GitHub - LaunchPlatform/beangrid: Open-source LLM powered plaintext based spreadsheet app

4 min read Original article ↗

BeanGrid

🚧 Prototype Alert: This is a prototype created by vibe coding. It's an experimental plaintext-based spreadsheet application that demonstrates the potential of plaintext-based spreadsheets in the AI era.

A plaintext spreadsheet application with real-time collaboration and AI assistance, built for the modern AI-powered workflow.

Demo

BeanGrid.-.Plaintext.Spreadsheet.-.Brave.2025-07-26.16-14-56.mp4

Why Plaintext-Based Spreadsheets?

1. Open Source & Vendor Independence

Excel and other traditional spreadsheets are closed-source and vendor-locked. We need open-source solutions that give users full control over their data and workflows.

2. AI-Native Design

In the AI era, plaintext-based spreadsheets are infinitely more accessible for LLMs to read, understand, and update. No more complex binary formats or proprietary APIs - just clean, human-readable YAML that AI can easily parse and modify.

3. Git Integration & Version Control

Keeping your spreadsheets in Git is now seamless. Track every change, see the complete history of modifications, and collaborate with confidence. Never lose your work again.

4. Open Source Collaboration

Share and discover spreadsheets like never before. Common workbooks like tax calculations, financial models, and business templates should be available as open-source sheets that anyone can reference, modify, and contribute to.

Features

  • 📝 Plaintext YAML Format: Store spreadsheets in human-readable YAML files
  • 🧮 Excel-like Formulas: Full formula support with cross-sheet references
  • 🤖 AI Assistant: Real-time chat with Grok for spreadsheet analysis and updates
  • 🌐 Modern Web Interface: React-based UI with real-time updates
  • 📊 Multi-sheet Support: Organize data across multiple sheets
  • 🔗 Git Integration: Version control for your spreadsheets
  • 💬 Real-time Chat: AI-powered assistance for complex calculations

YAML-Based Spreadsheet Example

Here's what a spreadsheet looks like in YAML format:

sheets:
  - name: Sales
    cells:
      - id: A1
        value: Product
      - id: B1
        value: Price
      - id: C1
        value: Quantity
      - id: D1
        value: Total
      - id: A2
        value: Widget A
      - id: B2
        value: "10.50"
      - id: C2
        value: "4"
      - id: D2
        formula: "=B2*C2"
      - id: A3
        value: Widget B
      - id: B3
        value: "15.75"
      - id: C3
        value: "6"
      - id: D3
        formula: "=B3*C3"
      - id: A4
        value: Total
      - id: B4
        value: ""
      - id: C4
        value: ""
      - id: D4
        formula: "=SUM(D2:D3)"
  - name: Summary
    cells:
      - id: A1
        value: Summary Report
      - id: A2
        value: Total Sales
      - id: B2
        formula: "=Sales!D4"
      - id: A3
        value: Average Price
      - id: B3
        formula: "=AVERAGE(Sales!B2:B3)"

Formula Examples

# Basic arithmetic
- id: C1
  formula: "=A1+B1"

# Cross-sheet references
- id: B2
  formula: "=Sales!D4"

# Functions
- id: D4
  formula: "=SUM(D2:D3)"

# Complex calculations
- id: E1
  formula: "=(A1+B1)*C1/100"

Configuration

LLM Settings

The application uses Grok for AI assistance. You can configure the LLM settings using environment variables:

# Set the LLM model (Grok)
export LLM_MODEL="xai/grok-3"

# Set the LLM API base URL (X.AI API)
export LLM_API_BASE="https://api.x.ai/v1"

# Set your X.AI API key
export LLM_API_KEY="your-xai-api-key-here"

Or you can set these in a .env file:

LLM_MODEL=xai/grok-beta
LLM_API_BASE=https://api.x.ai/v1
LLM_API_KEY=your-xai-api-key-here

Note: You'll need to obtain an API key from X.AI to use Grok.

Installation

Prerequisites

  • Python 3.8+
  • Git (for version control features)
  • X.AI API key (for AI features)

Quick Start

  1. Clone the repository

    git clone https://github.com/LaunchPlatform/beangrid.git
cd beangrid

  2. Install dependencies

  3. Set up your API key

    export LLM_API_KEY="your-xai-api-key-here"

  4. Run the server

    uv run python run_server.py

  5. Open your browser Navigate to http://localhost:8000

Usage

Creating a New Session

Click the "🆕 New Session" button in the top toolbar to create a fresh workspace with a new YAML file.

Editing Cells

  1. Click on any cell to select it
  2. Use the formula bar at the bottom to edit values or formulas
  3. Press Enter to save changes

AI Assistance

  1. Open the chat sidebar on the right
  2. Ask questions about your data or request updates
  3. The AI can suggest cell updates or entire workbook modifications

Version Control

  • All changes are automatically tracked in Git
  • Use the "Diff" tab to see what's changed
  • Commit changes with meaningful messages

Development

This is a prototype built with:

  • Backend: FastAPI + Python
  • Frontend: React + JavaScript
  • AI: Grok via X.AI API
  • Storage: YAML files with Git version control

Contributing

Since this is a prototype, we're exploring the concept of plaintext-based spreadsheets. Ideas and feedback are welcome!

License

MIT License - feel free to use this as inspiration for your own plaintext spreadsheet projects.

Exploring the future of spreadsheets in the AI era