UnitySkills is a REST API-based automation engine that lets AI coding agents control Unity Editor projects, including scenes and other editor operations. It is for developers using Unity with supported coding agents and IDEs. The catalogue skills and commands provide the agent workflows used to operate Unity projects through the engine.
Borrowing it
Nothing to install: this file belongs to Besty0728/Unity-Skills. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Besty0728/Unity-Skills/main/AGENTS.mdgit clone --depth 1 https://github.com/Besty0728/Unity-SkillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/besty0728/unity-skills/agents-md)<a href="https://agentmods.dev/instructions/besty0728/unity-skills/agents-md"><img src="https://agentmods.dev/badge/instructions/besty0728/unity-skills/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/besty0728/unity-skills/agents-md"><img src="https://agentmods.dev/badge/instructions/besty0728/unity-skills/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.03374 | $0.03374 |
| Opus 5.5 | $0.01350 | $0.01350 |
| Sonnet 5.5 | $0.00675 | $0.00675 |
| Haiku 4.5 | $0.00337 | $0.00337 |
Grade A, and why
Unity-Skills AGENTS.md 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 21d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
UnitySkills — instructions for AI agents developing this repo
Audience: agents editing this repository. Agents calling the REST API read SkillsForUnity/unity-skills~/SKILL.md instead — that file is the shipped protocol doc (8,192-byte hard cap; new content goes under references/) and the authoritative description of operating modes, permission endpoints, batch/diff semantics and observability. Read it before touching anything the client sees.
| Field | Value |
|---|---|
| Version | 2.9.1 |
| Stack | C# Unity Editor plugin (UPM com.besty.unity-skills) + Python client |
| Unity | 2022.3+, verified on 6000.x |
| License | MIT |
Architecture
AI agent → unity_skills.py → HTTP localhost:8090-8100 → SkillsHttpServer (producer-consumer) → SkillRouter (reflects [UnitySkill]) → 56 *Skills.cs / 54 SkillCategory / 839 skills. WorkflowManager = persistent undo/rollback, RegistryService = multi-instance discovery. 56 vs 54: BatchSkills.cs registers into Workflow + Validation, DiagnoseSkills.cs into Debug. The per-category table lives in README.md; /skillcheck keeps every count in sync — do not hand-edit counts.
- Threading (hard rule): the HTTP thread only enqueues; the Unity main thread drains the queue from
EditorApplication.update. ZeroUnityEngine.*/UnityEditor.*calls off the main thread. - Every request, including GET
/health/jobs/{id}/skills, goes through that main-thread queue (≤20 per tick). A skill that blocks the main thread also freezes liveness; the Python client pollsGET /jobs/{id}instead ofjob_waitfor this reason. - No auth + wildcard CORS is intentional (loopback bind is the accepted boundary). Do not report "add auth / tighten CORS / validate Origin" as a security finding.
- Optional-package modules (ProBuilder, XR, Netcode, YooAsset, DOTween, PrimeTween, Behavior, HybridCLR, Addressables, QFramework) detect their dependency and return
MISSING_PACKAGE; URP-family modules (Volume/PostProcess/Decal/URP) compile to same-namedNoURP()stubs withoutcom.unity.render-pipelines.universal. QFramework has no UPM package, so detection is by reflected anchor type and it must not declareRequiresPackages. - Optional-package behavior is tested for real in
Tests/Editor/OptionalPackages/: one fixture per module onOptionalPackageTestBase(skills called by name, state read back by reflection, fixture ignored when the probe skill's package is missing), with per-module floors of passed tests in.github/optional-package-floors.json; a new optional module adds a fixture and a floor. Package code that needs a built-in module (NGO'sNetworkRigidbody2Dneeds physics2d) mirrors the package'sCOM_UNITY_MODULES_*versionDefines, or a project without that module fails to compile the whole package. - 29 advisory modules under
unity-skills~/skills/are documentation only: no REST skills, no C# stub, ever.
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.
- 21d ago Changed a43edc5bef6b
- 1mo ago Changed · +4 lines · +482 tokens per session 05475d64a824
- 1mo ago First seen · 93 lines · 2,892 tokens per session scan A 2689fde7ddc6
Unity-Skills AGENTS.md is an instructions file published in the GitHub repository Besty0728/Unity-Skills (1,923 stars, last pushed yesterday), licensed MIT. It adds 3,374 tokens to every session, about $0.0135 per session on Opus 5.5. 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 2026-09-20.
Unity-Skills AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| Unity-Skills AGENTS.mdBesty0728/Unity-Skills | Instructions file | 1.9k | yesterday | MIT | A | 3,374 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |