DEV Community: Ashton Chew

👻 Ghostlet: We Built Something Useless for Hacktoberfest (And Loved It)

Ashton Chew — Wed, 04 Nov 2026 05:51:41 +0000

This is a submission for the MLH x DEV Writing Challenge

What I Built

Not every hackathon project needs to solve a problem. Sometimes, you just want to build something silly, cute and fun!

Meet Ghostlet 👻, a tiny pixel-art desktop companion for Windows, built by three friends for Hacktoberfest 2026 Week 4.

Leave your PC idle, and little ghosts start appearing around your cursor. The longer you're away, the more ghosts, pumpkins, graves and spooky decorations accumulate.

But move your mouse or press a key, and they'll panic, grab their decorations and run away!

Why something useless?

We initially weren't even planning to participate in Week 4.

There were plenty of interesting projects being built, many experimenting with AI and solving practical problems. We wanted to try something different.

AI models were still relatively new to us, so instead of forcing ourselves to build something unfamiliar, we decided to use technologies we already knew and create something we would genuinely enjoy.

Our idea? Something deliberately useless, but fun!

We wanted people to have a little companion running in the background while coding, especially during Halloween.

🎨 Bringing Ghostlet to life

We split the work based on what we enjoyed most. My teammates focused on the creative side, while I handled most of the programming.

Taking inspiration from perler bead art, we manually created pixel-art ghosts and decorations, giving each character its own personality. There are even several ghost variations, including a rare witch ghost! ✨

Technically, Ghostlet is built with Python, PySide6 and the Windows API, which detects system idle time and allows the ghosts to react to user activity.

PySide6 was new to me, so understanding the framework and getting everything working was one of my biggest challenges. I used AI assistance through Claude on the web to help me learn unfamiliar concepts and work through implementation details.

It was a fun opportunity to explore desktop application development using Python!

Demo

Want some little ghosts on your desktop?

GitHub: https://github.com/ashtonckj/ghostlet
Download: Windows executable on GitHub Releases

No Python installation required!

The project is open source under the MIT License. Feel free to explore, customise the ghosts or create your own decorations!

Partner Technologies

Ghostlet uses Python, PySide6 and the Windows API. We didn't integrate any MLH partner technologies into the application itself. Instead, we focused on building something with tools we were comfortable with while using AI assistance as a learning aid.

Hackathon Experience

We participated online as a team of three:

Honestly, I'm glad I convinced everyone to give Week 4 a try!

The teamwork was amazing. Everyone contributed in their own way, and we had fun bringing our little Halloween idea to life.

We didn't build something that would change the world, but we made something that could brighten someone's day while coding. And I think that's pretty cool.

Halloween might be over, but who knows? Maybe Ghostlet will return with some Christmas friends! 🎄

Happy coding, and don't leave your computer unattended for too long. You never know who's watching. 👻

👻 Ghostlet: While you’re away, the ghosts will play.

Ashton Chew — Fri, 30 Oct 2026 15:38:40 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine

What We Built

Meet Ghostlet 👻, a tiny pixel-art desktop companion for Windows!

The idea is simple: leave your PC idle, and little ghosts start appearing around your cursor. The longer you're away, the more ghosts, pumpkins, graves and spooky decorations accumulate.

Move your mouse or press a key, and the ghosts panic, grab their decorations and run away!

We wanted to turn an ordinary idle desktop into a little Halloween scene.

👻 Meet the Ghosts!

Classic	Cool	Crying	Hunting	Loving
Magic	Peeping	Shivering	Sleeping	Rare Witch! ✨

🎃 Meet the Decorations!

Pumpkin	Happy Pumpkin	Red Pumpkin	Medium Pumpkin	Jack-o'-Lantern	Spider
Skeleton	Cross Grave	Tall Grave	Wide Grave	Short Grave

Demo

Download Ghostlet and give it a try!
GitHub Releases (Windows executable, no Python installation required)

Code

The project is open source under the MIT License.

GitHub Repository – Ghostlet

ashtonckj / ghostlet

While you’re away, the ghosts will play. BOO!

Ghostlet

Step away from your PC. Tiny ghosts move in. 👻

Leave your mouse alone for a minute and pixel ghosts float up around your cursor, while pumpkins, graves and spooky friends pile up along the bottom of your screen. Come back and they panic, grab the decorations and run away! 🎃

👻 Meet the ghosts

Each ghost picks a random look.


classic	cool	crying

hunting	loving	magic

peeping	shivering	sleeping

✨ The rare one! 10× less likely to show up than any other ghost. ✨

🎃 Spooky props


pumpkin	happy pumpkin	red pumpkin	medium pumpkin	jack-o'-lantern

spider	skeleton	cross grave	tall grave	wide grave	short grave

The longer you're away, the more pile up. When you're back, the ghosts carry them off.

💾 Get it

Grab Ghostlet.exe from Releases and double-click it. No Python needed.

Ghostlet lives in your tray (bottom-right). Right-click the ghost for Settings.

Run from…

View on GitHub

How I Built It

Ghostlet is built with Python and PySide6, using the Windows API to detect system idle time.

Why Does Open Innovation Matter?

Open-source development makes it possible for people to inspect, modify and build upon existing work.

