MCP-Server

by sushant1827

297 downloads
Not rated
GitHub

About

This repository contains workflow configurations for a Model Context Protocol (MCP) server that integrates AI capabilities with n8n automation platform.

Details

Author
sushant1827
Downloads
297
Categories
Other

- AI-powered social media post generation via GPT-4o-mini
- Todo list management (list, add, update) through Airtable
- Built-in calculator tool for mathematical operations
- Extensible architecture for adding more AI tools

Import the two provided workflow JSON files into your n8n instance (version 1.0 or higher), configure OpenAI API credentials and an Airtable Personal Access Token, and activate both workflows. Interact with the server via its webhook endpoint at /test using the available tools.

MCP-Server

This repository contains workflow configurations for a Model Context Protocol (MCP) server that integrates AI capabilities with n8n automation platform.

Overview

The project consists of two main components:

1. MCP Server - A server that exposes AI tools through n8n's MCP Trigger and connects various services.
2. Social Media Post Generator - A workflow that uses AI to create professional social media content.

Features

- AI-powered social media post generation
- Todo list management via Airtable
- Calculator functionality
- Extensible architecture for adding more AI tools

Workflows

MCP Server

image

The MCP Server workflow acts as the central hub for all AI functionality. It:

- Exposes a webhook endpoint at /test
- Provides access to the following tools:
- Calculator
- Social Media Post Generator
- Airtable Todo List Management (List, Add, Update)

Social Media Post Generator

image

This workflow is designed to:

- Accept a query input
- Use OpenAI's GPT-4o-mini model to generate professional social media content
- Return the generated content to the calling workflow

Requirements

- n8n version 1.0.0 or higher
- OpenAI API account
- Airtable account with appropriate permissions

Setup Instructions

1. Import Workflows
- Import MCP_Server.json and Social_Media_Post_Generator.json into your n8n instance

2. Configure Credentials
- Set up OpenAI API credentials for the Chat Model node
- Configure Airtable Personal Access Token for the Airtable Tool nodes

3. Airtable Setup
- Ensure you have a base with ID appEVZyRnoMDAkayU (or update the configuration to match your base)
- Create a table named "Table 1" with columns:
- Name (string)
- Completed (boolean)

4. Activate the Workflows
- Activate both workflows in your n8n instance

Usage

Using the MCP Server

The MCP Server exposes various AI capabilities through the n8n interface. You can interact with it by:

1. Creating Social Media Posts

   Use the create_social_media_content tool with a query parameter

2. Managing Todo Items

   - List all todos using the List_ToDos tool
- Add a new todo with the Add_ToDo tool (provide Name and Completed status)
- Update an existing todo with the Update_ToDo tool (provide record ID, Name, and Completed status)

3. Performing Calculations

   Use the calculator tool for mathematical operations

Technical Details

Workflow Connections

- The MCP Server Trigger connects to all available tools
- The Social Media Post Generator is triggered by the Social Post Generator tool
- All Airtable operations are configured as separate tools

Configuration Notes

- The Social Media Post Generator uses the GPT-4o-mini model
- Airtable interactions are configured for a specific base and table

Contributing

Contributions are welcome! Here are some ways you can contribute:

- Add new AI tools to the MCP Server
- Improve existing tool functionality
- Create documentation and usage examples
- Fix bugs and optimize performance

Acknowledgements

- This project uses n8n (https://n8n.io) as the automation platform
- AI capabilities are powered by OpenAI's models
- Data storage is handled by Airtable

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.