Skip to content
Back to search

claude-code-game-studios

Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/donchitos--claude-code-game-studios/badge.svg)](/listings/donchitos--claude-code-game-studios)

HTML

<a href="/listings/donchitos--claude-code-game-studios"><img src="/api/listings/donchitos--claude-code-game-studios/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Prerequisites

- [Git](https://git-scm.com/) - [Claude Code](https://code.claude.com/docs/en/setup) — the native installer is recommended (`npm install -g @anthropic-ai/claude-code` also works, on Node.js 22+) - [Python 3](https://www.python.org/) — reads `project.yaml` for every skill and hook; without it every setting silently falls back to its default - Bash — runs every hook; built into macOS and Linux, and on Windows it comes with [Git for Windows](https://git-scm.com/downloads/win) - **Recommended**: [jq](https://jqlang.github.io/jq/) (for hook validation)

Hooks fail gracefully if optional tools are missing — you just lose validation — with one deliberate exception: a commit that stages data JSON is blocked when there is no Python to validate it.

Setup

1. **Clone or use as template**:

   git clone https://github.com/Donchitos/Claude-Code-Game-Studios.git my-game
   cd my-game

2. **Open Claude Code** and start a session:

   claude

3. **Run `/start`** — the system asks where you are (no idea, vague concept, clear design, existing work) and guides you to the right workflow. No assumptions.

Or jump directly to a specific skill if you already know what you need: - `/brainstorm` — explore game ideas from scratch - `/setup-engine godot 4.6` — configure your engine if you already know - `/project-stage-detect` — analyze an existing project

