mcpField
G
game-development

game-development

github.com/miles990Development26415 installsUpdated 41w ago
Repository

About game-development

A practical guide to game development across platforms, covering architecture, rendering, physics, AI, multiplayer, and optimization. It combines common patterns and code examples with references to game engines and related skills.

Key features

  1. 01Game loops and ECS architecture
  2. 022D animation and collision detection
  3. 03State machines, behavior trees, and pathfinding
  4. 04Multiplayer networking and state sync
  5. 05Rendering and memory optimization
  6. 06Game engine references

Use cases

  1. 01Design a game loop and entity system
  2. 02Implement 2D collisions and character animation
  3. 03Plan enemy AI and pathfinding
  4. 04Optimize rendering and object memory use

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/miles990/claude-software-skills --skill game-development

After installing, type/game-developmentin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

flame/flame-templates/references/
platformer.md
roguelike.md
rpg.md

Details

From the project README

Game Development

Overview

Patterns and practices for building games across platforms. Covers architecture, rendering, physics, AI, multiplayer, and optimization.


Game Architecture

Game Loop
┌─────────────────────────────────────────────────────────────┐
│                      Game Loop                              │
│                                                             │
│   ┌─────────┐    ┌─────────┐    ┌─────────┐    ┌─────────┐ │
│   │  Input  │ ─→ │ Update  │ ─→ │ Physics │ ─→ │ Render  │ │
│   │ Process │    │  Logic  │    │  Step   │    │  Frame  │ │
│   └─────────┘    └─────────┘    └─────────┘    └─────────┘ │
│        ↑                                             │      │
│        └─────────────────────────────────────────────┘      │
│                        Next Frame                           │
└─────────────────────────────────────────────────────────────┘
Fixed vs Variable Timestep
TypeUse CaseCode Pattern
FixedPhysics, determinismwhile (accumulator >= dt) { update(dt); }
VariableRendering, animationsupdate(deltaTime);
HybridMost gamesFixed physics, variable render
// Hybrid game loop
let accumulator = 0;
const FIXED_DT = 1/60;

function gameLoop(currentTime: number) {
  const deltaTime = currentTime - lastTime;
  accumulator += deltaTime;

  // Fixed timestep for physics
  while (accumulator >= FIXED_DT) {
    physicsUpdate(FIXED_DT);
    accumulator -= FIXED_DT;
  }

  // Variable timestep for rendering
  const alpha = accumulator / FIXED_DT;
  render(alpha); // Interpolate between states

  requestAnimationFrame(gameLoop);
}

Entity Component System (ECS)
┌─────────────────────────────────────────────────────────────┐
│                    ECS Architecture                         │
│                                                             │
│  Entity: Just an ID                                         │
│  ┌─────┐  ┌─────┐  ┌─────┐                                 │
│  │  1  │  │  2  │  │  3  │                                 │
│  └─────┘  └─────┘  └─────┘                                 │
│                                                             │
│  Components: Pure Data                                      │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐                  │
│  │ Position │  │ Velocity │  │  Sprite  │                  │
│  │ x, y, z  │  │ vx, vy   │  │ texture  │                  │
│  └──────────┘  └──────────┘  └──────────┘                  │
│                                                             │
│  Systems: Logic                                             │
│  ┌────────────────┐  ┌────────────────┐                    │
│  │ MovementSystem │  │ RenderSystem   │                    │
│  │ Position+Vel   │  │ Position+Sprite│                    │
│  └────────────────┘  └────────────────┘                    │
└─────────────────────────────────────────────────────────────┘
// Component definitions
interface Position { x: number; y: number; }
interface Velocity { vx: number; vy: number; }
interface Sprite { texture: string; width: number; height: number; }

// System
function movementSystem(entities: Entity[], dt: number) {
  for (const entity of entities) {
    if (entity.has(Position) && entity.has(Velocity)) {
      const pos = entity.get(Position);
      const vel = entity.get(Velocity);
      pos.x += vel.vx * dt;
      pos.y += vel.vy * dt;
    }
  }
}

