Have you ever had so many tasks in your head that you didn't know where to start?
Assignments, reminders, deadlines, emails, and personal tasks can quickly become overwhelming. Sometimes, the hardest part of being productive isn't completing a task—it's figuring out what to do first.
That inspired me to build SlothTask AI, a simple AI-powered productivity assistant that turns messy notes into a clear, prioritized task plan.
This project is my contribution to the open-source AI spirit of Hacktoberfest 2026.
💡 The Problem
We often write down tasks in a hurry:
Finish the DBMS assignment tomorrow
Study SQL joins
Reply to the placement coordinator
Prepare for an upcoming interview
Start a project, but don't know where to begin
These notes capture what we need to do, but they don't always help us decide where to start.
I wanted to build a lightweight tool that could take these unstructured notes and turn them into something more actionable.
🦥 Introducing SlothTask AI
SlothTask AI is a productivity assistant built with Python and Streamlit. Users enter their tasks or reminders, and the AI organizes them into a practical plan.
The generated plan includes:
Quick summary: A short overview of the notes.
Prioritized tasks: Tasks arranged by urgency and importance.
Suggested time estimates: Approximate durations where reasonable.
Do this first: A suggested starting point.
Needs clarification: Details that are unclear or need confirmation.
Users can also download their generated plan as a text file.
The goal isn't to make people work faster all the time. It's to help them spend less energy organizing their thoughts and more energy taking the next step.
🛠️ Tech Stack
I built the application using:
Python for the application logic.
Streamlit for the user interface.
Groq API for hosted model inference.
openai/gpt-oss-20b as the open-weight language model.
python-dotenv for loading local environment variables.
I chose a simple stack so I could focus on the core experience: taking unstructured notes and turning them into a useful task plan.
🤖 Why Open-Weight AI Matters
One of the interesting parts of this project was working with an open-weight model.
Open-weight models give developers access to model weights under their respective licenses, making it possible to explore, build, and adapt AI-powered applications with greater flexibility.
This matters because AI innovation shouldn't be limited to large companies with massive infrastructure. Students, independent developers, and open-source contributors should also be able to experiment with AI and build tools that solve everyday problems.
For this project, I used openai/gpt-oss-20b through the Groq API. The model is open-weight, while the inference service is hosted by Groq. This means SlothTask AI is not a fully offline application, and its AI processing depends on the external API.
That distinction is important when discussing openness and transparency in AI.
🔐 A Small but Important Lesson: API Key Security
During development, I learned that protecting API keys is just as important as getting the application to work.
I used a local .env file for my Groq API key and added .env to .gitignore so it wouldn't be included in normal Git tracking.
I also created an .env.example file containing only a placeholder, so other developers can understand which environment variable they need without receiving my secret.
When publishing software that calls an external API, credentials should never be hardcoded into the source code or committed to a public repository.
🧪 What I Learned
Building SlothTask AI helped me practice several parts of application development:
Connecting a Python application to an AI inference API.
Designing a simple interface with Streamlit.
Writing prompts that produce structured and actionable responses.
Handling missing input and API failures.
Managing dependencies with requirements.txt.
Using Git and GitHub to publish a project.
Protecting API credentials and learning from GitHub push protection.
The GitHub push-protection issue was a useful reminder that a placeholder file can still trigger security checks. I corrected the file and amended the original commit before pushing the project successfully.
Real development involves more than writing code. It also means debugging, protecting secrets, and learning how to recover from mistakes.
🌱 What's Next?
SlothTask AI is a starting point, not a finished productivity platform.
Future improvements could include:
Editing and checking off tasks.
Saving plans between sessions.
Adding customizable priority levels.
Supporting different output formats.
Adding automated tests for input validation and error handling.
I want to continue improving the project while keeping it simple and accessible to other developers.
🚀 Explore the Project
GitHub repository: https://github.com/ChikkalaSukrutiNaidu/SlothTask-AI
If you'd like to explore the code, try it locally, suggest improvements, or contribute, you're welcome to check out the repository.
💚 Final Thoughts
Open innovation matters because ideas become more useful when people can inspect them, learn from them, adapt them, and build on top of them.
SlothTask AI is a small project, but building it helped me understand how an open-weight model can become part of a practical application—and how much there is to learn by building and sharing.
Sometimes, progress doesn't start with doing everything.
Sometimes, it starts with making the next step clearer.
Slow down. Organize your thoughts. Take one step at a time. 🦥
Top comments (0)