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 527 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <p align="center">
  2. <h1 align="center">Claude Code Game Studios</h1>
  3. <p align="center">
  4. Describe a game. Claude Code asks a few questions, writes a one-page brief,
  5. and builds a playable browser game from it.
  6. <br />
  7. Then you change it, grow it and finish it, one step at a time.
  8. </p>
  9. </p>
  10. <p align="center">
  11. <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue.svg" alt="MIT License"></a>
  12. <a href="https://code.claude.com/docs"><img src="https://img.shields.io/badge/built%20for-Claude%20Code-f5f5f5?logo=anthropic" alt="Built for Claude Code"></a>
  13. <a href="https://www.buymeacoffee.com/donchitos3"><img src="https://img.shields.io/badge/Buy%20Me%20a%20Coffee-Support%20this%20project-FFDD00?logo=buymeacoffee&logoColor=black" alt="Buy Me a Coffee"></a>
  14. <a href="https://github.com/sponsors/Donchitos"><img src="https://img.shields.io/badge/GitHub%20Sponsors-Support%20this%20project-ea4aaa?logo=githubsponsors&logoColor=white" alt="GitHub Sponsors"></a>
  15. </p>
  16. ---
  17. ## Make me a game
  18. Open Claude Code in a copy of this repository and say what you want:
  19. ```
  20. /prototype a fishing game where the rare fish only bite at dusk
  21. ```
  22. 1. **A few questions.** One round, picked from the genre: the choices that
  23. shape the game, how it should feel, how it looks, how hard it is, the goal.
  24. Each has a usual answer, so "you pick" works. A vague idea ("a racing game")
  25. gets two or three one-line pitches to choose from first.
  26. 2. **A one-page brief.** `design/game-brief.md`: your words kept as you typed
  27. them, the look, the first minute, the challenge, the features and what is
  28. left for later. You read it and say Build it, or change something first.
  29. 3. **A first build.** A browser game in `prototypes/<name>/` that opens with a
  30. double-click: every figure drawn, no emoji art, no copy of another game, and
  31. not trivially easy. Your session builds it at its own effort ("build it at
  32. medium" is cheaper and thinner).
  33. 4. **Checked before you get it.** A few scripted key presses in a headless
  34. browser show it opens with no errors and responds; text that is too small,
  35. overlapping or off screen and screens that freeze are fixed in one round; and
  36. every line of the brief is marked built, missing or broken from the
  37. screenshots and the code, or not assessed when it could not be checked. Your
  38. play is the real test: the handover card says what passed, what is still broken
  39. and what was guessed, and ends "Not yet played by a person".
  40. Then it is yours to play. After that:
  41. - **Change it by asking.** Each change is checked against what you said to
  42. keep and to avoid, and each one you accept is a checkpoint you can go back
  43. to. Ask to tune it and it gains a tweak panel with a slider for every value.
  44. - **Share it.** A clean copy without the panel and the working files, ready
  45. for itch.io or a link.
  46. - **Grow it.** When you like it, it becomes a real web project (Vite and
  47. TypeScript, its values in data files, tests and a build), and you add to it
  48. one feature a sitting: say the next feature, and it is built, checked and
  49. committed. After every feature the same seeded run is played again and any
  50. change is shown to you. Stories (`/create-stories`, `/dev-story`,
  51. `/story-done`) only when you ask for them.
  52. - **Finish it.** `/cut` proposes what to cut and a polish list, and a date
  53. when the game is to sell.
  54. Already have a browser game? `/prototype --from <path>` brings a copy in as it
  55. is, so it can be changed and grown the same way.
  56. ## What it has been shown to do
  57. - **The first build.** In a blind comparison on three game ideas, the brief
  58. made from its questions was picked over a hand-written brief all three
  59. times. All six games built from the two kinds of brief were ones the tester
  60. said they would build on, so on the games themselves the two kinds of brief
  61. tied.
  62. - **Growing.** One game was grown from its first build through six stories
  63. into a game the tester would keep building. That is one game, not yet a
  64. pattern.
  65. - **Not yet shown:** the engine route (Godot, Unity, Unreal) and deepening one
  66. system with a design document. Both are built and pass the framework's own
  67. checks, but neither has had a test like the two above.
  68. The games themselves are deliberately **not** shipped here: this is a template
  69. you clone to start your own project, and somebody else's game in your repo is
  70. clutter, not evidence.
  71. ---
  72. ## Table of Contents
  73. - [Getting Started](#getting-started)
  74. - [Engines](#engines)
  75. - [Configuration](#configuration)
  76. - [Upgrading](#upgrading)
  77. - [Project Structure](#project-structure)
  78. - [Hooks, Permissions and Rules](#hooks-permissions-and-rules)
  79. - [Studio Mode](#studio-mode)
  80. - [Design Philosophy](#design-philosophy)
  81. - [Customization](#customization)
  82. - [Platform Support](#platform-support)
  83. - [Community](#community)
  84. - [Supporting This Project](#supporting-this-project)
  85. - [License](#license)
  86. ---
  87. ## Getting Started
  88. ### Prerequisites
  89. - [Git](https://git-scm.com/)
  90. - [Claude Code](https://code.claude.com/docs/en/setup) — the native installer is recommended (`npm install -g @anthropic-ai/claude-code` also works, on Node.js 22+)
  91. - [Python 3](https://www.python.org/) — reads `project.yaml` for every skill and hook; without it every setting silently falls back to its default
  92. - Bash — runs every hook; built into macOS and Linux, and on Windows it comes with [Git for Windows](https://git-scm.com/downloads/win). Without it on Windows the hooks skip themselves, and session start says their checks are off
  93. - **Chrome, Edge or Chromium** — `/prototype` plays and screenshots each game in it; without one, those checks say they were not assessed
  94. - **Recommended**: [Node.js](https://nodejs.org/) (to grow a game into a web project, for a 3D game's three.js, and for the web engine's build and tests) and [jq](https://jqlang.github.io/jq/) (for hook validation)
  95. Hooks fail gracefully if optional tools are missing — you just lose validation — with one deliberate exception: a commit that stages data JSON is blocked when there is no Python to validate it.
  96. ### Setup
  97. 1. **Clone or use as template**:
  98. ```bash
  99. git clone https://github.com/Donchitos/Claude-Code-Game-Studios.git my-game
  100. cd my-game
  101. ```
  102. 2. **Open Claude Code in the project's root folder** (the one holding
  103. `CLAUDE.md` and `project.yaml`) and start a session:
  104. ```bash
  105. claude
  106. ```
  107. Starting it in a subfolder such as `src/` skips the project's hooks, deny
  108. rules and ask-first permission mode.
  109. 3. **Type `/prototype` and your idea**, or **run `/start`**, which asks where
  110. you are (make me a game, an idea or none yet, clear design, existing work)
  111. and guides you to the right workflow. No assumptions.
  112. Other ways in:
  113. - `/next-step` — names the one next step wherever you are
  114. - `/brainstorm` — explore game ideas from scratch
  115. - `/setup-engine godot 4.7` — configure an engine if you already know you want one
  116. - `/project-stage-detect` — analyze an existing project
  117. ## Engines
  118. | Engine | `engine.name` | Code root | Notes |
  119. |--------|---------------|-----------|-------|
  120. | **Web** | `Web` | `src/` | The default for a new project. A one-shot needs no engine at all; a grown game is a web project. `/setup-engine web` offers a starter: nothing extra, three.js, Phaser or PixiJS |
  121. | **Godot 4** | `Godot` | `src/` | GDScript, C# or GDExtension |
  122. | **Unity** | `Unity` | `Assets/` | Unity compiles only `Assets/` and `Packages/` |
  123. | **Unreal Engine 5** | `Unreal` | `Source/<Module>/` | C++ and Blueprints |
  124. `/setup-engine` pins the engine version and writes its reference docs
  125. (`docs/engine-reference/<engine>/`). With no engine set, `/prototype` builds
  126. in the browser first; you pick an engine after you have played the game. Each
  127. engine's folders and default commands come from one table,
  128. `.claude/engines/engines.tsv`. If your code or tests live somewhere else, set
  129. `engine.code_root` or `engine.test_root` in `project.yaml`.
  130. ## Configuration
  131. `project.yaml` at the repo root is the single source of truth for how the
  132. studio behaves on your project — engine, naming conventions, and a `modes`
  133. block that controls how much process every skill enforces. `/start` writes it
  134. for you; you rarely need to hand-edit it.
  135. The one question that matters most is **`modes.rigor`** — `minimal`, `standard`,
  136. or `full`. It's a single front door over six separate knobs (`workflow` —
  137. which GDD sections are required, `docs.density`, `qa.level`,
  138. `story_granularity`, `review_mode` — how many director agents review your work,
  139. and `team.size`), so `/start` asks it once instead of six times:
  140. - **`minimal`** (default) — the light path. No GDDs required, terse docs, minimal QA evidence, solo review. `/prototype` (questions, a brief, a first build) → play and change it → grow it on PROCEED → one feature a sitting: name it, and it is built, checked and committed. Stories (`/create-stories` → `/dev-story` ↔ `/story-done`) only when you ask.
  141. - **`standard`** — 5 required GDD sections, balanced doc depth, standard QA evidence, lean review.
  142. - **`full`** — all 8 GDD sections, thorough docs, full QA evidence on every story type, full director review.
  143. `minimal` is the default because the heavier tier costs more to reach working
  144. code without producing a better result, and gives nothing back when someone new
  145. picks the project up. Raise it in one `/project-settings` call when your project grows
  146. — `/next-step` and `/gate-check` will suggest it when they see the signs.
  147. Two escape hatches keep one setting from being a blunt instrument.
  148. **`system_overrides`** holds a single system to a higher standard than the rest
  149. of the project — your combat system gets the full treatment while everything
  150. else stays light. **`testing.strict`** sets, per test type (logic, integration,
  151. visual, UI, config), whether missing evidence blocks a story or merely warns.
  152. Any of the six underlying knobs can still be set individually if you want a mix
  153. (e.g. `workflow: full` with `docs.density: terse` — comprehensive but compact).
  154. Two of them — `review_mode` (`full` / `lean` / `solo`) and `team.size` — are
  155. also **personal**: rigor sets the team-wide default, but you can override them
  156. per-developer in a gitignored `project.local.yaml` without touching the shared
  157. config. `modes.automation` (how often the AI pauses to ask before it acts) sits
  158. outside rigor and is likewise personally overridable. `project.goal` (a jam
  159. game, a hobby, a game to sell, and so on) is what you want from the game;
  160. `/start` guesses it, and `/cut` reads it.
  161. View or change any setting with the `/project-settings` skill:
  162. ```
  163. /project-settings # view effective config
  164. /project-settings modes.review_mode # view one setting
  165. /project-settings modes.rigor=full # write to project.yaml (team-wide)
  166. /project-settings --local modes.review_mode=solo # write to project.local.yaml (just you)
  167. ```
  168. Full schema and per-setting behavior: `.claude/docs/effects-map.md`.
  169. ## Upgrading
  170. Already using an older version of this template? See [UPGRADING.md](UPGRADING.md)
  171. for step-by-step migration instructions, a breakdown of what changed between
  172. versions, and which files are safe to overwrite vs. which need a manual merge.
  173. ## Project Structure
  174. ```
  175. CLAUDE.md # Master configuration
  176. project.yaml # Project settings: engine, modes, goal
  177. .claude/
  178. settings.json # Hooks, permissions, safety rules
  179. agents/ # 49 agent definitions (markdown + YAML frontmatter)
  180. skills/ # 75 skills, plus three old names (a subdirectory each)
  181. hooks/ # 18 scripts (bash) — 15 event hooks, two helpers they
  182. # source (yaml-helper.sh, update-check.sh), and one
  183. # opt-in diagnostic you wire yourself
  184. scripts/ # Helper scripts the skills and hooks run; each .sh has --help
  185. rules/ # 14 path-scoped coding standards
  186. genres/ # Genre sheets and look guidance /prototype asks and
  187. # builds from
  188. engines/ # The engine table (engines.tsv) and the web starter
  189. statusline.sh # Status line script (context%, model, stage, epic breadcrumb)
  190. docs/
  191. workflow-catalog.yaml # 7-phase pipeline definition (read by /next-step)
  192. templates/ # 62 document templates (+ per-section guidance)
  193. prototypes/ # One-shots, one folder each (isolated from src/)
  194. references/ # Optional: images and sounds you like, for /prototype
  195. src/ # Game source code (Godot, Web). The code root is
  196. # ENGINE-SPECIFIC: Unity compiles only
  197. # Assets/, Unreal builds from Source/<Module>/
  198. assets/ # Art, audio, VFX, shaders, data files
  199. design/ # The game brief, GDDs, narrative docs, level designs
  200. docs/ # Technical documentation, ADRs, engine references
  201. tests/ # Test suites (unit, integration, performance, playtest)
  202. tools/ # Build and pipeline tools
