Anyone else building a 2D RPG on SRD 5e rules? Curious how you handled the rules engine.
I'm building a 2D turn-based RPG that uses the SRD 5.2 ruleset (the Creative Commons core of D&D 5e) and ran into a problem I suspect I'm not alone in: implementing the rules correctly is genuinely hard, and the further you get into it, the harder it gets.
Not the basics — rolling dice, tracking HP, applying conditions. Those are fine. I mean the interaction surface. Reactions firing mid-resolution. Opportunity attacks triggering during a move that itself was triggered by a spell. Concentration checks that depend on which damage instance hit first. Area of effect needing to know about cover. The rules are a graph, not a list, and I kept finding that my ad hoc implementation would handle 80% of cases cleanly, then fall apart on the edges.
I looked at what was available before building my own, and there are some decent options out there — nothing I want to disparage, they're built by people who clearly put real work in. But nothing I found quite fit what I needed: something truly headless (no assumptions about my renderer or game loop), content-agnostic (rules engine separate from the spell and monster data), performant enough for responsive user interactions (and possibly run simulation workloads, ect), and robust enough to trust as a foundation for a serious project.
So I've been building one. It's written in Rust, compiles to WebAssembly, and is designed to plug into whatever — a browser game, a Godot project, a native app. I want to be able to reuse it in later projects. The content (classes, spells, skills, monsters,...) lives in separate content packages (JSON), so the engine itself has no opinion about what "Fireball" does, for example. I'm building it for my own game, but I've put real thought into the architecture, and it's starting to feel like something other developers might actually use.
Before I decide whether to open source it, I want a gut check from people who've actually tried to build in this space:
- Have you implemented SRD 5e (or similar TTRPG) rules for a project? Where did it get painful?
- Is "headless rules engine, you bring your own renderer and content" a thing you'd actually want, or would you want more batteries included?
- Would a well-documented, open-source Rust/WASM engine with TypeScript bindings be something you'd consider building on — or is that too low-level for most indie projects?
Curious whether the problem I ran into is common or just my particular situation. There is a lot of extra work I'd need to put in to make it open source, so I don't want to go down that road if it's not a common pain point for others. Happy to go into more detail on the architecture if there's interest.