This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady
What I Built
I built SlothTidier, a lightweight Python utility that scans messy directories (like a Downloads folder) and automatically sorts files into contextual categories (Documents, Images, Code, Archives, Misc) using a local open-weight model. It saves you from manual digital housekeeping so you can channel your inner sloth.
Demo
Code
You can find the full source code and setup guide on GitHub:
👉 GitHub Repository
How I Built It
I wrote a Python script utilizing requests to communicate with an OpenAI-compatible local inference endpoint. By running an open-weight model (Qwen 2.5 Coder) locally via LM Studio, the script evaluates filenames semantically and executes local file movements with zero external API calls.
Why Does Open Innovation Matter?
Open innovation and local open-weight models made this project possible in three major ways:
- Total Privacy: My personal file and document names never leave my machine or hit third-party servers.
- Zero Cost: Running inference locally means $0 in cloud API subscription fees.
- Complete Offline Autonomy: It runs seamlessly even without an internet connection.
The Lazy Payoff
Digital clutter is one of those annoying micro-tasks everyone puts off. SlothTidier runs locally in the background, instantly turning a chaotic folder into an organized workspace with zero manual clicking.
Prize Categories
- Open-Source AI Track
Top comments (0)