mcpbeat Sign in

Unity Scriptdesign Agent Skill

Advises on Unity gameplay script quality — code review, reducing coupling, improving maintainability, and refactoring. Use when reviewing code quality, untangling tightly-coupled scripts, or planning a refactor for maintainability, even if the user just says "看看我代码" or "代码有点乱". 为 Unity 游戏脚本质量提供建议(代码审查、降低耦合、提升可维护性、重构);当用户要审查代码质量、理顺高耦合脚本、或为可维护性规划重构时使用。

1k tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
1536
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/Besty0728/Unity-Skills --skill unity-scriptdesign

The instruction itself

7 sections, as written by the author

Unity Script Design Review

Use this skill before creating gameplay scripts, or after scripts are generated and need a design pass.

Review Checklist

  • Responsibility: does the script have one clear job?
  • Role: should it really be a MonoBehaviour, ScriptableObject, or plain C# class?
  • Coupling: are dependencies explicit instead of hidden globals or deep scene lookups?
  • Communication: should this be a direct reference, interface call, or event?
  • Performance: is there unnecessary Update, repeated Find, avoidable allocation, or reflection in hot paths?
  • Lifecycle: are subscriptions, timers, and async work cleaned up clearly?
  • Inspector UX: are serialized fields private, grouped, and explained?
  • Testability: can the core logic move into a plain C# class?
  • Naming: do class and field names explain intent without cryptic abbreviations?

Data Lifecycle Boundary

The Review Checklist above asks "where does this *class* live". Ask the same question for every *field*. Every piece of state has one of three lifecycles, and putting a field on the wrong one is the most common cause of "why did this break when the designer tweaked a value" and "why are my unit tests flaky".

| Lifecycle | When the value is decided | Where it belongs | Typical idiom |

|-----------|--------------------------|------------------|---------------|

| Authoring-time | By a designer in the Editor, before Play | ScriptableObject asset, or [SerializeField] private on a prefab | Immutable at runtime; read via _config.Speed |

| Composition-time | Once per scene/instance, at Awake/Start | private field, assigned from GetComponent / GetComponentInChildren / ctor arg | Cached reference, no per-frame lookup |

| Runtime-mutable | Every frame or on gameplay events | private backing field + public read-only property + event | Exposed via public float Health { get; private set; } + OnHealthChanged |

Typical assignments

  • Weapon damage / fire rate / clip size → Authoring-time (ScriptableObject so balance can be hot-swapped).
  • Enemy AI's current target TransformComposition-time if set once at spawn, Runtime-mutable if re-targeted each frame.
  • Player current HP → Runtime-mutable with event. Never public float hp;.
  • Reference to Rigidbody/Animator on the same GameObject → Composition-time, cached in Awake.
  • Level music track → Authoring-time via ScriptableObject level descriptor.
  • "Is in combat" flag → Runtime-mutable, but usually derived from other state — review whether it should be a field at all.

Why the separation matters

Mixing the three lifecycles is what turns a clean class into a god object. A MonoBehaviour whose public float speed is edited by both the Inspector and a power-up script has two owners and no invariant; a bug in either path corrupts the other. The ECS baking pipeline makes this distinction a hard architectural boundary (Authoring → Baker → System), and the discipline transfers directly: if you would not mix an Authoring component with runtime write-back in ECS, do not mix them in a MonoBehaviour either. *Source: EntitiesSamples/Docs/baking.md:5-16.*

Guardrails

> Mode: Documentation only — no REST skills to gate; load freely under any operating mode (Approval / Auto / Bypass).

  • Prefer descriptive names over local shorthand.
  • Do not “optimize” readability away for imagined productivity gains.
  • Do not recommend complex patterns if a smaller refactor fixes the real problem.

Output Format

  • Keep: what is already good
  • Simplify: what should stay straightforward
  • Refactor: the highest-value structural change
  • Performance notes: only real hotspots, not theoretical micro-optimizations
  • Maintainability notes: naming, ownership, coupling, editor usability

Other skills for the same job

different authors, same section of the catalogue
Receiving Code Review
by ZhanlinCui
×7

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

2k tokens
Requesting Code Review
by ZhanlinCui
×6

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

2k tokens
Git Commit
by github
vendor ×3

Execute git commit with conventional commit message analysis, intelligent staging, and message generation. Use when user asks to commit changes, create a git commit, or mentions "/commit". Supports: (1) Auto-detecting type and scope from changes, (2) Generating conventional commit messages from diff, (3) Interactive commit with optional type/scope/description overrides, (4) Intelligent file staging for logical grouping

799 tokens
Github Code Review
by ComeOnOliver
×3

Comprehensive GitHub code review with AI-powered swarm coordination

13k tokens
Karpathy Guidelines
by hyyhf
×3

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

629 tokens
Code Reviewer
by google-gemini
vendor ×2

Use this skill to review code. It supports both local changes (staged or working tree) and remote Pull Requests (by ID or URL). It focuses on correctness, maintainability, and adherence to project standards.

795 tokens
Agent MD Refactor
by softaworks
×2

Refactor bloated AGENTS.md, CLAUDE.md, or similar agent instruction files to follow progressive disclosure principles. Splits monolithic files into organized, linked documentation.

4k tokens
Commit Work
by softaworks
×2

Create high-quality git commits: review/stage intended changes, split into logical commits, and write clear commit messages (including Conventional Commits). Use when the user asks to commit, craft a commit message, stage changes, or split work into multiple commits.

2k tokens

How to use it

Copy the folder

Take besty0728/unity-scriptdesign from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.