🌍 EarthQuest AI — Open-Source AI That Gets You Off the Screen and Into the Real World
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass
What I Built
EarthQuest AI is a local-first AI-powered outdoor adventure companion designed to turn screen time into real-world exploration.
The idea is simple:
Use AI for a few seconds. Then put the phone away and go outside.
EarthQuest AI asks users about their available time, preferred activity, and optionally their location. It then generates a personalized outdoor quest using AI.
Instead of endlessly scrolling through apps, users receive a simple mission they can complete in the real world.
For example:
- 🌳 Explore a nearby green space and discover something you've never noticed.
- 🚶 Take a short walking adventure and document interesting things you find.
- 🌿 Observe plants, trees, or wildlife around you.
- 📸 Complete a photography challenge.
- 🧭 Create your own mini exploration challenge.
After completing the quest, users can return to EarthQuest AI, record their experience, and track their outdoor activity.
Who is it for?
EarthQuest AI is designed for:
- Students who spend too much time on screens
- Developers and tech enthusiasts
- People who want simple outdoor activities
- Anyone looking for a reason to go outside
- People who want to use AI without constantly being online
The main philosophy behind the project is:
The better you use EarthQuest AI, the less time you spend using EarthQuest AI.
Demo
🌐 Live Website:
https://aditya-narain-shukla.github.io/EARTHQUEST-AI/
The website allows users to create an outdoor quest, generate an AI-powered activity, start the quest, complete it, and reflect on their experience.
Code
💻 GitHub Repository:
https://github.com/aditya-narain-shukla/EARTHQUEST-AI
The project is built using simple web technologies so that it remains lightweight, understandable, and easy to run.
How I Built It
EarthQuest AI is built using:
- HTML — structure and interface
- CSS — responsive design, animations, cards, and visual styling
- JavaScript — application logic, quest management, timers, statistics, local storage, and AI communication
- Ollama — local AI inference
- Llama 3.2 — the open-weight AI model used to generate personalized outdoor quests
The browser communicates with Ollama running locally on the user's computer.
The AI receives information such as:
- Available time
- Preferred activity
- Optional location/context
- The goal of getting the user outdoors
It then generates a personalized outdoor quest.
I also included a fallback quest-generation system so the application can still provide useful experiences when the local AI model is unavailable.
The project intentionally avoids frameworks and unnecessary dependencies. The core application can be understood through three simple files:
index.html
style.css
script.js
This makes the project easy for beginners to understand, modify, and contribute to.
Why Does Open Innovation Matter?
Open innovation is especially important for a project like EarthQuest AI because the goal is not to create another service that constantly sends user data to a remote server.
With local AI through Ollama, the AI inference can happen directly on the user's own machine.
This provides several important advantages.
🔒 Privacy
Users can keep their activity preferences and optional location information on their own computer instead of automatically sending everything to a third-party AI service.
💰 Lower Running Cost
There is no per-request cloud AI API cost for generating quests when inference is performed locally.
🧠 Model Freedom
Because the project uses Ollama, users can experiment with different compatible open-weight models instead of being locked into a single proprietary AI provider.
🛠️ Customization
Developers can change the prompts, behavior, interface, and AI workflow according to their needs.
🌐 Less Dependence on the Cloud
Local inference makes it possible to build AI experiences that don't require every interaction to pass through a centralized cloud service.
Most importantly, open AI fits the philosophy of EarthQuest AI:
Technology should help people experience the real world, not keep them trapped inside a screen.
Instead of building another AI chatbot that encourages users to spend more time talking to AI, EarthQuest AI uses AI as a short bridge between the digital world and the physical world.
My Agent Session
The main EarthQuest AI experience is focused on the user's interaction with the local AI model.
Agent Session: Not included in this submission.
Prize Categories
🏆 Hacktoberfest Open-Source AI Challenge — Touch Grass
EarthQuest AI is specifically built around the Touch Grass theme.
The project uses open-weight AI to encourage people to:
Plan → Go Outside → Explore → Experience → Return → Reflect
The screen is intentionally designed to be only a small part of the experience.
What Makes EarthQuest AI Different?
Most AI applications try to increase engagement:
More messages → More screen time → More AI usage
EarthQuest AI tries to reverse that:
Short AI interaction → Real-world quest → Screen away → Outdoor experience
That is the core idea behind the project.
🌍 AI should not always keep us online.
Sometimes, the best AI experience is the one that tells you to close the laptop, put down your phone, and go outside.
Future Ideas
Some features I would like to add in future versions include:
- 🐦 Offline bird identification
- 🌱 Plant and tree identification
- 🗺️ Smarter outdoor route generation
- 🏃 Personalized running and walking challenges
- 🌦️ Weather-aware quests
- 📍 More advanced local exploration
- 🎮 Achievement and reward systems
- 👥 Community-generated quests
- 🤖 Support for additional local AI models
- 📱 Progressive Web App support
- 🔌 More complete offline functionality
Final Thought
EarthQuest AI started with a simple question:
What if we used AI to help people spend less time with technology instead of more?
The result is a small experiment in using open
Top comments (0)