scene-architecture

scene-architecture is a skill for Claude Code from TheArcForge/Hades. It costs 31 tokens per session (5,078 once invoked), scanned A, original, MIT.

A Unity guide for organizing game scenes and moving between them. It covers single scenes, additive loading (combining scenes), startup scenes, persistent services, and streamed worlds.

In plain words
What is it for?
Use it to choose a scene structure, place persistent managers, design loading and transition flows, decide when to use DontDestroyOnLoad, or plan an open-world scene setup.
Why use it?
It helps prevent unclear ownership and fragile transitions when game content or shared services span multiple scenes.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the hades plugin — 22 skills, 6 commands shipped together

Good fit Use it to choose a scene structure, place persistent managers, design loading…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/thearcforge/hades/scene-architecture
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.

Any agent
npx skills add TheArcForge/Hades --skill scene-architecture
Clone the repo
git clone --depth 1 https://github.com/TheArcForge/Hades

Made for: Claude Code.

Or install hades, the plugin that ships this one along with the rest of its 22 skills, 6 commands.

Wrote 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.

agentmods badge for scene-architecture

README.md
[![agentmods](https://agentmods.dev/badge/skills/thearcforge/hades/scene-architecture.svg)](https://agentmods.dev/skills/thearcforge/hades/scene-architecture)
Your own site
<a href="https://agentmods.dev/skills/thearcforge/hades/scene-architecture"><img src="https://agentmods.dev/badge/skills/thearcforge/hades/scene-architecture.svg" alt="Measured on agentmods" height="20"></a>
Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,078 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00031 $0.05078
Opus 5 $0.00015 $0.02539
Sonnet 5 $0.00006 $0.01016
Haiku 4.5 $0.00003 $0.00508

Measured 6d ago against content hash a706b6fac3fe, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

scene-architecture 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 6d 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.

ClaudeCodePlugin/skills/scene-architecture/SKILL.md · 641 lines

How it starts

The opening of the file, as written. The whole thing — 641 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Scene Architecture

Provides decision frameworks and implementation patterns for structuring Unity scenes — from small single-scene games to large streaming worlds with persistent services. Covers bootstrap setup, scene transitions, additive loading, and persistent manager strategies.

When to Apply

Activate when the conversation involves:

  • Choosing between a monolithic scene, per-state scenes, or additive loading
  • Deciding where persistent managers (audio, save, analytics) should live
  • Designing scene transitions — hard cut, fade, or loading screen with progress
  • Deciding whether to use DontDestroyOnLoad vs a dedicated bootstrap scene
  • Setting up a streaming open-world or hub-and-spoke scene graph
  • Questions about who owns a scene transition and how to trigger it
  • Scene dependency: one scene needs data from a previously loaded scene

Do NOT activate for narrow scene-authoring tasks (placing objects, lighting, baking) — those go to hades:scene-authoring.

Project Context Check

Before making recommendations, gather current project state:

  1. Check existing scene structure:

    • Call get_project_summary() to see scene count, project scale, render pipeline, and asset volumes — a 3-scene mobile game warrants different advice than a 40-scene open-world PC title
    • Call get_scene_summary("<main_scene>") to understand what is already in the main or boot scene (existing managers, lighting, canvas hierarchy)
    • Call search_by_name("*Scene*") to discover scene-related assets and scripts (loaders, transition controllers, scene reference SOs)
  2. Check team decisions in memory:

    • Call recall_memory("scene loading architecture") to find documented scene strategy
    • Call recall_memory("persistent manager bootstrap") to find existing manager lifetime decisions
    • If a recalled decision shows warning, surface the conflict to the user before proceeding
  3. Adapt based on findings:

    • If graph shows an existing BootstrapLoader or SceneTransitionManager → align with it rather than introducing a parallel system
    • If no loading infrastructure exists → introduce from scratch and prompt the user to record the decision
    • If memory documents a scene strategy → follow it unless there is a strong technical reason to deviate

Read the full file on GitHub · 641 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. 6d ago First seen · 641 lines · 31 tokens per session scan A a706b6fac3fe

Subscribe to this mod's changes

scene-architecture is a skill published in the GitHub repository TheArcForge/Hades (27 stars, last pushed yesterday), licensed MIT. It adds 31 tokens to every session and 5,078 once invoked, about $0.0002 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.