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 skills/bigdra50/unity-cli/unity-verifynpx skills add bigdra50/unity-cli --skill unity-verifygit clone --depth 1 https://github.com/bigdra50/unity-cliWhat 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 | $0.00060 | $0.01116 |
| Opus 5 | $0.00030 | $0.00558 |
| Sonnet 5 | $0.00012 | $0.00223 |
| Haiku 4.5 | $0.00006 | $0.00112 |
Grade A, and why
unity-verify 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
unity-verify
PREREQUISITE:
../unity-shared/SKILL.md（Relay Server 経由で Unity Editor が起動/アクティブであること）
Quick Verify
コード変更するたびに実行。unity-shared の Quick Verify (#quick-verify) を参照。skill 経由のコマンドは必ず -i <instance> を付ける (unity-shared #インスタンス指定)。
u -i <instance> console clear
→ u -i <instance> refresh
→ isCompiling ポーリング (2s 間隔, 最大 30s)
→ u -i <instance> console get -l E,W | head -50
結果判定
u console get の出力に対して以下のテーブルで判定。出力ゼロ行 = エラー無しとして扱う (空 = OK)。
| 出力内容 | 判定 | 次のアクション |
|---|---|---|
| 空 (0 行) | クリーン | 完了。次工程 (Full Verify 等) に進める |
[ERROR] 行のみ無し ＋ [WARN] 行あり |
クリーン (Warning 許容) | Warning を 1〜3 行に要約してユーザー報告、続行 |
[ERROR] 行あり |
エラー | 下記「修正ループ」へ |
-l E,W は Error + Warning 両方を取得する。Warning だけなら続行可、Error が 1 件でもあれば停止して修正ループに入る。
修正ループ (Error 検出時)
最大 3 回までの「修正 → Quick Verify 再実行」サイクル。
- 修正前にユーザー確認 (default): 以下のいずれかに該当する場合は自動修正せずユーザーに修正方針を確認する
- ファイル名 prefix が
_(例:_Broken*.cs) または "Test" "Fixture" 等を含む - エラーファイル内コメントに
intentional,TODO,FIXME,tuning等の意図的破壊を示唆する語がある - 1 回の修正で 3 ファイル以上に変更が及ぶ場合
- 既存ロジックの大幅変更が必要な場合 (型変更、API 置換、import 追加以外の構造変更)
- ファイル名 prefix が
- 自動修正可: 単純なタイプミス (セミコロン抜け、スペル間違い、import 不足) で 1 ファイル限定の場合は試行 OK
- ループ運用: 修正 → Quick Verify を最大 3 回繰り返す。3 回で解消しない場合はユーザーに状況を報告し判断を仰ぐ
- 副作用回避: 1 回の修正は 1 ファイル単位。無関係な変更 (フォーマット整形、リファクタ) を混ぜない
Full Verify
ユーザーが要求した場合のみ。Quick Verify + EditMode テスト。
Quick Verify 実行
→ クリーン (Error 0、Warning は許容) なら u -i <instance> tests run edit → 結果確認
→ Fail あり: 報告、修正して Quick Verify から再実行 (修正ループの規定に従う)
Runtime Check
要求された場合のみ。Play Mode でランタイムエラーを検出。
u -i <instance> console clear
→ u -i <instance> play
→ isPlaying ポーリング
→ 3秒待機
→ u -i <instance> console get -l +E+X
→ u -i <instance> stop
報告のみ。自動修正せずユーザーに判断を委ねる。
Auto-trigger
以下の編集後に Quick Verify を自動実行:
.cs/.shader/.compute.asmdef/.asmref- Unity パッケージ関連 (
package.json/manifest.json)
スキップ: コメントのみの変更、プロジェクト外ファイル、ユーザー指示。
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 · 89 lines · 60 tokens per session scan A e20ced00ce17
unity-verify is a skill published in the GitHub repository bigdra50/unity-cli (39 stars, last pushed 2mo ago), licensed MIT. It adds 60 tokens to every session and 1,116 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
unity
Compile, test, and drive Unity for this repo's C# packages (unity/core, jint, quickjs, clearscript) and the two Unity projects (tests/, kitchen-sink/). Use when a change touches C# under unity/, when Unity test results are needed, when a rendering snapshot has to be checked or regenerated, or when the app has to be…
uloop-execute-dynamic-code
Execute C# with Unity APIs when existing uloop tools cannot inspect or edit enough. Use for reachable scene/component state, scene/prefab/menu automation, and PlayMode checks.
uloop-simulate-mouse-input
Simulate Mouse.current input in PlayMode through Unity Input System. Use for gameplay mouse clicks, long-press (LongPress), movement delta (MoveDelta/SmoothDelta), or scroll. Use --dry-run to check what a Game View coordinate hits in 3D physics before clicking (works in EditMode; no Input System required). Use…
uloop-pause-point
Pauses Unity playback at any source file:line without editing code or recompiling, and returns a snapshot of the locals, parameters, and instance fields at that exact frame. Use for bug investigation, PlayMode/E2E verification, checking variable values at a specific frame, or confirming that a code path executed.
uloop-screenshot
Capture Unity Editor windows or Game View rendering as PNG. Use for visual checks, debugging, documentation, or annotated UI element coordinates.
uloop-simulate-keyboard
Simulate keyboard input in PlayMode through Unity Input System. Use for key presses, holds (via Press --duration or KeyDown/KeyUp), releases, and game controls such as WASD or Space. Requires the Input System package (com.unity.inputsystem).