  203. production/ # Sprint plans, milestones, release tracking
  204. studio-tests/ # QA for the skills and agents themselves — see below
  205. ```
  206. ### Testing your customizations
  207. Claude Code Game Studios is a template you are meant to edit. `studio-tests/` is how
  208. you check that an edited or newly written skill still holds up — a catalog of all
  209. 75 skills and 49 agents, per-category quality rubrics, behavioral specs, and
  210. templates for writing specs of your own.
  211. ```
  212. /skill-test static [name] # structural linter — 7 compliance checks
  213. /skill-test category [name] # category-specific quality rubric
  214. /skill-test spec [name] # behavioral spec assertions
  215. /skill-test audit # coverage report across skills and agents
  216. /skill-improve [name] # test-fix-retest loop; keeps or reverts on score change
  217. ```
  218. It tests the *framework*, never your game — game tests live in `tests/`.
  219. ## Hooks, Permissions and Rules
  220. **Hooks** run automatically on every session:
  221. | Hook | Trigger | What It Does |
  222. |------|---------|--------------|
  223. | `validate-commit.sh` | PreToolUse (Bash, PowerShell) | Checks for hardcoded values, TODO format, JSON validity, design doc sections — exits early if the command is not `git commit` |
  224. | `validate-push.sh` | PreToolUse (Bash, PowerShell) | Warns on pushes to protected branches, and on `git tag`, `git merge`, `gh release`, `butler push` and `steamcmd +run_app_build` — exits early for any other command |
  225. | `validate-read.sh` | PreToolUse (Read) | Refuses a whole-file read of a design doc over 40 KB and answers with its headings, so the retry reads one section (`features.read_guard`) |
  226. | `validate-write.sh` | PreToolUse (Write/Edit) | Warns, never blocks, when a write would break a rule the skills state: a file they never overwrite, the config guard, the prototype boundary, a localization freeze |
  227. | `validate-registry.sh` | PreToolUse (Write/Edit) | Warns when a write to the entity or TR-ID registry would delete or renumber an entry, or set a status its header does not list |
  228. | `validate-assets.sh` | PostToolUse (Write/Edit) | Checks data-file JSON under `assets/`, `Assets/` or `Content/`, and asset naming in Godot's `assets/` only — exits early if the file is under none of them |
  229. | `session-start.sh` | Session open | Shows current branch and recent commits for orientation |
  230. | `detect-gaps.sh` | Session open | Detects fresh projects (suggests `/start`) and missing design docs when code or prototypes exist |
  231. | `pre-compact.sh` | Before compaction | Logs the compaction (its output does not reach Claude — `session-start.sh` restores context afterwards) |
  232. | `notify.sh` | Permission prompt, or idle about a minute | Shows Windows toast notification via PowerShell |
  233. | `session-stop.sh` | Every response end, in the background | Archives `active.md` to session log and records git activity |
  234. | `log-agent.sh` | Agent spawned | Audit trail start — logs subagent invocation |
  235. | `log-agent-stop.sh` | Agent stops | Audit trail stop — completes subagent record |
  236. | `validate-agent-artifact.sh` | Agent stops | Warns when a subagent stops without writing the file its brief named |
  237. | `validate-skill-change.sh` | PostToolUse (Write/Edit) | Advises running `/skill-test` after any `.claude/skills/` change |
  238. > **Note**: `validate-commit.sh`, `validate-push.sh`, `validate-read.sh`, `validate-write.sh`, `validate-registry.sh`, `validate-assets.sh` and `validate-skill-change.sh` fire on every Bash, PowerShell, Read, Write or Edit tool call and exit immediately (exit 0) when the command or file path is not relevant. This is normal hook behavior — not a performance concern.
  239. **Permission rules** in `settings.json` auto-allow test runs and the framework's read-only helper scripts, and block dangerous operations (force push, `rm -rf`, reading `.env` files) in both the Bash and PowerShell tools. Read-only commands such as `git status`, `git log` and `ls` need no rule: Claude Code runs them without asking. The `git` rules match anywhere after `git`, so a git command whose text only mentions `push` followed later by ` +` or ` -f`, `clean` followed by ` -f`, or `reset` followed by `--hard` — a commit message such as `git commit -m "push + pull"` included — is denied too: reword the message.
  240. ### Path-Scoped Rules
  241. Coding standards are automatically enforced based on file location. The paths
  242. below are Godot's. Each rule also matches the Unity and Unreal layouts: the code
