I also compiled the full timeline of development, which you can browse here: https://wordminos.com/making-of/timeline.html

which bit of polish changed how people reacted the most?

I think it's still to be seen in the long run but so far two main ones come to mind

- Adding a mascot(s) helped immensely for creating an identity / making it not another puzzler

- Tuning the puzzle generation constraints: how frequent 2-,3-,4-,5- size pieces are, how many shared colors, how colored markers are laid out relative to one another in the board, all helped tune the puzzles from not being outlandishly difficult or mundanely boring.

Great work, was fun to play!

Thanks! It took a lot of tuning!

Wowza... I am so bad at this. lol

One thing I would say is that when there is a "done" or a "next" button, it took me a sec to see it as a button because the actual button color was too close to white on my monitor. So I thought it was a click anywhere kinda "done".

Thanks for the feedback! Maybe I should add another touch animation like the one that teaches you about the levels menu chevron. Or just add some light glows every so often

This barely feels like a Web game. Well done, especially in 12 days. What was your workflow like? How much use could you get out of LLMs past the basic mechanics (if at all)?

Thanks!

Webgames have really evolved far and away past what you may imagine them being able to support

This was actually my first entirely agent-coded, no code read game i made (in 20yrs of coding)

The timeline is here: https://wordminos.com/making-of/timeline.html

My main contributions were the polish, design, character, packaging ideas

Cool stuff it reminds me of Tiled Words a bit.

I'd really be interested in seeing more about your "no code read" development process. Not worried about things like incorrect local storage writes? Or patterns that will be difficult to modify later when making changes?

Edit: Also, tech stack? Did you choose the tech stack? Or just let AI choose?

With this project really just let the agent do its thing. I had some very minor scaffolding to ask that it write english specs out but otherwise i havent really been on top of that.

I chose the “tech stack” but it wasnt much in the way of direction: plain js and webgl, cause i planned to have that be wrapped for ios and android after

Actually a very fun game! The concept is brilliant if I may say so. I can see this being a very successful mobile game.