This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass
What I Built
I built TouchGrass AI, a local AI-powered outdoor activity planner.
The idea is simple:
Let AI plan it. You go outside. 🌱
TouchGrass AI helps people turn their free time into simple outdoor activities instead of spending more time on their screens.
Users can enter their:
- Activity type
- Available time
- Energy level
- Environment
- Solo or group preference
- Personal interests
- Optional location or context
The AI then generates a personalized outdoor mission.
For example, a user can enter:
- Activity: Walking
- Time: 30 minutes
- Energy: Medium
- Environment: College campus
- Interest: Nature
The application can generate a mission such as:
30-Minute Campus Nature Hunt
- Walk for 5 minutes without checking your phone.
- Find three different types of leaves.
- Sit quietly and observe your surroundings.
- Listen for different natural sounds.
- Find something interesting in nature.
- Take a different route back to your starting point.
The goal is to make the screen the shortest part of the experience.
Generate the mission → Put the phone away → Go outside. 🌱
GitHub Repository
https://github.com/NavyaPachigolla/TouchGrass-AI
Demo
If the project is currently running locally, you can write:
The project currently runs locally using Flask and Ollama. A demo video/screenshot can be added here.
The basic flow is:
Enter preferences → Generate outdoor mission → Start mission → Put the phone away → Go outside
Code
The complete source code is available on GitHub:
GitHub Repository:
https://github.com/NavyaPachigolla/TouchGrass-AI
The project contains:
- Flask backend
- HTML frontend
- CSS styling
- JavaScript functionality
- Ollama local AI integration
- Outdoor mission generation
- Mission countdown timer
How I Built It
I built TouchGrass AI using a simple web stack:
- Python
- Flask
- HTML
- CSS
- JavaScript
- Ollama
- Qwen2.5 3B Instruct
The backend receives the user's preferences and creates a prompt for the local AI model.
The local model then generates a personalized outdoor mission.
The generated mission contains:
- Mission title
- Description
- Duration
- Difficulty
- Outdoor steps
- What to bring
- Safety guidance
- Screen-off challenge
The frontend displays the generated mission in a simple and nature-focused interface.
I also added a mission timer so that after generating the plan, the user can start the activity and reduce interaction with the screen.
The main idea behind the application is that AI should help the user leave the screen, not keep the user on the screen.
Why Does Open Innovation Matter?
Open innovation is important for this project because the purpose of TouchGrass AI is to help people spend less time looking at screens.
The project uses an open-weight AI model locally through Ollama instead of depending on a closed cloud AI API.
This provides several benefits.
1. Local AI
The AI can run locally on the user's computer.
The user's activity preferences can therefore stay on their own machine instead of automatically being sent to a remote AI service.
2. No dependency on a paid AI API
The core AI functionality does not require a paid proprietary AI API.
This makes the project easier to experiment with and run locally.
3. Freedom to experiment
Using an open-weight model allows developers to experiment with different models and approaches.
The model can potentially be replaced or customized as the project evolves.
4. Less vendor lock-in
The application is not designed around one proprietary AI provider.
This gives developers more flexibility when experimenting with local AI.
5. The AI has a specific purpose
The AI is not being used just to create another chatbot.
Its purpose is to generate a real-world activity that encourages the user to stop using the screen and experience the outdoors.
The AI creates the plan.
The real experience happens outside. 🌱
My Agent Session
I used an AI coding agent while developing TouchGrass AI.
[Add your DevRelay agent session here if you have one.]
If you don't have a DevRelay session, this section can be removed because it is optional.
Prize Categories
AI / Open-Source AI
What I Learned
While building this project, I learned how local AI can be integrated into a practical application instead of relying entirely on cloud-based AI APIs.
I also learned that AI doesn't always have to increase screen time.
It can also be used as a tool to help people step away from technology and interact with the real world.
That is the idea behind TouchGrass AI:
Use AI for a few minutes. Then go touch grass. 🌱
Top comments (0)