Ghostlet is released under MIT, so anyone can customise the ghosts, create new pixel-art decorations or even add their own behaviours.

Sometimes, even a small silly project can become something others want to experiment with!

The Spirit of It

We think the funniest part is coming back to your PC and finding a whole bunch of ghosts having a little party on your desktop.

And then watching them desperately run away when they realise you're back. 🎃

👥 Built by three humans

@ashtonckj
@fionakhl
@moltenice3

Happy Halloween! 👻🎃

🦥 We built a pixel-art office where five cats do your data analysis while a sloth naps 🐱

Ashton Chew — Sun, 25 Oct 2026 06:44:33 +0000

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

Every dataset starts the same way: df.head(), count the missing values, find the duplicates, draw a histogram, repeat. So we handed that work to a team of cats.

Purrrfessorrr Slothhh is a local-first data analysis assistant. Upload a CSV, TSV, or Excel file, and a very sleepy sloth professor delegates the work to five specialist cats. Your data never leaves your machine.

What I Built

🕵️ Detective Cat finds outliers, duplicates, and type problems
🎨 Artist Cat draws the charts
📝 Scribe Cat explains the findings in plain English
💻 Coder Cat writes Python and R code for every analysis
🧹 Janitor Cat suggests how to clean the data

Ask follow-up questions in chat, then export everything as a ZIP or a Jupyter notebook.

Demo

Try it in two minutes. No account and no API key needed:

git clone https://github.com/ashtonckj/Purrrfessorrr-Slothhh.git
cd Purrrfessorrr-Slothhh
.\run.bat        # macOS / Linux: ./run.sh

Code

GitHub: https://github.com/ashtonckj/Purrrfessorrr-Slothhh

How I Built It

Python + FastAPI, vanilla JavaScript, and optional Gemma 3 running locally through Ollama.

The key decision was to keep calculation and conversation separate:

Python computes every number. The analysis is a plain, documented pipeline, so the same file always gives the same report.
Gemma only explains. It sees a summary of the computed results, not your raw rows. A guard then removes any sentence containing a number that isn't in the real data.
No model? Still works. Without Ollama, the full analysis and rule-based chat keep running.

Why Does Open Innovation Matter?

A dataset is often the most sensitive thing on your laptop. Open-weight models let us build AI-assisted analysis that needs no hosted API and sends nothing to a third party. We enforced it in code: the app refuses to talk to any model that isn't on localhost.

The Lazy Payoff

The repetitive first pass of exploratory analysis (profiling, quality checks, starter charts, boilerplate code) happens in one guided workflow, so you spend your time on what the data means. We haven't benchmarked the time saved, so we won't claim a number.

Prize Categories

Best Use of Gemma (featured): Gemma 3 runs locally via Ollama to write the morning digest, grounded strictly on the model's own output, with a deterministic offline fallback so it never hard-fails.

👥 Built by three humans (and five fictional cats)

Ashton — Github: @ashtonckj · Dev: @ashtonckj
Fiona — Github: @fionakhl · Dev: @fionakhl
Jocelyn — Github: @JocelynKoh3 · Dev: @moltenice3

🎨 Third-party Credits

While the application itself was developed by us, some of the illustrations and icons were sourced from artists and creative communities across the web. 🐈✨

A huge thank you to the creators whose artwork helped give this project its playful personality! 💛

We've documented the sources of these assets in my GitHub repository under ASSETS.md, where you can explore the original artwork and its attribution details.

All credit for the third-party illustrations belongs to their respective creators.

The sloth is slow. The workflow doesn't have to be. 🦥💤

Bringing Better Classic ASP Support to VS Code in 2026

Ashton Chew — Mon, 12 Oct 2026 12:40:48 +0000

Bringing Better Classic ASP Support to VS Code in 2026

Classic ASP may be old, but legacy applications still need to be maintained. When working with these applications, the editor should help with more than syntax highlighting.

I started building Classic ASP Language Support because I wanted a more complete development experience in VS Code. What began as an extension for a legacy language has grown into a project covering formatting, IntelliSense, navigation, diagnostics and several other language features.

In this article, I'll explain the problems I wanted to solve, some of the challenges of supporting VBScript alongside HTML and JavaScript, and what I've learned from building and maintaining the extension.

Why Classic ASP is difficult to support

A single .asp file can contain several different languages. The editor has to understand where server-side VBScript begins and ends, how it interacts with HTML, and how embedded JavaScript, CSS and SQL should be handled.

Simply highlighting keywords is not enough when you also want useful completion, navigation and diagnostics.

Building a better development experience

I'll walk through some of the features I've implemented, including:

Formatting mixed-language ASP files.
IntelliSense and context-aware completions.
Navigating definitions across #include files.
Renaming symbols and finding references.
Detecting common VBScript and markup mistakes.

The challenges behind the features

The interesting part isn't just adding commands to an editor. It's making different language features cooperate without interfering with each other, while keeping the extension responsive on large files.

I'll share examples of the edge cases I've encountered and the lessons I've learned from improving the implementation.

What's next

Classic ASP is a legacy technology, but that doesn't mean maintaining it should be unnecessarily difficult.

I'm continuing to improve the extension and would welcome feedback from developers who work with real Classic ASP applications.

Try it: Classic ASP Language Support on the VS Code Marketplace

Source code: GitHub repository