Dhruv Mahajan, CM Shaik Abdul Rahim Batcha, Monisha P
We have constructed two ways of breaking free from the algorithm.First being a passive way that gives us notifications and helps us work to our goal. The second method is more active and changes the algorithm itself via interactions.
A focus app that helps break scrolling habits using AI-powered notifications.
- Smart notification generation using LLMs (Gemini or Local Llama via Ollama)
- Context-aware task selection
- Progress tracking and performance analysis
- Fallback templates when LLM is unavailable
- Adaptive feedback system
- Clone the repository:
git clone https://github.com/shcbswvc/port_22.git
cd port_22- Install dependencies:
pip install -e .- Set up environment variables:
cp .env.example .env
# Edit .env with your configuration- For local LLM support:
- Install Ollama from ollama.ai
- Pull the Llama model:
ollama pull llama3.2:3b
The app can be configured to use different LLM providers:
- Gemini (Cloud-based)
- Local (Ollama with Llama3.2)
- None (Uses template-based notifications)
Run the demo:
python -m src.demosrc/core/: Core application logicsrc/notifications/: Notification generation and templatessrc/database/: Data persistencesrc/models/: Data models
The main concept is that the user first gives input of the topics they like and so on, based on this the LLM generates some users that they should follow, the users followed help the algorithm change as the algorithm depends on who and how we interact with other accounts, due to Instagram being really good at detecting automated accounts, I will advice not to use this script.
MIT