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/jason-hub-star/unityctl/unityctl-workflowsnpx skills add Jason-hub-star/unityctl --skill unityctl-workflowsgit clone --depth 1 https://github.com/Jason-hub-star/unityctlWhat 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 | $0.00079 | $0.00903 |
| Opus 5 | $0.00039 | $0.00451 |
| Sonnet 5 | $0.00016 | $0.00181 |
| Haiku 4.5 | $0.00008 | $0.00090 |
Grade A, and why
unityctl-workflows scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
unityctl Workflows
Use unityctl as the execution and evidence layer. Keep source-code edits in normal filesystem tools; use unityctl for Unity-owned state and live Editor operations.
Establish the target
- Resolve the Unity project root containing
Assets/,Packages/, andProjectSettings/. - Pass
--projectexplicitly on every project-scoped command. Discovery commands such aseditor currentandeditor instancesare global and do not accept it. - Check the installed CLI and project:
unityctl --version
unityctl doctor --project "$P" --json
- If the bridge is missing, install the bundled bridge, then open Unity and wait for stability:
unityctl init --project "$P"
unityctl await-ready --project "$P" --timeout 300 --json
Do not run init again when doctor reports a healthy matching bridge.
Discover, do not guess
Use the live command surface instead of memorizing flags:
unityctl tools --json
unityctl <command> --help
tools --json returns a top-level array. Use names returned by it; the machine
command catalog uses hyphenated names while the CLI may expose grouped verbs.
unityctl tools --json | jq '.[].name'
Run the closed loop
- Read the current state, including
scene hierarchy --summary. - If any affected scene is already dirty, stop before writing. Ask whether to save the existing work or let the user revert it; do not mix ownership.
- Compare
editor current --jsonwith the intended project. Before writes, select the intended target explicitly when they differ:
unityctl editor current --json
unityctl editor select --project "$P" --json
- Make the smallest Unity-owned change.
- Read back the exact changed field or object.
- Save the affected scene or asset.
- Run the narrowest relevant check.
- Produce artifact evidence for visual or multi-step work.
Prefer stable globalObjectId values after saving a scene. Name targeting is acceptable for initial creation but is ambiguous once duplicates exist.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 96 lines · 79 tokens per session scan A f2a6916ebd86
unityctl-workflows is a skill published in the GitHub repository Jason-hub-star/unityctl (18 stars, last pushed 12d ago), licensed MIT. It adds 79 tokens to every session and 903 once invoked, about $0.0004 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-runtime-bug-hunt
Unity Editor PlayMode 中のランタイム挙動不具合・例外・応答不能の原因特定スキル。uloop execute-dynamic-code でランタイム状態を広く浅くダンプし、JetBrains Rider debugger で狭く深く変数/BPヒットを観測する。TRIGGER when: (1) ユーザーのメッセージに PlayMode 実行時のランタイム例外スタックトレースが貼られている（NullReferenceException / TimeoutException / IndexOutOfRangeException 等）(2)…
recipe-chain-machine-count
Use when the user wants a moorestech item produced at a target throughput — whether they ask "how many machines" explicitly OR just state a desired production rate (per minute or per second) without mentioning machine count. The rate may be phrased per second (個/秒・毎秒N個) — convert to per minute (×60) before computing.…
unity-playmode-recorded-playtest
Unity Editor を PlayMode 起動し、録画付きで end-to-end gameplay を検証する枠組み。第一選択はプレイテストDSL（Client.Playtest asmdef + 本スキル同梱の scripts/run-scenario.sh）による1コマンド一発実行で、preflight→PlayMode起動→シナリオ投入→result.json回収まで自動化される（実測ready26秒）。UI経路設置（ビルドメニュー→クリック/ドラッグ）とホットバー割当（建築ショートカット。歯車チェーンポール等）もDSLで操作可能。ユースケース別の詳細は references/…
PuerTS Agent Development Guide
Guide for developing LLM agents based on PuerTsAgent framework — covers resource directory structure, system-prompt, skills, builtin modules, and best practices.
editmode-in-playing-test
Skill "editmode-in-playing-test" from moorestech/moorestech, covering editmodeinplayingtest creator, 概要, 重要な前提, ワークフロー and 1. テストクラスを作成.
Install PuerTS for Unity
Guide for downloading and installing PuerTS UPM packages into a Unity project — covers version selection, package dependencies, download/extract and git URL installation methods, and Editor Assistant setup.