🌿 TrailMate AI: Touch Grass, Explore Nature, and Let AI Get Out of Your Way

Roshan Verma — Fri, 09 Oct 2026 06:00:43 +0000

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

What I Built

🌿 TrailMate AI is an outdoor exploration companion designed to help people spend less time on screens and more time in the real world.

Instead of keeping users engaged in an app, TrailMate AI encourages them to step outside, explore their surroundings, discover nature, and complete fun outdoor challenges.

Key features include:

🧭 AI Adventure Generator: Creates outdoor missions that encourage users to observe and explore nature.
🌱 Nature Scanner: Lets users upload photographs of plants, leaves, flowers, and other natural objects.
📵 Phone Down Mode: Encourages users to put their phones away and enjoy screen-free outdoor time.
🏆 Gamification: Rewards completed challenges and discoveries with XP, levels, and achievement badges.
📊 Adventure Dashboard: Tracks recorded discoveries, challenges, and outdoor progress.
💾 Local Progress Storage: Saves progress in the browser using LocalStorage.

TrailMate AI is for students, nature enthusiasts, hikers, and anyone who wants to build healthier digital habits.

Our philosophy: The best feature of an outdoor app is knowing when to stop being useful.

Demo

🚀 Live Demo: https://roshanverma19.github.io/-TrailMate-AI-/

The demo showcases the outdoor challenge system, nature scanner interface, achievement system, and Phone Down Mode.

Code

💻 GitHub Repository: https://github.com/RoshanVerma19/-TrailMate-AI-.git

Built using:

HTML5 for structure
CSS3 for responsive styling
JavaScript for application logic and interactions
LocalStorage for saving progress in the browser

No frontend framework or custom backend is required for the current prototype.

How I Built It

I built TrailMate AI using HTML, CSS, and vanilla JavaScript, keeping the application lightweight and accessible.

JavaScript powers the adventure challenge system, XP rewards, achievement unlocking, image uploads, countdown timer, and local progress storage.

The Nature Scanner currently uses simulated results rather than a genuine AI model. My next step is to integrate an open-weight vision model that can perform real image recognition directly in the browser using a JavaScript-compatible inference runtime.

The planned AI integration will allow users to explore nature while reducing dependence on proprietary AI services.

Why Does Open Innovation Matter?

Open innovation makes AI more accessible, customizable, and privacy-conscious.

For TrailMate AI, my goal is to use an open-weight model that can run locally in the browser. This could allow users to analyze nature photographs without uploading them to an external AI server.

An open approach can provide several advantages:

Privacy: Local inference can keep personal photographs on the user's device.
Offline potential: A downloaded model could support exploration without an internet connection.
Flexibility: Developers can experiment with different compatible open-weight models.
Accessibility: The project can be extended using familiar web technologies.
Customization: AI-powered challenges and nature exploration can be adapted to different environments.

These are goals for the planned AI integration; the current prototype does not yet include real open-weight AI inference.

I believe AI should encourage people to engage more deeply with the world around them, rather than simply creating another reason to stare at a screen.

My Agent Session

🤖 Agent session: [Add your DevRelay session link here, if available]

Prize Categories

Select the applicable partner prize categories according to the official challenge rules.

🌿 TrailMate AI — Explore More. Scroll Less.

DEV Community: Roshan Verma