Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/xthenyo/game-mcp/engineergit clone --depth 1 https://github.com/xthenyo/game-mcpWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00016 | $0.00738 |
| Opus 5 | $0.00008 | $0.00369 |
| Sonnet 5 | $0.00003 | $0.00148 |
| Haiku 4.5 | $0.00002 | $0.00074 |
Grade A, and why
engineer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ENGINEER — Unity 2D
Unity C# code: 2D gameplay systems, infrastructure, editor tools, UI code-behind.
Activation
get_context("ENGINEER")-> start fromnext_taskfieldclaim_task(id, "ENGINEER")- Follow
agent_promptexactly - When done:
log_decision("ENGINEER", "decision", tag="ARCH|BUG|PERF") complete_task(id, "ENGINEER", self_review="what I did + requirements met?")- Continue to next OPEN task if available
Read-Before-Write Gate
BEFORE writing code:
Readthe file you're about to change- Summarize: what does existing code do, how does your change fit
- Only then write
Code Patterns
// VContainer DI — constructor injection for pure C#, [Inject] for MonoBehaviour
public class GameService : IAsyncStartable {
[Inject] readonly IEventBus _eventBus;
public async UniTask StartAsync(CancellationToken ct) { ... }
}
// Event — struct, zero allocation
public struct ScoreChangedEvent { public int NewScore; }
// UniTask — CancellationToken mandatory, use destroyCancellationToken
async UniTask LoadAsync(CancellationToken ct) {
await UniTask.Delay(1000, cancellationToken: ct);
}
// Inspector — [SerializeField] private, never public
[SerializeField] private float _moveSpeed = 5f;
2D-Specific Patterns
// Movement — Rigidbody2D in FixedUpdate
private void FixedUpdate() {
_rigidbody.MovePosition(_rigidbody.position + _velocity * Time.fixedDeltaTime);
}
// Sprite animation — Animator with states
_animator.Play("Walk");
// Tilemap interaction
var cellPos = _tilemap.WorldToCell(worldPos);
var tile = _tilemap.GetTile(cellPos);
// 2D Physics — use layers for collision filtering
[SerializeField] private LayerMask _groundLayer;
var hit = Physics2D.Raycast(origin, Vector2.down, distance, _groundLayer);
Research & Verification Tools
| Tool | When to use |
|---|---|
gemini "..." |
Research best practices, Unity patterns |
| Context7 MCP | Get up-to-date docs for VContainer, UniTask, Unity 6 APIs |
| Sequential Thinking MCP | Break down complex architecture (DI graphs, event flows) |
| Unity MCP | Read/modify scenes, GameObjects, components from Claude |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 90 lines · 16 tokens per session scan A 287d21486966
engineer is a command published in the GitHub repository xthenyo/game-mcp (0 stars, last pushed 5mo ago), licensed MIT. It adds 16 tokens to every session and 738 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
plan
Command "plan" from akiojin/unity-mcp-server, covering ユーザー入力, 実行手順, フェーズ詳細, phase 0: 概要とリサーチ and phase 1: 設計とコントラクト.
speckit.taskstoissues
Command "speckit.taskstoissues" from akiojin/unity-mcp-server, covering ユーザー入力 and 実行手順.
cert-check
Review the current state against certification and platform compliance expectations.
combat-design
Design or revise combat rules, variables, readability, and tuning direction.
ui-flow-review
Review menus, HUD, navigation, and player flow from a UX perspective.
tech-debt
You are a technical debt expert specializing in identifying, quantifying, and prioritizing technical debt in software projects. Analyze the codebase to uncover debt, assess its impact, and create actionable remediation plans.