A large-scale cellular-automaton battle simulator. A grid of very small units, each a single "fighter," continuously battles its neighbors using configurable Rock-Paper-Scissors-style rules — from classic 3-type RPS up to higher-order systems with 5, 7, 13, 37+ types. Each fighter's color represents its current move; on screen it renders as a small configurable block of pixels rather than a literal single pixel.
Pre-implementation. Documentation scaffold is in place; language, framework, and game design are undecided.
| Document | Description |
|---|---|
| AGENTS.md | Operating manual for AI agents |
| plan/README.md | Development plan, phases, and decision log |
| INPUTS.md | Append-only log of every user prompt |
The plan lives in plan/ as a set of feature files indexed by plan/README.md.
| Area | Status | Candidates |
|---|---|---|
| Language | TBD | Kotlin, Zig |
| Graphics / rendering | TBD | raylib, others |
| Build tooling | TBD | — |
| CI / asset pipeline | TBD | — |
Not yet applicable — no build system or source code exists.
TBD