n8n-mcp-Automation

No description

GitHub Stars

0

User Rating

Not Rated

Favorites

0

Views

27

Forks

0

Issues

0

README
AI Automation with n8n, Docker CLI, and mCP – Step-by-Step Guide

Hey everyone! 🚀
In this post, you’ll learn how to install n8n using Docker CLI (no Docker Desktop required) and take your automation skills to the next level using mCP (Model Context Protocol).

Imagine your AI being able to connect with any tool or API automatically – no complicated code, no headaches. That’s what mCP does – it makes automation 10x easier and smarter.


🔧 What You’ll Learn

✅ Install n8n for free using Docker CLI
✅ Set up mCP in n8n to unlock AI superpowers
✅ Let AI dynamically choose and execute tools
✅ Automate web searches and scraping
✅ Bonus: Learn a powerful trick even experts miss!


🚀 Step 1: Install Docker CLI

If you haven’t already, install Docker Engine on your system:

  • Linux: Follow your distro’s instructions
  • macOS & Windows: Install Docker Desktop, then use CLI commands (Docker Desktop still required on these platforms for the engine)

To verify Docker is working:

bash
docker --version


🐳 Step 2: Run n8n Container Using Docker CLI
1. Create a local folder for storing n8n data

bash
mkdir -p ~/n8n-data

2. Run n8n container with required settings

bash
docker run -it --name n8n-container
-p 5678:5678
-v ~/n8n-data:/home/node/.n8n
-e N8N_ENABLE_COMMUNITY_NODES=true
n8nio/n8n

Explanation:

  • -it: Run interactively
  • --name: Name your container
  • -p 5678:5678: Map local port 5678 to container
  • -v: Mount volume for data persistence
  • -e: Set environment variable to enable community nodes
  • n8nio/n8n: Official n8n Docker image

Once the container starts, open your browser and go to:


Screenshot 2025-05-14 212600


http://localhost:5678 (aws ip address)

You should see the n8n signup page!


Screenshot 2025-05-14 212319


🧠 Step 3: Create a New Account and Start Your Workflow
  1. Sign up with your email and password
  2. Click to create a new workflow
  3. You’re now inside n8n and ready to build!

Screenshot 2025-05-14 212549


🧩 Step 4: Install mCP Nodes in n8n
  1. Go to Settings → Community Nodes
  2. Search for n8n-nodes-mcp
  3. Click Install
  4. Wait for installation to finish

Now you can use mCP in your automations!


🛠 Step 5: Create Your First AI Agent Workflow
Add Nodes:
  • Chat Trigger Node – To start on user input
  • AI Agent Node – Set up with OpenAI (add your API key)
  • Simple Memory Node – Helps maintain conversation history

Now let’s bring in the mCP power!


Screenshot 2025-05-14 222855


🔌 Step 6: Connect mCP with n8n
Add mCP Client Node:
  1. Action: List Tools
  2. Set up credentials using Brave API
  3. use this repository to setup brave using api: https://github.com/Swayamnakshane/servers.git
  4. Use CLI or browser to get your API key from Brave Search

bash

This is an example of mCP setup inside n8n via node config
Fill in API key and Brave server name in node settings

Test the node. It should show available tools from the Brave mCP server (like search, details, etc.)


🤖 Step 7: Let the AI Use Those Tools
  1. Add a new mCP node to Execute Tool

  2. Choose a tool like “search”

  3. Let AI define parameters automatically

  4. Add a System Message like:

    "You are a helpful assistant using Brave search to answer questions."

Make sure all nodes are connected correctly.


🧪 Step 8: Run and Test!

Open the chat input and ask:

“Find a nearby coffee shop.”

The workflow will:

  • Capture your question
  • AI will decide to use Brave Search
  • Get the result
  • Format and show it back to you

Screenshot 2025-05-14 223633


🎉 Conclusion

By using Docker CLI + n8n + mCP, you’ve created an AI system that can use real tools, make decisions, and fetch live data — all with zero code.

connect with me :
linkedin: Swayamnakshane4

https://github.com/user-attachments/assets/0876df7e-c508-450d-b918-b1908b2417da