Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问clear审计通过

game-developer游戏开发商

Agent Skill

game-developer 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

5,508

周安装

225

GitHub Stars

76

下载量

1,782
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:game-developer(游戏开发商)
来源仓库:https://github.com/404kidwiz/claude-supercode-skills
仓库路径:skills/game-developer
安装命令:
npx skills add https://github.com/404kidwiz/claude-supercode-skills --skill game-developer
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/404kidwiz/claude-supercode-skills --skill game-developer

简介

game-developer 提供游戏开发专业支持,专注 Unity 和 Unreal Engine 项目。

  • 适用于原型设计、图形优化、多人联机及 VR/AR 体验开发等场景。
  • 通过关键词或任务定位候选方案,适合在主流 AI 宿主中快速检索相关技术建议。
  • 安装前请确认权限范围和维护状态,注意可能涉及联网或文件读写操作。
  • 建议结合原始 README 和仓库内容核验具体实现细节与使用限制。

SKILL.md

Game Developer

Purpose

Provides interactive entertainment development expertise specializing in Unity (C#) and Unreal Engine (C++). Builds 2D/3D games with gameplay programming, graphics optimization, multiplayer networking, and engine architecture for immersive gaming experiences.

When to Use

  • Prototyping game mechanics (Character controllers, combat systems)
  • Optimizing graphics performance (Shaders, LODs, Occlusion Culling)
  • Implementing multiplayer networking (Netcode for GameObjects, Mirror, Unreal Replication)
  • Designing level architecture and streaming systems
  • Developing VR/AR experiences (OpenXR, ARKit)
  • Creating custom editor tools and pipelines


2. Decision Framework

Engine Selection

Which engine fits the project?
│
├─ **Unity**
│  ├─ Mobile/2D/VR? → **Yes** (Best ecosystem, smaller build size)
│  ├─ Team knows C#? → **Yes**
│  └─ Stylized graphics? → **Yes** (URP is flexible)
│
├─ **Unreal Engine 5**
│  ├─ Photorealism? → **Yes** (Nanite + Lumen out of box)
│  ├─ Open World? → **Yes** (World Partition system)
│  └─ Team knows C++? → **Yes** (Or Blueprints visual scripting)
│
└─ **Godot**
   ├─ Open Source requirement? → **Yes** (MIT License)
   ├─ Lightweight 2D? → **Yes** (Dedicated 2D engine)
   └─ Linux native dev? → **Yes** (Excellent Linux support)

Multiplayer Architecture

ModelDescriptionBest For
Client-Hosted (P2P)One player is host.Co-op games, Fighting games (with rollback). Cheap.
Dedicated ServerAuthoritative server in cloud.Competitive Shooters, MMOs. Prevents cheating.
Relay ServerRelay service (e.g., Unity Relay).Session-based games avoiding NAT issues.

Graphics Pipeline (Unity)

PipelineTargetPros
URP (Universal)Mobile, VR, Switch, PCHigh perf, customizable, large asset store support.
HDRP (High Def)PC, PS5, Xbox Series XPhotorealism, Volumetric lighting, Compute shaders.
Built-inLegacyAvoid for new projects.

Red Flags → Escalate to graphics-engineer (Specialist):

  • Writing custom rendering backends (Vulkan/DirectX/Metal) from scratch
  • Debugging driver-level GPU crashes
  • Implementing novel GI (Global Illumination) algorithms


Workflow 2: Unreal Engine Multiplayer Setup

Goal: Replicate a variable (Health) from Server to Clients.

Steps:

  1. Header (Character.h) UPROPERTY(ReplicatedUsing=OnRep_Health) float Health; UFUNCTION() void OnRep_Health(); void GetLifetimeReplicatedProps(TArray<FLifetimeProperty>& OutLifetimeProps) const override;
  2. Implementation (Character.cpp) void AMyCharacter::GetLifetimeReplicatedProps(TArray<FLifetimeProperty>& OutLifetimeProps) const {Super::GetLifetimeReplicatedProps(OutLifetimeProps); DOREPLIFETIME(AMyCharacter, Health);} void AMyCharacter::TakeDamage(float DamageAmount) {if (HasAuthority()) {Health -= DamageAmount; // OnRep_Health() called automatically on clients // Must call manually on Server if needed OnRep_Health();}}
  3. Blueprint Integration

- Bind UI Progress Bar to Health variable. - Test with "Play as Client" (NetMode).



Workflow 4: VFX Graph & Shader Graph (Visual Effects)

Goal: Create a GPU-accelerated particle system for a magic spell.

Steps:

  1. Shader Graph (The Look)

- Create Unlit Shader Graph. - Add Voronoi Noise node scrolling with Time. - Multiply with Color property (HDR). - Connect to Base Color and Alpha. - Set Surface Type to Transparent / Additive.

  1. VFX Graph (The Motion)

- Create Visual Effect Graph asset. - Spawn Context: Constant Rate (1000/sec). - Initialize: Set Lifetime (0.5s - 1s), Set Velocity (Random Direction). - Update: Add Turbulence (Noise Field) to simulate wind. - Output: Set Quad Output to use the Shader Graph created above.

  1. Optimization

- Use GPU Events if particles need to trigger gameplay logic (e.g., damage). - Set Bounds correctly to avoid culling issues.



5. Anti-Patterns & Gotchas

❌ Anti-Pattern 1: Heavy Logic in Update()

What it looks like:

  • Performing FindObjectOfType, GetComponent, or heavy math every frame.

