GridSnap

A Python web application that uses Google's Gemini Vision AI to extract data from screenshots of tables and imports it directly into Smartsheet.

What It Does

GridSnap bridges the gap between visual data and Smartsheet by:

• Using Gemini AI to identify and extract table structures from images
• Automatically detecting column headers and data types
• Formatting and transferring the extracted data to Smartsheet
• Providing a simple browser-based interface for the entire process

Features

• Upload screenshots of tables (Excel, PDFs, etc.)
• Verify Smartsheet ID and connection
• Extract column headers and row data
• Preview data before pushing to Smartsheet
• Clean interface with drag-and-drop support

Prerequisites

• Python 3.7 or higher
• Google Gemini API key - Get it here
• Smartsheet API token - Available in your Smartsheet account under Account > Personal Settings > API Access

Setup

1. Clone the repository

2. Create a virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Set up environment variables

   • Copy .env.template to .env
   • Add your API keys:

     GEMINI_API_KEY=your_gemini_api_key_here
     SMARTSHEET_API_TOKEN=your_smartsheet_api_token_here
     

5. Run the application

   python main.py

6. Access the app

   • Open http://localhost:5001

Usage Guide

1. Enter your Smartsheet ID and verify access
2. Upload a screenshot of your table
3. Add context if needed (optional)
4. Review the extracted data
5. Push to Smartsheet with one click

Limitations

• Works best with simple grid structures
• Does not support merged cells or complex formatting
• Accuracy depends on image quality

Future Enhancements

• Interactive table preview
• CSV export option
• Support for more complex table structures