Excel Automator MCP

This is a Model Context Protocol (MCP) server that enables LLMs to interact with Google Sheets. It provides tools to read, write, update, and delete rows in a Google Sheet.

Problem Solved

This project allows AI agents to directly manipulate data in Google Sheets, enabling automation of tasks like inventory management, data entry, and report generation directly from a chat interface.

Prerequisites

• Node.js installed
• A Google Cloud Service Account with access to the Google Sheets API
• A service.json credential file from your Google Cloud project

Setup

1. Install Dependencies

   npm install

2. Configure Credentials Place your Google Cloud service account credentials file named service.json in the src/ directory.

   Note: The application expects service.json to be available to the running script. Ensure it is copied to the build directory or accessible.

3. Build the Project

   npm run build

   You may need to manually copy src/service.json to build/service.json if the build script doesn't handle it.

Usage

To run the server:

node build/index.js

Available Tools

• read-sheet: Read all rows from the configured Google Sheet.
• write-sheet: Append a new row (specifically formatted for inventory: apples, bananas, oranges, grapes).
• update-sheet-row: Update a specific row.
• delete-sheet-row: Delete a specific row.

Configuration

The spreadsheet ID is currently hardcoded in src/index.js. You may need to update SPREADSHEET_ID with your own sheet ID.