Rock Paper Scissors
Level: Beginner 15–30 minConcepts: Algorithms
Solutions: C# | TypeScript | Python
Rock Paper Scissors is a game involving two players making pre-defined hand gestures at each other. The gesture that each player uses is played against the other, with a winner being decided based on the rules being used.
The three gestures used in base Rock Paper Scissors are… well… rock, paper, and scissors. The way these are scored is as such: Rock beats Scissors, Scissors beats Paper, and Paper beats Rock. It gets a lot more complicated when you introduce new gestures, but let’s keep it simple for now.
Now create a backend for the game that we can use to hook up to our many game clients we’re going to be creating.
Test Cases
| Player Move | Opponent Move | Result |
|---|---|---|
| Paper | Scissors | Player Loses |
| Paper | Paper | Tie |
| Rock | Scissors | Player Wins |
| Rock | Paper | Player Loses |
| Rock | Rock | Tie |
| Scissors | Paper | Player Wins |
| Scissors | Rock | Player Loses |
| Scissors | Scissors | Tie |
Bonus
Extend the game engine to include the rules for Rock, Paper, Scissors and Spock. The new moves to include are:
- Spock smashes Scissors
- Paper disproves Spock
- Spock vaporizes Rock
After implementing the additional rules make sure your code has no if statements.
Reference Walkthrough
Reference implementations in C#, TypeScript, and Python live at tddbuddy-reference-katas/rock-paper-scissors. Nine scenarios — the full 3×3 matrix of plays from player 1’s perspective — are shared across all three languages, each a single pure function decide(Play, Play) → Outcome. The Spock extension and the no-if-statements constraint from the bonus section are intentionally out of scope for the reference; the three-way base game is sufficient to demonstrate the F1 shape with typed enums.
- C# (.NET 8, xUnit, FluentAssertions 6.12.0) — walkthrough
- TypeScript (Node 20, Vitest 1.6, TS 5 strict) — walkthrough
- Python (3.11, pytest) — walkthrough
This kata ships in Agent Full-Bake mode at high gear (F1 tier): the algorithm is small enough to land as one commit per language, with a brief walkthrough noting there are no builders — but the plays and outcomes are modelled as typed enums (C# enum, TypeScript string-literal unions, Python StrEnum) rather than bare strings. The closed set of three plays and three outcomes is the ubiquitous language, and typed enums prevent "Rock" vs "rock" drift at call sites. See the repo’s Gears section for when high gear is the right call.
Related reading
- Katas Are Rehearsal, Not Performance
You don't practice TDD on production code. You practice on katas and bring the muscle memory to production. The gap between knowing TDD and doing TDD is reps. - Test Deletion Is a Privileged Operation
The cheapest way for an agent to make a failing test pass is to delete it. That is logical for the agent and catastrophic for the codebase. Tests are append-only by default. Deletion needs a human author, a separate commit, and a separate review. - Refactor Was Always Where Compounding Lived
Red and green were the visible two thirds of TDD. Refactor was the optional polish. Codebases that compound did the refactor pass; codebases that accrete did not. Agents skip refactor by default because it has no green-bar reward, and a year of skipped refactors is how a codebase loses its shape.