Why it fails:

  • Kills CPU performance.
  • Drains battery on mobile.

Correct approach:

  • Cache references in Start() or Awake().
  • Use Coroutines or InvokeRepeating for logic that doesn't need to run every frame (e.g., AI pathfinding updates every 0.5s).

❌ Anti-Pattern 2: Trusting the Client

What it looks like:

  • Client sends "I shot player X" to server.
  • Server applies damage immediately.

Why it fails:

  • Cheaters can send fake packets.

Correct approach:

  • Authoritative Server: Client sends "I fired". Server calculates hit. Server tells Client "You hit".
  • Use prediction/reconciliation to mask latency for the local player.

❌ Anti-Pattern 3: God Classes

What it looks like:

  • PlayerController.cs has 2000 lines handling Movement, Combat, Inventory, UI, and Audio.

Why it fails:

  • Spaghetti code.
  • Hard to debug.

Correct approach:

  • Composition: PlayerMovement, PlayerCombat, PlayerInventory.
  • Use components to split responsibility.


7. Quality Checklist

Performance:

  • Frame Rate: Stable 60fps on target hardware.
  • GC Alloc: 0 bytes allocated per frame in main gameplay loop.
  • Draw Calls: Batched appropriately (check Frame Debugger).
  • Load Times: Async loading used for scenes/assets.

Code Architecture:

  • Decoupled: Systems communicate via Events/Interfaces, not hard dependencies.
  • Clean: No "God Classes" > 500 lines.
  • Version Control: Large binaries (textures, audio) handled via Git LFS.

UX/Polish:

  • Controls: Input remapping supported.
  • UI: Scales correctly for different aspect ratios (16:9, 21:9, Mobile Notches).
  • Feedback: Audio/Visual cues for all player actions (Juice).

Examples

Example 1: 2D Platformer Game Development

Scenario: Building a commercial 2D platformer with physics-based gameplay.

Implementation:

  1. Physics: Custom physics engine for responsive platforming
  2. Animation: Sprite-based animation with state machines
  3. Level Design: Tilemap-based levels with procedural elements
  4. Audio: Spatial audio system with adaptive music

Technical Approach:

# Character controller pattern
class PlayerCharacter:
    def update(self, dt):
        input = self.input_system.get_player_input()
        velocity = self.physics.apply_gravity(velocity, dt)
        velocity = self.handle_movement(input, velocity)
        displacement = self.physics.integrate(velocity, dt)
        self.handle_collisions(displacement)
        self.animation.update_state(velocity, input)

Example 2: VR Experience Development

Scenario: Creating an immersive VR experience for Oculus/Meta Quest.

VR Implementation:

  1. Locomotion: Teleportation and smooth movement options
  2. Interaction: Hand tracking with gesture recognition
  3. Optimization: Single-pass stereo rendering
  4. Comfort: Comfort mode options for sensitive users

Key Considerations:

  • 72Hz minimum frame rate for comfort
  • Motion sickness avoidance in design
  • Hand physics for realistic interaction
  • Battery optimization for standalone headsets

Example 3: Multiplayer Battle Royale

Scenario: Developing a competitive multiplayer game with 100 players.

Multiplayer Architecture:

  1. Networking: Client-side prediction with server reconciliation
  2. Lag Compensation: Interpolation and extrapolation techniques
  3. Anti-Cheat: Server-side validation, cheat detection
  4. Matchmaking: Skill-based matchmaking with queue optimization

Best Practices

Game Development

  • Core Loop First: Prototype and refine the core gameplay loop
  • Modular Architecture: Decouple systems for maintainability
  • Performance Budgeting: Define and monitor performance targets
  • Data-Driven Design: Use configuration files for game balance
  • Version Control: Handle large binary assets appropriately

Physics and Movement

  • Determinism: Ensure consistent physics across networked games
  • Collision Detection: Optimize for minimal false positives
  • Character Controllers: Separate physics from character logic
  • Ragdoll Physics: Use for death animations and interaction
  • Performance: Profile physics update time, optimize as needed

Graphics and Rendering

  • Batching: Group draw calls for GPU efficiency
  • Level of Detail: Implement LOD for models and textures
  • Shaders: Optimize shader complexity, use shared materials
  • Lighting: Balance quality and performance, use baked lighting
  • Post-Processing: Apply selectively, profile GPU impact

Audio Implementation

  • Spatial Audio: 3D positioning for immersion
  • Adaptive Music: Dynamic soundtrack based on gameplay
  • Performance: Stream large audio files, pool sound effects
  • Compression: Use appropriate audio compression formats
  • Accessibility: Provide audio cues as alternatives to visual feedback

Testing and Quality

  • Playtesting: Regular playtesting sessions for feedback
  • Performance Profiling: Monitor frame rate, memory, load times
  • Platform Testing: Test on target hardware, not just dev machines
  • Accessibility: Implement accessibility features from start
  • Localization: Plan for international markets early

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

27.78%
按下载量换算495

OpenCode

24.63%
按下载量换算439

Codex

16.61%
按下载量换算296

Cursor

11.94%
按下载量换算213

Gemini CLI

8.61%
按下载量换算153

windsurf

3.79%
按下载量换算68

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/404kidwiz/claude-supercode-skills --skill game-developer;npx skills add 404kidwiz/claude-supercode-skills --skill "game-developer" 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills