unity-cli-loop AGENTS.md

Project instructions for uloop, a command-line tool that communicates with the Unity game editor over a network connection. They describe the project structure, generated files, and rules for avoiding editor freezes.

In plain words
What is it for?
Use them when developing uloop, regenerating its instruction files, running compile or log commands, and adding or running Unity tests.
Why use it?
They reduce mistakes when changing the project, especially editing generated files or running Unity tests unsafely.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/hatayama/unity-cli-loop/agents-md
Clone the repo
git clone --depth 1 https://github.com/hatayama/unity-cli-loop

Made for: Codex, OpenCode.

Per session 3,014 This file is loaded in full into every session.
When invoked 3,014 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.03014 $0.03014
Opus 5 $0.01507 $0.01507
Sonnet 5 $0.00603 $0.00603
Haiku 4.5 $0.00301 $0.00301

Measured yesterday against content hash f8f0ddcb1ef8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

unity-cli-loop 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 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.

AGENTS.md · 201 lines

How it starts

The opening of the file, as written. The whole thing — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Architecture Overview

This project provides a CLI tool (uloop) that communicates with Unity Editor over local IPC (a Unix domain socket on macOS/Linux, a named pipe on Windows — not TCP). AI agents interact with Unity through uloop CLI commands (e.g., uloop get-logs, uloop compile). The Unity Editor side hosts a local project IPC server that accepts short-lived CLI command sessions.

Repository Map

Directory-level responsibilities. Kept deliberately coarse — check the directory itself for file-level detail.

  • Packages/src/ — the Unity package (C#). Each first-party tool lives in Editor/FirstPartyTools/<Tool>/ with its implementation and agent skill (Skill/SKILL.md, optional Skill/references/).
  • Packages/src/Editor/CliOnlyTools~/<Tool>/Skill/ — skills for CLI-only commands (launch, pause-point, etc.) that have no Unity tool class. Tilde-suffixed folders are ignored by Unity, so files there need no .meta; files under FirstPartyTools do (run uloop compile to let Unity generate them).
  • Assets/ — the development/test Unity project, including custom-command samples under Assets/Editor/CustomCommandSamples/.
  • Assets/RegressionHarness/<TrapName>/ — permanent manual repro scenes for verification-round "traps" (pause-point, simulate-keyboard, physics callbacks, etc.), paired with a driver script under scripts/regression-harness-<trap-name>.sh. See docs/regression-harness.md.
  • cli/dispatcher/ — the globally installed uloop entry command (also owns uloop skills install and skill syncing).
  • cli/project-runner/ — the per-project CLI runner that talks to the Unity-side IPC server.
  • cli/common/ — Go modules shared by dispatcher and project runner. Tool parameter schemas live in cli/common/tools/default-tools.json; skill discovery in cli/common/skillscan/.
  • cli/release-automation/ — Go logic backing GitHub Actions release/CI workflows.
  • tools/UnityCliLoop.DeadCodeScanner/ — the C# dead-code scanner described below.
  • docs/ — reference docs for the rules in this file. docs/adr/ holds architecture decision records: read the relevant one before reopening a settled design question, and note its stated reversal condition rather than relitigating from scratch.
  • dist/ — locally built development binaries (dist/darwin-arm64/uloop, etc.); never committed.
  • .claude/, .agents/ — generated skill copies; never edit directly (see Generated Skill Files).
  • .uloop/ — runtime state and command outputs (screenshots, test results, hierarchy dumps).

Read the full file on GitHub · 201 lines

Changes

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.

  1. yesterday Changed · +161 lines · +2,255 tokens per session f8f0ddcb1ef8
  2. yesterday First seen · 40 lines · 759 tokens per session scan A d6e3e1118dfd

Subscribe to this mod's changes

unity-cli-loop AGENTS.md is an instructions file published in the GitHub repository hatayama/unity-cli-loop (522 stars, last pushed yesterday), licensed MIT. It adds 3,014 tokens to every session, about $0.0151 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.

Related

Other instructions, from other repositories

core CLAUDE.md

Instructions for ReactUnity/core, covering claude.md, what this repo is, two package universes, pnpm workspace membership and commands.

ReactUnity/core · 6,014 tokens

core AGENTS.md

Instructions for ReactUnity/core, covering release workflow, write changelog files, example, fix button hover state and package references.

ReactUnity/core · 245 tokens

unity-mesh-fracture AGENTS.md

Instructions for sinanata/unity-mesh-fracture, covering agents.md, what this project is, golden rules (do not violate), wire it into your project and public api.

sinanata/unity-mesh-fracture · 2,647 tokens

Aspid.FastTools CLAUDE.md

Instructions for VPDPersonal/Aspid.FastTools, covering обзор репозитория, сборка, архитектура, карта функциональности and соглашения по коду редактора.

VPDPersonal/Aspid.FastTools · 2,948 tokens

unity-cli CLAUDE.md

Instructions for bigdra50/unity-cli, covering claude.md, project overview, commands, install globally (provides: unity-cli, u, unity commands) and run relay server standalone.

bigdra50/unity-cli · 985 tokens

unity-mesh-fracture copilot-instructions.md

Instructions for sinanata/unity-mesh-fracture: This repository follows the cross-tool AGENTS.md standard. Read /AGENTS.md for the full guide: the MeshFracture API, the bake-at-load / look-up-at-impact pipeline, the material rules, build and preview commands, and the pull-request checklist.

sinanata/unity-mesh-fracture · 346 tokens