  243. rules match C# under `Assets/**/Gameplay/` and C++ under `Source/**/Gameplay/` (and `Core`,
  244. `AI`, `Networking`, `UI`), data matches JSON under `Assets/**/Data/` and
  245. `Content/**/Data/`, shaders match by file type, and tests match
  246. C# under `Assets/Tests/` and C++ under `Source/**/Tests/`.
  247. | Path | Enforces |
  248. |------|----------|
  249. | `src/gameplay/**` | Data-driven values, delta time usage, no UI references |
  250. | `src/core/**` | Zero allocations in hot paths, thread safety, API stability |
  251. | `src/ai/**` | Performance budgets, debuggability, data-driven parameters |
  252. | `src/networking/**` | Server-authoritative, versioned messages, security |
  253. | `src/ui/**` | No game state ownership, localization-ready, accessibility |
  254. | `design/gdd/**` | Required sections per `modes.workflow`, formula format, edge cases |
  255. | `design/narrative/**` | Lore consistency, character voice, canon levels |
  256. | `assets/data/**/*.json` | JSON validity, naming conventions, schema rules |
  257. | `assets/shaders/**` | Naming conventions, performance targets, cross-platform rules |
  258. | `**/*.tscn`, `**/*.tres` | Scene and resource files changed in the editor where possible; safe hand-edit rules |
  259. | `tests/**` | Test naming, coverage requirements, fixture patterns |
  260. | `prototypes/**` | A one-shot's tuning values in `config.js` and seeded randomness; relaxed standards for engine and paper prototypes |
  261. ## Studio Mode
  262. Everything above is the light path: one brief, a game you play, one feature a
  263. sitting as it grows. For a team, a long project, or a person who wants the paperwork, the
  264. same project also has **studio mode**: 49 specialized agents in a studio
  265. hierarchy, a design document per system, architecture decisions, epics and
  266. sprints, phase gates, and `team-*` pipelines that coordinate several agents on
  267. one feature. `/next-step` suggests it once when a grown game gets big (about
  268. nine stories or features).
  269. ### What's Included
  270. | Category | Count | Description |
  271. |----------|-------|-------------|
  272. | **Agents** | 49 | Specialized subagents across design, programming, art, audio, narrative, QA, and production |
  273. | **Skills** | 75 | Slash commands for every workflow phase (`/prototype`, `/start`, `/design-system`, `/create-epics`, `/create-stories`, `/dev-story`, `/story-done`, etc.) |
  274. | **Hooks** | 15 | Automated validation on commits, pushes, asset changes, design-doc reads, writes and registries, session lifecycle, agent audit trail, and gap detection |
  275. | **Rules** | 14 | Path-scoped coding standards enforced when editing gameplay, engine, AI, UI, network code, scene and asset files, and more |
  276. | **Templates** | 62 | Document templates for GDDs, UX specs, ADRs, sprint plans, HUD design, accessibility, and more |
  277. ### Studio Hierarchy
  278. Agents are organized into three tiers, matching how real studios operate:
  279. ```
  280. Tier 1 — Directors
  281. creative-director technical-director producer
  282. Tier 2 — Department Leads
  283. game-designer lead-programmer art-director
  284. audio-director narrative-director qa-lead
  285. release-manager localization-lead
  286. Tier 3 — Specialists
  287. gameplay-programmer engine-programmer ai-programmer
  288. network-programmer tools-programmer ui-programmer
  289. systems-designer level-designer economy-designer
  290. technical-artist sound-designer writer
  291. world-builder ux-designer prototyper
  292. performance-analyst devops-engineer analytics-engineer
  293. security-engineer qa-tester accessibility-specialist
  294. live-ops-designer community-manager
  295. ```
  296. Each agent declares a model in its frontmatter: the three directors ask for
  297. Opus, `lead-programmer` and fifteen specialists name Sonnet, and the remaining thirty
  298. **inherit your session's model**. Treat the tiers as seniority of role, not as
  299. a billing plan — most of the studio runs on whatever model you are already using.
  300. ### Engine Specialists
  301. The template includes agent sets for the three major engines. Use the set that
  302. matches your project (a web project uses the general programmers):
  303. | Engine | Lead Agent | Sub-Specialists |
  304. |--------|-----------|-----------------|
  305. | **Godot 4** | `godot-specialist` | GDScript, Shaders, GDExtension |
  306. | **Unity** | `unity-specialist` | DOTS/ECS, Shaders/VFX, Addressables, UI Toolkit |
