Week 3 hacktoberfest challenge post

Nayeemul Islam — Fri, 23 Oct 2026 17:08:07 +0000

Downloads Sloth 🦥: Slow, Steady, and Lazy File Sorting with Open-Weight AI

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

🏆 The Challenge: Slow and Steady

The prompt for Week 3 was to build something with open-source AI at its core that rewards laziness .

My Downloads folder is a graveyard of forgotten installers, random screenshots, and invoices named document(1).pdf. Manual sorting is tedious, and giving a closed-source AI access to my personal files felt risky.

Enter Downloads Sloth: a lazy, careful, open-weight file tidier. It embraces the "Slow and Steady" philosophy by waiting for downloads to finish, respecting API rate limits, and previewing moves before executing them. It rewards laziness by running silently in the background, sorting your mess, and writing a daily digest—so you never have to look at your Downloads folder again.

🎥 Demo Video

Here is a quick code-only demo showing the Sloth in action: checking API limits, running a safe dry-run, sorting files for real, and using the local dashboard.

🦥 What I Built

Downloads Sloth is a Python automation that watches your Downloads folder, reads new files, gives them sensible names, sorts them into categories, and logs every decision to SQLite.

How it rewards laziness:

Set it and forget it: Run it via Windows Task Scheduler on logon. It watches continuously.
Zero API waste: Obvious files (.zip, .exe, .mp4) are sorted locally via extension rules.
Privacy-first: Files matching patterns like passport, bank, tax, or resume are never sent to the AI. They are sorted locally by name.
Safe by default: It defaults to a dry run. It predicts where files will go without moving them. When you're ready, you sort for real.
Fully undoable: Made a mistake? python sloth.py undo --last 5 puts everything back exactly where it was.

🛠️ How I Built It

The project is intentionally lightweight and relies on open-source tools:

Core: Python 3.11+ (using standard library sqlite3, http.server, and tomllib).
The Brain: Gemma 4 31B (an open-weight model under Apache 2.0) via OpenRouter.
Vision/Docs: Pillow for downscaling images, pypdf and python-docx for extracting text excerpts.
UI: A zero-dependency, local-only web dashboard built with Python's standard library http.server and vanilla JS. It shows the "pile" of waiting files, the "shelves" of sorted files, and a live console.

🌍 Why Open Innovation Matters for this Project

The challenge asked us to explain why open-source AI matters for what we built . For a file sorter, open innovation is a strict requirement for trust.

Privacy and Transparency: You are asking an AI to read your financial statements, medical prescriptions, and personal resumes. With closed-source models, you are sending that text into a black box. By using an open-weight model like Gemma 4, the community can audit how the model behaves.
No Vendor Lock-in: Because the Sloth uses an OpenAI-compatible API endpoint, you aren't tied to one provider. If you want maximum privacy, you can change one line in config.toml to point to a local Ollama server running on your own hardware. Open-source AI gives users the freedom to own their inference.
Community Extensibility: Hacktoberfest is about community. Because the sorting logic, privacy rules, and categories are exposed in a simple config.toml and standard Python code, anyone can fork the repo, add new extension rules, or translate the dashboard without fighting a proprietary API.

💻 Try It Yourself (The Lazy Way)

Repo:

Krak3n909 / HacktoberFest-Week3

Downloads Sloth

A lazy file tidier powered by an open-weight model (Gemma 4 31B via OpenRouter) Drop files in Downloads. The sloth reads them, renames them, sorts them, and leaves a short digest. It never deletes anything, and every move can be undone.

Setup (Windows, PowerShell)

pip install -r requirements.txt
setx OPENROUTER_API_KEY "sk-or-your-key"     # then close and reopen PowerShell

Edit config.toml if your folders differ (defaults: ~/Downloads -> ~/Downloads/_Sorted).

Use

python sloth.py check             # test key + show your real OpenRouter limits
python sloth.py once --limit 10   # dry run on 10 files (default mode: moves nothing)
python sloth.py once --live       # actually move files
python sloth.py run --live        # watch continuously
python sloth.py undo --last 5     # put the last 5 moves back
python sloth.py digest            # write today's summary
python sloth.py ui                # local web dashboard (needs nothing extra)

Dashboard

python sloth.py ui opens http://127.0.0.1:8765. It…

View on GitHub

# 1. Install dependencies
pip install -r requirements.txt

# 2. Set your OpenRouter API key (free tier works fine!)
setx OPENROUTER_API_KEY "sk-or-your-key"

# 3. Check your limits and test the model
python sloth.py check

# 4. Preview what the sloth wants to do (moves nothing)
python sloth.py once --limit 10

# 5. Sort for real
python sloth.py once --live

# 6. Open the local web dashboard
python sloth.py ui

🌱 Hacktoberfest Contribution Ideas

Want to earn your Hacktoberfest PRs? The Sloth needs your help:

Add more local extension rules or privacy patterns.
Improve the vanilla JS dashboard styling.
Add support for local audio transcription for .mp3 sorting.
Write unit tests for the privacy regex engine.

Check out the code

Krak3n909 / HacktoberFest-Week3

Downloads Sloth

Setup (Windows, PowerShell)

pip install -r requirements.txt
setx OPENROUTER_API_KEY "sk-or-your-key"     # then close and reopen PowerShell

Edit config.toml if your folders differ (defaults: ~/Downloads -> ~/Downloads/_Sorted).

Use

python sloth.py check             # test key + show your real OpenRouter limits
python sloth.py once --limit 10   # dry run on 10 files (default mode: moves nothing)
python sloth.py once --live       # actually move files
python sloth.py run --live        # watch continuously
python sloth.py undo --last 5     # put the last 5 moves back
python sloth.py digest            # write today's summary
python sloth.py ui                # local web dashboard (needs nothing extra)

Dashboard

python sloth.py ui opens http://127.0.0.1:8765. It…

View on GitHub

, drop a star, and let the Sloth tidy up your digital life! 🦥

Week 2 hacktoberfest challenge post

Nayeemul Islam — Mon, 12 Oct 2026 20:46:20 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built

ReplyPal writes replies to customer reviews. You paste a review, say what your business is, pick a tone (warm, professional, or playful), and get a short reply back. It thanks the customer, answers their specific points, and if the review is negative, it apologizes and offers a fix.

It's for small business owners: Etsy sellers, cafe owners, freelancers. They know replying to reviews helps, but every reply takes time, and angry reviews are hard to answer calmly. A few dollars a month is an easy decision for someone who gets dozens of reviews.

Demo

Try it live: hacktoberfest-week-3.streamlit.app

It's on a free tier, so it may take a few seconds to wake up, and it may hit rate limits if a lot of people use it at once.

Code

Krak3n909 / HacktoberFest-Week-3

The whole app is one file, app.py, about 35 lines.

How I Built It

Model: openai/gpt-oss-120b, an open-weight model, served through Groq's API.
Interface: Streamlit, with a text input for the business, a tone dropdown, and a text area for the review.
Logic: One prompt that includes the business description, the tone, a length limit, and the review text. It makes one API call and shows the result.
Deployment: Streamlit Community Cloud, with the API key stored in the app's secrets rather than in the repo.

I kept it deliberately small. The whole product is one prompt and one call, so the model choice is the only real dependency. That made it easy to build in an afternoon and easy to reason about.

Why Does Open Innovation Matter?

Because the model is open-weight, I'm not tied to one vendor's pricing or terms. The model name is a single line of code. If Groq changes its prices or retires a model, I change that line, or I run the same weights somewhere else. For a tool that would sell for a few dollars a month, that matters: each reply costs a fraction of a cent, and no one else controls that cost.

Open weights also leave the door open for things a closed API wouldn't allow, like running it locally so a shop's customer data never leaves their machine, or fine-tuning it on one owner's past replies so it sounds like them.

Prize Categories

I'm not entering any partner categories.

DEV Community: Nayeemul Islam

Week 3 hacktoberfest challenge post

Downloads Sloth 🦥: Slow, Steady, and Lazy File Sorting with Open-Weight AI

🏆 The Challenge: Slow and Steady

🎥 Demo Video

🦥 What I Built

🛠️ How I Built It

🌍 Why Open Innovation Matters for this Project

💻 Try It Yourself (The Lazy Way)

Krak3n909 / HacktoberFest-Week3

Downloads Sloth

Setup (Windows, PowerShell)

Use

Dashboard

🌱 Hacktoberfest Contribution Ideas

Krak3n909 / HacktoberFest-Week3

Downloads Sloth

Setup (Windows, PowerShell)

Use

Dashboard

Week 2 hacktoberfest challenge post

What I Built

Demo

Code

Krak3n909 / HacktoberFest-Week-3

How I Built It

Why Does Open Innovation Matter?

Prize Categories