Ashen Crown is a roguelite autobattler where you assemble a fighter from dozens of weapons, classes, and skills before watching them battle through 20 procedural encounters to face a final boss. With hundreds of viable build combinations across specializations and skill synergies, the game rewards both focused mastery of single mechanics and chaotic experimentation. It suits players who enjoy discovering broken interactions and watching their custom builds succeed or fail in real-time combat.
About this game
A single-player roguelite autobattler built almost entirely with AI. Pick a fighter, climb 20 battles to a final boss, and win in the build you assemble: 12 weapons, 24 classes, 53 subspecs, 247 skills. Go deep on one path or sprawl across many and break the game open.
What players are saying
▼ Not recommended1 hrs
An interesting AI experiment, but not a game I can recommend in its current state.
Ashen Crown is unusual because the project openly presents itself as an experiment in AI-assisted game development. The developer has also made the source code available and appears transparent about the game’s limitations and future ambitions. I genuinely respect that approach, and as a technical or creative experiment, the project is worth following.
However, Steam reviews ultimately need to answer a different question: is the product currently worth buying and playing?
After spending around 90 minutes with Ashen Crown, my answer is unfortunately no.
The game has a potentially interesting core idea, but the current experience feels more like an early prototype than a finished commercial release. The presentation is extremely basic, the combat lacks satisfying feedback, and the limited strategic depth becomes noticeable very quickly. Many encounters feel repetitive, while the interface and overall pacing make the experience harder to enjoy than it should be.
The AI-generated nature of the project is an interesting starting point, but it cannot replace polish, balance, engaging progression and meaningful gameplay decisions. At the moment, Ashen Crown feels more valuable as a public development experiment than as a complete game for players.
This is not meant as an attack on the developer or the idea behind the project. I appreciate the transparency, the accessible source code and the willingness to explore a new development process. But based on the version currently being sold, I cannot recommend purchasing it yet.
My current rating: 3/10.
I would be happy to revisit the game after substantial updates, especially if future development improves the combat feedback, strategic variety, interface and overall presentation.
Hi, and nice to see you found this page.Ashen Crown is an experiment we've brought to an end. The full results are written up at https://quahjit.com/ashencrownproject/, and the source code is there too (once the game launches). If you just want to poke around and see how it's built, go ahead, we're more than happy for you to.On the whole it helped us explore, on the creative front, how far we could take an idea on our own before bringing in actual artists and professionals to flesh it out. Which is to say: the game isn't really done yet.So if you do support it, here's our pledge. If Ashen Crown reaches 1,000 sales, the proceeds go toward hiring local talent to create proper creative assets for the game, released as a free update to everyone who backed it early.That's the experiment's next chapter, and you'd be part of it.
Posts come from Steam's official announcements feed.