Marcin.Wojtala

Side project reveal 👑🐌 @royalslugs Since July I've been building Royal Slugs: a browser party game where you and a mate secretly plan your slugs' launches, lock in, then watch both plans collide at once. Phones become controllers with a QR code. Built end to end by directing AI agents, using the practices from my #Agentic #Playbook. So it's also my biggest experiment in how far that approach goes. It's in public alpha today. Expect wobbles, physics chaos and a lot of build-in-public posts from me. #indiedev #gamedev

And here it is 🎬 the official trailer.

5/ The agents wrote a lot of it. I made the calls, reviewed every commit, and rewrote the multiplayer myself when their approach kept failing. Break it and tell me what happens. @royalslugs Play free at royalslugs.com

4/ Royal Tournament: 4 or 8 friends, invite link only. Others can watch through a private link, and the winner gets a champion poster.

3/ Online 1v1 is early alpha. Join an open game or send a friend an invite link.

2/ Same sofa? Your phones become the controllers. Scan the QR code and play.

First crash on my Grok Bot team last night and I'm not even mad... In fact I'm impressed with @bot DR capabilities! My bot's render task crashed the system and took down my other bots mid-task 🫤 13 minutes later it was back on its own. No progress lost. The bots reinstalled, reran 1,535 tests and carried on. The @royalslugs release still opened a day early. Amazing work @ericzakariasson @mattyp @poteto, and the rest of the product engineering team! @bot snapshot recovery saved me.

My definition of done for @royalslugs isn't "it works." It's "it works, the docs are updated, and the agent has challenged what it just built." Every change ends with the agent asking what could be better, in the game and in the documentation. Some of my best decisions came out of an agent pushing back on its own work. The docs grow with the game, not after it. What's in your definition of done when you build with agents? #indiedev #gamedev

Royal Slugs has been in public alpha for a week, and I've mostly watched my mates play it. Would be amazing to get some more feedback from other users, pushing the limits of my game physic engine. Jump in, play on the same screen or your phones as the controllers/ If you've got 15 minutes and a mate (or a second phone), come help me to test @royalslugs and tell me what felt wrong. I read every reply. royalslugs.com/?utm_source=x…

Don't forget, sometimes, the best agent session are the ones, when you decided NOT to build something. I co-authored a full server-side copy of the @royalslugs whole physics game as an ADR. The agent kept me convinced that this is the right direction. After a separate research agent session, I came up with a much simpler solution (inspired by multiplayer online games) Lesson: don't blindly follow your agent, if you are not sure about something, it oftens benefits to park the conversation, clear the context, and reset. Do your agents ever talk you out of things? #agentic #copilot #indiegame

My first phone controllers implementation in @royalslugs streamed the whole game to your phone 15fps, SYNCING but not really as excing as on the large TV or Monitor📱 I got an invulauble feedback from the mate: Then what is the point of showing it on the TV as well? Nobody looks up, they stay locked in on their phone during the action. You should keep the focus on the large screen, use the small one for actions. A true first-to-second screen handover. Thanks to that feedback, we changed it. Aim privately, one Lock in, EYES UP warning. Best thing you've ever deleted?

It's an alpha and it wobbles, but it's real and it's fun. Come and outsmart your friends: royalslugs.com/?utm_source=x… Reach out with the feedback please! The method behind it is in my Agentic Playbook.

5. Engineer the loop, not just the prompt. The agents run the browser tests the change affects, read the evidence, fix, rerun. I run them too on my machine. Then I review the code and push every commit myself. My job is designing that loop and catching it when it's confidently wrong.

4. Direction shift Not everything went according to the plan. I had to rewrite the Multiplayer engine to utilise web sockets; agents' recommendation (even in x high thinking mode) was not producing any tangible results. It's okay to go back to the research, you need to drive the decisions.

3. Small, reviewable steps. The desync: Soccer rules ran on the render loop, so each browser fitted a different number of physics steps per frame. The fix wasn't a rewrite. One rule: anything that affects the game runs on the fixed 120 Hz clock.