TerraAI: Touch Grass — An Open-Source AI That Turns Screen Time into Green Time

Astha Soni — Sun, 11 Oct 2026 13:10:16 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

Demo

Code

How I Built It

Why Does Open Innovation Matter?

My Agent Session

Prize Categories

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

🌿 TerraAI — Touch Grass: Less Screen Time, More Green Time

I built TerraAI, a web-based outdoor activity planner designed to help people take a break from their screens and reconnect with nature.

In a world where we spend hours studying, working, and scrolling on our devices, sometimes the best thing technology can do is help us step away from it.

TerraAI encourages users to spend more time outdoors by suggesting simple activities based on their mood, available time, and interests.

🌱 What can you do with TerraAI?

🚶 Discover mindful walking activities.
🌿 Find simple gardening ideas.
🐦 Explore birdwatching activities.
🧘 Take a mindful outdoor break.
⏱️ Choose an activity that fits your available time.
🌙 Switch between light and dark themes.

The project is designed for students, people working at desks, nature lovers, and anyone who wants to spend less time scrolling and more time experiencing the world.

My goal: Make the screen the shortest part of the outdoor experience.

Demo

🌐 Try TerraAI here:

https://asthasonimicrosoft-code.github.io/terra-ai/

Open the website, enter how you're feeling, select your available time, choose an activity, and discover your next outdoor adventure!

Code

💻 GitHub Repository:

https://github.com/asthasonimicrosoft-code/terra-ai

The project uses three main files:

index.html — Website structure and planner interface.
style.css — Visual design, responsive layout, and themes.
script.js — Activity recommendation logic and AI integration.

The repository is publicly available, and I welcome suggestions and contributions that make TerraAI more useful.

How I Built It

I built TerraAI using HTML, CSS, and JavaScript to keep the project lightweight, accessible, and easy to run in a web browser.

The project explores browser-based AI through Transformers.js and an open-weight model from the Hugging Face ecosystem. It also includes a JavaScript fallback recommendation system for situations where the AI model cannot load.

Here's how the main components work:

Frontend: HTML provides the page structure, while CSS handles the outdoor-inspired design and responsive interface.
Interactions: JavaScript collects the user's mood, available time, and activity preference.
AI exploration: Transformers.js provides a way to load compatible machine-learning models directly in the browser.
Fallback: If model loading fails, the application can still suggest an activity using its built-in recommendation logic.
Deployment: GitHub Pages hosts the website, so users can access it through a public link.

I deliberately chose a browser-based approach instead of requiring users to install Ollama or run a separate Python backend.

The current version is a prototype, and the AI model integration is experimental. The model requires an internet connection to download its files unless a compatible model has already been cached. The fallback recommendations do not themselves require an AI model.

Why Does Open Innovation Matter?

For me, open innovation means being able to experiment, learn, and build without being locked into a single proprietary AI service.

TerraAI explores several important advantages:

1. Accessibility for beginners

Using open-source web libraries and available open-weight models makes AI experimentation more approachable for developers who are still learning.

2. More flexibility

I can experiment with compatible models, compare their performance, and change the implementation as the project grows.

3. Fewer infrastructure requirements

A browser-based approach means I don't need to maintain a dedicated AI server for this prototype or ask users to install Ollama.

4. More control over the experience

The application combines AI experimentation with my own interface and recommendation logic, allowing me to shape how the technology serves the user's needs.

5. Technology with a real-world purpose

The most important part of TerraAI isn't keeping users engaged with another screen. It's using technology to encourage real-world experiences.

Open-weight models still have their own licences, hardware requirements, and limitations, and browser-based AI may require internet access. However, open innovation gives me the freedom to explore alternatives and learn how these systems work.

My Agent Session

I don't have a DevRelay agent-session link to share yet. I'll add it here if I publish a session documenting the development process.

Prize Categories

Primary theme: Touch Grass — outdoor activities and real-world experiences supported by open-source AI.

TerraAI explores browser-based AI and open-weight models to help users discover activities beyond their screens.

Note: This submission focuses on the challenge theme. Any additional partner prize categories should be selected only if the project meets their specific eligibility requirements.

🌿 Less Screen Time. More Green Time.

TerraAI is my attempt to explore a simple idea: technology doesn't always have to keep us online. Sometimes, it can help us find a reason to go outside.

If you like the idea, check out the project, try the demo, and share your feedback!

Hacktoberfest #OpenSourceAI #TouchGrass #TerraAI #JavaScript #TransformersJS #HuggingFace

DEV Community: Astha Soni

TerraAI: Touch Grass — An Open-Source AI That Turns Screen Time into Green Time

What I Built

Demo

Code

How I Built It

Why Does Open Innovation Matter?

My Agent Session

Prize Categories

What I Built

Demo

Code

How I Built It

Why Does Open Innovation Matter?

My Agent Session

Prize Categories

Hacktoberfest #OpenSourceAI #TouchGrass #TerraAI #JavaScript #TransformersJS #HuggingFace