# SlothSort: A Slow and Steady AI Agent That Organizes Your Downloads Folder 🦥

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

Meet SlothSort, a local AI-powered Downloads folder organizer that takes the boring work of managing files off your hands.

We all know how quickly the Downloads folder turns into a mess of PDFs, documents, images, and randomly named files. Finding something important becomes a task of its own.

SlothSort makes this easier by acting as a local file-organizing agent. It can:

• Analyze files in your Downloads folder.
• Organize files into appropriate folders.
• Rename files with more meaningful names.
• Extract text from supported text and PDF files to understand their contents.
• Identify images using their filenames and extensions.
• Maintain activity logs so users can track what the agent does.
• Support undo functionality to reverse organizational changes.

The most important rule: SlothSort never deletes your files.

It works quietly in the background, helping you maintain an organized workspace without manually sorting everything.

Code

GitHub Repository: https://github.com/yaansa/slothsort

SlothSort is open source, and contributions are welcome!

How I Built It

I built SlothSort around Gemma 3, an open-weight AI model, running locally through Ollama.

The project combines local AI inference with file-system operations to create a practical desktop automation agent.

Here's the basic workflow:

1. The agent scans the Downloads directory.
2. It examines filenames, extensions, and supported file contents.
3. The local AI model helps interpret file context and determine meaningful organization.
4. The agent renames and moves files into appropriate folders.
5. Every operation is recorded in an activity log.
6. Users can undo changes when necessary.

The technology behind the project includes:

• Gemma 3: Local AI reasoning.
• Ollama: Running the open-weight model locally.
• Python: File processing and automation.
• PDF and text extraction: Understanding supported document contents.
• File-system operations: Organizing and renaming files.

The focus was not on building another chatbot, but on creating an AI agent that performs a useful real-world task.

Why Does Open Innovation Matter?

Open innovation was essential to building SlothSort.

Using an open-weight model like Gemma 3 through Ollama allowed me to experiment with AI locally instead of depending on a closed, paid API.

This made several things possible:

• Privacy: Files can be processed locally instead of being uploaded to an external AI service.
• Accessibility: No paid API subscription or API key is required.
• Independence: The project can run without relying on a cloud AI provider.
• Experimentation: I can customize the workflow and experiment with local models.
• Transparency: Developers can inspect, modify, and improve the implementation.

For a tool that interacts with personal files, local processing is especially valuable.

Open-source AI makes it possible for developers to build useful, accessible tools while retaining greater control over their data and workflows.

The Lazy Payoff 🦥

The goal of SlothSort is simple: spend less time organizing files and more time actually using them.

Instead of manually opening, renaming, and moving downloaded files, users can let the agent handle the repetitive organization.

It's a small automation, but one that can make everyday computer usage a little less frustrating.