This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass
What I Built
WildQuest AI 🌿 — Less Scrolling. More Exploring.
What if technology helped us spend less time staring at a screen and more time discovering the world around us?
That is the idea behind WildQuest AI: a mobile-friendly nature adventure companion that turns an ordinary walk into an opportunity to explore, observe, and learn.
Instead of endless scrolling, users choose a 15-, 30-, or 60-minute outdoor quest across six categories: plants, birds, trees, mindful walking, biodiversity, and litter awareness.
- Nature Quests: Step-by-step outdoor activities with difficulty levels, equipment, completion criteria, and safety guidance.
- Screen-Off Pocket Mode: A distraction-minimizing experience that encourages users to put their phones away and focus on their surroundings.
- Nature Field Journal: Save observations, notes, and photographs locally on the device.
- Explorer Dashboard: Track completed quests, outdoor minutes, streaks, and achievement badges.
- Offline-Friendly Adventures: Revisit cached quests and locally saved journal entries without a constant internet connection.
The goal is simple: make the screen the shortest part of an outdoor adventure.
Demo
Try WildQuest AI: https://github.com/ankit20012006/Wildquest-ai)
Code
GitHub Repository: https://github.com/ankit20012006/Wildquest-ai))
How I Built It
WildQuest AI uses a modern full-stack web architecture:
- React, TypeScript, and Vite for the frontend.
- Node.js, Express, and TypeScript for the backend.
- Progressive Web App technologies for installation and offline caching.
- IndexedDB and local storage for local-first journal entries and progress.
- Structured validation and safety checks to validate generated quests and reject unsafe instructions.
- Backboard integration for configurable model-based quest generation.
The application currently includes curated seed quests that keep the core experience useful when remote AI generation is unavailable.
The local automated test suite passed 18 tests, and the frontend production build and backend TypeScript compilation completed successfully.
The Backboard request integration has been updated to follow the documented request structure. However, a successful live inference response from a verified open-weight model has not yet been confirmed. The current demo fallback should not be mistaken for a response generated by an LLM.
Why Does Open Innovation Matter?
Outdoor exploration should not depend entirely on a proprietary AI service being available every time someone steps outside.
An open-weight approach offers flexibility: developers can evaluate different models, inspect their licences, choose appropriate inference providers, and adapt model behaviour to a specific use case.
For WildQuest, these principles influence three important design decisions.
Model flexibility: The AI integration is designed to support a configurable model rather than permanently tying the application to one model. This makes experimentation and replacement possible once a compatible open-weight model is verified.
Privacy-conscious journaling: Personal observations and photographs can remain on the user's device. Any future cloud-based AI analysis should send only explicitly selected information, with user consent.
Offline resilience: Cached adventures and locally stored journal entries remain useful without connectivity. Remote AI generation still needs a network connection unless local inference is separately implemented.
Open innovation is valuable because it gives developers more control over how AI systems are configured, deployed, and adapted. It also encourages us to be transparent about what runs locally and what relies on a remote service.
WildQuest AI explores how those principles can support a healthier relationship with technology: using digital tools to get people outside, rather than keeping them online.
My Agent Session
Optional: Add a DevRelay agent-session link if you have saved one. Otherwise, remove this section.
Top comments (0)