

• Want to learn how to use n8n + ChatGPT to build an automated content distribution system? Watch: How to build an AI content automated distribution system with n8n + ChatGPT: a complete 30-minute tutorial
• Want to know more practical business cases of n8n paired with AI? Watch: Indie Developer: n8n + OpenClaw Automation Workflow Earning $5,000/mo