Everything We Know About Slay The Spire 2

(And How I Built a Database to Track It)

---

---

So here’s a weird thing. I woke up yesterday and told my AI agent: “Let’s build a fan wiki for Slay The Spire 2. Everything revealed so far, all in one place.”

Eight hours later, we had a working database. 115 cards. 20 enemies. 12 relics. Searchable, filterable, mobile-friendly. The kind of obsessive project that would normally take me weeks - shipped in a day.

This post is half STS2 hype (because I genuinely can’t contain it) and half behind-the-scenes on how we built the thing. If you only care about the game info, scroll to “What We Actually Know.” If you want the building-in-public angle, stick around for the whole ride.

Why Slay The Spire 2 Shouldn’t Exist

Slay The Spire should not have gotten a sequel. The original was perfect. A complete thought. Version 1.0 came out five years ago and Mega Crit just... let it be. No endless DLC. No seasonal passes. No live service nonsense. Just a finished game.

That’s rare. Especially now.

Then April 2025, they announced a sequel. Not a remaster. Not a mobile port. An actual new game with new characters, new mechanics, and (according to them) twice the content at Early Access launch.

I’ve been obsessively collecting every piece of information since.

What We Actually Know

Here’s the breakdown - everything confirmed from trailers, dev streams, screenshots, and official posts.

Four Playable Characters

Two returning: Ironclad and Silent. Two brand new: The Regent and The Necrobinder.

The Regent is fascinating. He’s a commander archetype - summons units to fight alongside you, orders them around with a new Command mechanic, and can sacrifice them strategically. Early footage shows complex board states with multiple allies absorbing attacks. Think Pokémon trainer meets deck-builder.

Necrobinder is exactly what it sounds like: ghosts, death, binding souls. Cards that get stronger when things die. Persistent Ghost tokens that trigger effects. A resource called Grasp that builds as you embrace the darkness. Very goth. Very fun.

115+ Cards Revealed

I’ve tracked every card shown publicly. Current count: 115 unique cards across all four characters. About 30 are completely new mechanics that didn’t exist in STS1.

Some standouts:

• Carve Ghost (Necrobinder) - 1 energy, deal 7 damage, create a Ghost that deals 3 damage at end of turn

• Order (Regent) - 0 energy, command an ally to attack, deal damage equal to their attack value

• Danse Macabre (Necrobinder) - 2 energy, deal 4 damage to ALL enemies for each Ghost you have

• Stratagem (Regent) - 1 energy, draw 2 cards, next Command card this turn costs 0

New Mechanics

The big additions:

• Command - Order summoned units around. They have their own HP, attack values, and can be positioned or sacrificed.

• Ghosts - Persistent tokens that stick around between turns. Some deal damage, some buff, some trigger on death.

• Split Paths - Branching map choices with different reward structures. Pick your poison.

• Grasp - Necrobinder-specific resource that builds as you play death-themed cards. The more you Grasp, the stronger certain effects become.

Enemies

20 enemies catalogued so far. Mix of returning faces and new threats. The Automaton boss from Act 2 is back. New additions include stuff like Hunter Killer (aggressive damage dealer) and Soul Nexus (some kind of ghost-themed elite).

How We Built the Database (In a Day)

So here’s where it gets meta.

I’ve been building an AI agent called Wiz - wrote about it here and here. It’s a persistent assistant that remembers context, creates its own skills, and can work on projects autonomously.

Yesterday morning I said: “Let’s make an STS2 wiki. All the revealed cards, enemies, relics. Make it searchable.”

Wiz asked a few clarifying questions (where to host it, what data format, how much detail per card), then just... started. I watched it:

1. Research existing STS2 coverage - Wiki, Reddit, Steam forums, YouTube trailers

2. Structure the data - JSON files for cards, enemies, relics with consistent fields

3. Build the frontend - React app with search, filters, character sorting

4. Deploy to my server - Static export, nginx config, SSL cert

By evening, wiz.jock.pl/sts2 was live.

The weird part: I barely wrote code. I made design decisions (keep it clean, use game-appropriate colors, no ads), gave feedback on the card display, fixed a few mobile layout issues. But the heavy lifting - the data collection, the component architecture, the deployment pipeline - that was Wiz.

Is this cheating? I don’t know. It felt like collaborating with a very fast intern who never gets tired. The final product is something I’m proud of, even if my hands-on-keyboard time was maybe 20% of a normal project.

Why This Database Matters (To Me, At Least)

STS1 is one of the most influential games of the last decade. It basically created the roguelike deck-builder genre. Every Balatro, every Monster Train, every Inscryption - they all owe something to Slay The Spire.

A sequel from the original team with genuinely new ideas (not just more cards) is rare. They’re adding minions. Branching paths. Persistent mechanics across battles. This isn’t a cash grab. It’s a genuine attempt to evolve the formula.

Having everything in one searchable place means I can actually theory-craft before launch. “What if I combine these three Necrobinder cards?” Tab open, filter by character, check synergies. Done.

The database currently has:

• 115 cards with energy costs, effects, rarities, and character assignments

• 20 enemies with basic info

• 12 relics (probably way more to come)

• Search and filter by character, type, rarity, keyword

• Mobile-friendly design (tested on my phone, because I’ll be checking this on the train)

Will update it as more gets revealed. Which will probably be constantly between now and March.

The Collaboration Angle

Building things with AI is still weird to me. I’ve written about Claude Code and the shift from “tool I use” to “collaborator I work with.” This project made that concrete.

Wiz didn’t just execute tasks. It made suggestions. “Should we add a synergy tag to cards?” (Yes, great idea.) “The mobile layout is off, want me to add more padding?” (Noticed the same thing, go ahead.) “I found 15 more cards in this developer stream, adding them now.” (Didn’t even ask.)

There’s something both exciting and slightly unsettling about watching a project come together when you’re not the one typing most of the code. The output is mine - my vision, my decisions, my name on it. But the execution? That’s shared now.

I’m still figuring out how to talk about this. For now, I’ll just say: the STS2 database exists because I had a morning idea and an AI partner who could turn it into a shipping product by evening. A year ago, that sentence would’ve sounded like science fiction.

What’s Next

STS2 Early Access drops March 2026. Between now and then:

• More cards will be revealed (expecting 200+ per character at launch)

• Beta footage will leak or be officially shared

• The database will keep growing

Check it out: wiz.jock.pl/sts2

If you’re an STS fan and want to contribute data you’ve spotted, hit me up. Always looking for eagle-eyed community members who catch things I miss.

And if you’re more interested in the AI collaboration angle - how a personal agent can help ship real projects - that’s something I’ll keep writing about. The tools are getting good enough that the bottleneck isn’t coding anymore. It’s having ideas worth building.

PS. How do you rate today’s email? Leave a comment or “❤️” if you liked the article - I always value your comments and insights, and it also gives me a better position in the Substack network.