2D Game Development

Sprite Animation
interface Animation {
  frames: string[];       // Texture names
  frameDuration: number;  // Seconds per frame
  loop: boolean;
}

class AnimatedSprite {
  private currentFrame = 0;
  private elapsed = 0;

  update(dt: number) {
    this.elapsed += dt;
    if (this.elapsed >= this.animation.frameDuration) {
      this.elapsed = 0;
      this.currentFrame++;
      if (this.currentFrame >= this.animation.frames.length) {
        this.currentFrame = this.animation.loop ? 0 : this.animation.frames.length - 1;
      }
    }
  }

  get texture(): string {
    return this.animation.frames[this.currentFrame];
  }
}
Collision Detection
MethodComplexityUse Case
AABBO(n²) → O(n log n)Boxes, simple shapes
CircleO(n²)Projectiles, characters
SATO(n²)Complex convex polygons
Pixel PerfectExpensivePrecise collision
// AABB collision
function aabbIntersect(a: AABB, b: AABB): boolean {
  return a.x < b.x + b.width &&
         a.x + a.width > b.x &&
         a.y < b.y + b.height &&
         a.y + a.height > b.y;
}

// Spatial hash for broad phase
class SpatialHash {
  private cells = new Map<string, Entity[]>();
  private cellSize: number;

  insert(entity: Entity) {
    const key = this.getKey(entity.position);
    if (!this.cells.has(key)) this.cells.set(key, []);
    this.cells.get(key)!.push(entity);
  }

  query(position: Vector2): Entity[] {
    // Check neighboring cells
    const nearby: Entity[] = [];
    for (let dx = -1; dx <= 1; dx++) {
      for (let dy = -1; dy <= 1; dy++) {
        const key = this.getKey({ x: position.x + dx * this.cellSize, y: position.y + dy * this.cellSize });
        nearby.push(...(this.cells.get(key) || []));
      }
    }
    return nearby;
  }
}

Game AI

Finite State Machine
interface State {
  enter(): void;
  update(dt: number): void;
  exit(): void;
}

class EnemyAI {
  private states = new Map<string, State>();
  private currentState: State;

  transition(stateName: string) {
    this.currentState?.exit();
    this.currentState = this.states.get(stateName)!;
    this.currentState.enter();
  }
}

// Example: Patrol → Chase → Attack
class PatrolState implements State {
  enter() { this.setAnimation('walk'); }
  update(dt: number) {
    this.patrol();
    if (this.canSeePlayer()) {
      this.fsm.transition('chase');
    }
  }
  exit() {}
}
Behavior Trees
┌─────────────────────────────────────────────────────────────┐
│                    Behavior Tree                            │
│                                                             │
│                      [Selector]                             │
│                     /          \                            │
│              [Sequence]      [Patrol]                       │
│              /        \                                     │
│        [CanSee?]    [Attack]                               │
│            │                                                │
│         [Chase]                                             │
└─────────────────────────────────────────────────────────────┘
Pathfinding (A*)
function aStar(start: Node, goal: Node): Node[] {
  const openSet = new PriorityQueue<Node>();
  const cameFrom = new Map<Node, Node>();
  const gScore = new Map<Node, number>();
  const fScore = new Map<Node, number>();

  gScore.set(start, 0);
  fScore.set(start, heuristic(start, goal));
  openSet.enqueue(start, fScore.get(start)!);

  while (!openSet.isEmpty()) {
    const current = openSet.dequeue()!;

    if (current === goal) {
      return reconstructPath(cameFrom, current);
    }

    for (const neighbor of getNeighbors(current)) {
      const tentativeG = gScore.get(current)! + distance(current, neighbor);

