TouchGrass Buddy：用 Ollama 与 Llama 3.2 生成户外冒险的开源 AI 应用

AI-Powered Outdoor Adventure Generator

TouchGrass Buddy is a local AI-powered application that generates personalized outdoor adventures based on your available time, location, and interests.

Most AI applications are designed to keep users glued to screens. TouchGrass Buddy does the opposite by encouraging exploration, mindfulness, exercise, creativity, and real-world experiences.

Built using:

🐍 Python
🎈 Streamlit
🦙 Ollama
🤖 Llama 3.2

✨ Features

Generate personalized outdoor adventures
Multiple adventure styles
- Nature Exploration
- Photography
- Fitness
- Family Fun
- Mindfulness
Progress tracking
Mission completion tracking
Start new adventure anytime
Completely offline
No API keys required
No cloud dependencies
Privacy-friendly
Powered by open-source AI

🖼️ Screenshots

Home Screen

Generated Mission

Mission Complete

🚀 How It Works

Choose how much time you have available.
Choose where you're going.
Select an adventure type.
Generate a personalized outdoor mission using a local AI model.
Complete the challenges.
Track your progress.
Start a new adventure whenever you're ready.

💡 Why Open-Source AI?

For this project, open-source AI was a better choice than a closed AI service because the application can run entirely on a laptop without an internet connection. Users do not need API keys, subscriptions, or cloud access, and their prompts never leave the device. This makes the solution private, portable, and cost-effective.

Open AI Stack

Ollama for local inference
Llama 3.2 as the language model
Streamlit for the user interface

Benefits

✅ Runs completely offline

✅ No internet connection required

✅ No API costs

✅ No vendor lock-in

✅ No user data leaves the device

✅ Users remain in complete control of their data

✅ Models can be replaced or customized

This demonstrates how open-source AI can enable private, affordable, and accessible applications without relying on cloud-hosted AI services.

🌳 Touch Grass Challenge Alignment

This project was built for the Hacktoberfest 2026 Open-Source AI Challenge.

Theme: Touch Grass

The challenge encourages building applications that use open-source AI to help people spend less time online and more time engaging with the real world.

TouchGrass Buddy aligns with this goal by:

Using local open-source AI
Encouraging outdoor exploration
Promoting physical activity
Supporting mindfulness and nature appreciation
Keeping screen interaction brief and purposeful

The ideal user journey is:

Generate Mission
       ↓
Go Outside
       ↓
Complete Challenges
       ↓
Track Progress
       ↓
Start New Adventure

The screen should be the shortest part of the experience.

⚙️ Installation

Clone the Repository

git clone https://github.com/vhudlikar/touchgrass-buddy.git
cd touchgrass-buddy

Install Dependencies

pip install -r requirements.txt

Install Ollama

Download and install Ollama:

https://ollama.com

Pull Llama 3.2

ollama pull llama3.2

Start Ollama

ollama serve

Run the Application

streamlit run app.py

🛠️ Technology Stack

Python
Streamlit
Ollama
Llama 3.2
Git
GitHub

🔮 Future Enhancements

GPS-aware adventures
Mobile-friendly interface
Nature scavenger hunts
Achievement badges
Group challenges
Route-based adventures
Local park recommendations
Wellness and fitness integrations

🤝 Contributing

Contributions are welcome.

If you have ideas, improvements, or bug fixes:

Fork the repository
Create a feature branch
Commit your changes
Open a pull request

📜 License

This project is licensed under the MIT License.

See the LICENSE file for details.

👨‍💻 Author

Viraj Hudlikar

Built as part of Hacktoberfest 2026 to explore how local open-source AI can encourage people to spend more time outdoors and less time staring at screens.

🎃 Hacktoberfest 2026

Challenge Theme: Touch Grass

Using local open-source AI to create real-world outdoor adventures instead of more screen time.