SlothSort AI: Let Open-Source AI Handle Your Digital Clutter

Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady Submission 🦥

Slow down. Sort smarter. Keep your originals safe.

Ever opened your Downloads folder and found yourself staring at hundreds of files, promising you'll organize them someday?

I decided to build something to make that task easier.

Meet SlothSort AI, a local-first file organization assistant built for the Hacktoberfest Open-Source AI Challenge: Slow and Steady.

SlothSort AI helps users declutter their digital workspace without manually sorting every file.

It combines a dark-themed dashboard, automatic file categorization, organization statistics, and AI-generated suggestions in one simple application.

The idea is simple: spend less time managing digital clutter and more time doing what matters.

🧠 The Open-Source AI Behind the Project

SlothSort AI uses Qwen 2.5 3B through Ollama for local inference.

The application is built with Python and Streamlit, with Requests connecting the interface to the local Ollama API.

Instead of depending on a hosted AI API for every request, the application can use a model running on the user's own computer.

For this project, openness isn't just a technical choice. It changes how the tool can be used and improved.

File organization can involve sensitive filenames. Using a local model allows AI suggestions to be generated without sending filenames to a cloud AI API. The current application is designed around local inference.

After installation, local inference does not require a recurring cloud AI API charge. Users still need suitable hardware and storage, and electricity and other costs may apply.

An open-weight model gives developers the opportunity to experiment with model configurations and compatible alternatives. SlothSort AI can evolve as its model integration improves.

The project uses familiar Python tools and a small language model, making it a practical starting point for learning about local AI applications.

Once the model and required dependencies are installed, core AI inference can run without internet access, as long as the local services are available.

That's the benefit of building with open AI: you can run it, learn from it, adapt it, and improve it.

Automation should reduce effort without creating unnecessary risk.

SlothSort AI previews its proposed organization before asking for confirmation. It copies files into category folders rather than moving or deleting the originals, and it avoids overwriting existing destination files by generating alternative filenames.

Its current categorization is extension-based, while the AI provides additional advice based on filenames. It does not claim to understand the actual contents of every file.

I want to explore custom categories, better duplicate detection, copy-history tracking, undo support, and additional local models.

The goal is to keep making digital organization simpler without sacrificing user control.

SlothSort AI began with a familiar problem: file clutter and the never-ending promise to organize it later.

By combining local AI with a safety-first workflow, I wanted to create a tool that rewards doing less manually while giving users more control over their digital workspace.

Open innovation makes that possible by giving developers the freedom to build beyond hosted services and create tools that people can adapt to their needs.

Sometimes the smartest way to get organized is to move at a sloth's pace. 🦥

Project repository: https://github.com/vanshikamishra1909-cmd/SlothSortAI

Built for the Hacktoberfest Open-Source AI Challenge — Slow and Steady.