      if (tentativeG < (gScore.get(neighbor) ?? Infinity)) {
        cameFrom.set(neighbor, current);
        gScore.set(neighbor, tentativeG);
        fScore.set(neighbor, tentativeG + heuristic(neighbor, goal));
        openSet.enqueue(neighbor, fScore.get(neighbor)!);
      }
    }
  }

  return []; // No path found
}

Multiplayer Games

Network Architecture
ModelLatencyComplexityUse Case
Peer-to-PeerLowMediumFighting games, small lobbies
Client-ServerMediumHighMost online games
Authoritative ServerHigherHighestCompetitive games
Lag Compensation
// Client-side prediction
class NetworkedPlayer {
  private pendingInputs: Input[] = [];
  private serverState: PlayerState;

  update(input: Input) {
    // 1. Apply input locally (prediction)
    this.applyInput(input);
    this.pendingInputs.push(input);

    // 2. Send to server
    this.sendInput(input);
  }

  onServerUpdate(state: PlayerState, lastProcessedInput: number) {
    // 3. Reconcile with server
    this.serverState = state;

    // 4. Re-apply unprocessed inputs
    this.pendingInputs = this.pendingInputs.filter(i => i.id > lastProcessedInput);
    for (const input of this.pendingInputs) {
      this.applyInput(input);
    }
  }
}
State Synchronization
// Delta compression
interface StateDelta {
  timestamp: number;
  changes: Map<EntityId, ComponentChanges>;
}

function computeDelta(prev: GameState, curr: GameState): StateDelta {
  const changes = new Map();
  for (const [id, entity] of curr.entities) {
    const prevEntity = prev.entities.get(id);
    if (!prevEntity || hasChanged(prevEntity, entity)) {
      changes.set(id, getChangedComponents(prevEntity, entity));
    }
  }
  return { timestamp: curr.timestamp, changes };
}

Game Optimization

Rendering Optimization
TechniqueBenefitImplementation
BatchingReduce draw callsCombine sprites with same texture
CullingSkip invisible objectsFrustum culling, occlusion culling
LODReduce polygon countDistance-based model switching
InstancingEfficient duplicatesGPU instancing for repeated objects
Memory Optimization
// Object pooling
class ObjectPool<T> {
  private pool: T[] = [];
  private factory: () => T;

  acquire(): T {
    return this.pool.pop() ?? this.factory();
  }

  release(obj: T) {
    this.reset(obj);
    this.pool.push(obj);
  }
}

// Usage
const bulletPool = new ObjectPool(() => new Bullet());

function fireBullet() {
  const bullet = bulletPool.acquire();
  bullet.init(position, direction);
  activeBullets.add(bullet);
}

function onBulletHit(bullet: Bullet) {
  activeBullets.delete(bullet);
  bulletPool.release(bullet);
}

Game Engines Reference

EngineLanguageBest ForPlatformSkill
UnityC#Mobile, indie, VRAll-
UnrealC++, BlueprintAAA, realisticPC, Console-
GodotGDScript, C#Indie, 2DAll-
FlameDartFlutter 2D, casualAllflame/
PhaserJavaScriptWeb 2DBrowser-
Three.jsJavaScriptWeb 3DBrowser-
BevyRustPerformanceDesktop-
LÖVELuaSimple 2DDesktop-
Flame Engine (Flutter)

專為 Flutter 開發者設計的 2D 遊戲引擎,詳細文件請參考 flame/SKILL.md:

  • flame-core/ - 組件、輸入、碰撞、相機、動畫、音效、粒子
  • flame-systems/ - 14 個遊戲系統(任務、對話、背包、戰鬥等)
  • flame-templates/ - RPG、Platformer、Roguelike 模板

Related Skills

  • [[performance-optimization]] - General optimization
  • [[realtime-systems]] - WebSocket, networking
  • [[cpp]] - Performance-critical code
  • [[javascript-typescript]] - Web games

Comments

Markdown supported