  307. | **Unreal Engine 5** | `unreal-specialist` | GAS, Blueprints, Replication, UMG/CommonUI |
  308. ### Slash Commands
  309. Type `/` in Claude Code to access all 75 skills (plus three old names that point to their new ones):
  310. **Make a Game**
  311. `/prototype` `/cut`
  312. **Onboarding & Navigation**
  313. `/start` `/next-step` `/project-stage-detect` `/setup-engine` `/adopt` `/project-settings`
  314. **Game Design**
  315. `/brainstorm` `/map-systems` `/design-system` `/quick-design` `/review-all-gdds` `/propagate-design-change`
  316. **Art & Assets**
  317. `/art-bible` `/asset-spec` `/asset-audit`
  318. **UX & Interface Design**
  319. `/ux-design` `/ux-review`
  320. **Architecture**
  321. `/create-architecture` `/architecture-decision` `/architecture-review` `/create-control-manifest`
  322. **Stories & Sprints**
  323. `/create-epics` `/create-stories` `/dev-story` `/sprint-plan` `/sprint-status` `/story-readiness` `/story-done` `/estimate`
  324. **Reviews & Analysis**
  325. `/design-review` `/code-review` `/balance-check` `/content-audit` `/scope-check` `/perf-profile` `/tech-debt` `/gate-check` `/consistency-check` `/security-audit`
  326. **QA & Testing**
  327. `/qa-plan` `/smoke-check` `/soak-test` `/regression-suite` `/test-setup` `/test-helpers` `/test-evidence-review` `/test-flakiness` `/skill-test` `/skill-improve`
  328. **Production**
  329. `/milestone-review` `/retrospective` `/bug-report` `/bug-triage` `/reverse-document` `/playtest-report`
  330. **Release**
  331. `/release-checklist` `/launch-checklist` `/write-changelog` `/patch-notes` `/hotfix` `/day-one-patch`
  332. **Creative & Content**
  333. `/vertical-slice` `/onboard` `/localize`
  334. **Team Orchestration** (coordinate multiple agents on a single feature)
  335. `/team-combat` `/team-narrative` `/team-ui` `/team-release` `/team-polish` `/team-audio` `/team-level` `/team-live-ops` `/team-qa`
  336. ### Does studio mode produce games, or documents?
  337. A fair question, and the answer is a measurement. Games were built from one
  338. brief at different `modes.rigor` levels, and the documents each wrote
  339. **before its first line of game code** were counted. The lightest and the
  340. heaviest:
  341. | `modes.rigor` | Docs before first line of game code | Time |
  342. |---|---:|---:|
  343. | `minimal` | **1** (the one-page brief) | 0 min |
  344. | `standard` | 30 | 58 min |
  345. All of them produced a running, tested game. Two blind reviewers and a human
  346. playtest then ranked the `standard` build **last** — the extra 29 documents
  347. bought traceability, not a better game. That is why the light path is the
  348. default.
  349. ### It runs the game and looks at it
  350. A passing test says nothing about a menu drawn off-screen, so a story that
  351. changes anything the player sees is not closed until the game has been
  352. launched, observed, and a screenshot retained in `production/qa/evidence/`. A
  353. parse check is not a run. This holds at every rigor level, including
  354. `minimal`, where automated tests are waived but the look is not.
  355. ### Agent Coordination
  356. Agents follow a structured delegation model:
  357. 1. **Vertical delegation** — directors delegate to leads, leads delegate to specialists
  358. 2. **Horizontal consultation** — same-tier agents can consult each other but can't make binding cross-domain decisions
  359. 3. **Conflict resolution** — disagreements escalate up to the shared parent (`creative-director` for design, `technical-director` for technical)
  360. 4. **Change propagation** — cross-department changes are coordinated by `producer`
  361. 5. **Domain boundaries** — agents don't modify files outside their domain without explicit delegation
  362. ### Collaborative, Not Autonomous
  363. This is **not** an auto-pilot system. Every agent follows a strict collaboration protocol:
  364. 1. **Ask** — agents ask questions before proposing solutions
  365. 2. **Present options** — agents show 2-4 options with pros/cons
  366. 3. **You decide** — the user always makes the call
  367. 4. **Draft** — agents show work before finalizing
  368. 5. **Approve** — nothing gets written without your sign-off
  369. You stay in control. The agents provide structure and expertise, not autonomy.
  370. That is how a terminal session started at the project root behaves out of the
  371. box. The `guided` and `autonomous` automation settings, and Claude Code's own
  372. auto mode, trade some of these prompts for speed; see
  373. [setup-requirements.md](.claude/docs/setup-requirements.md) for the routes into
  374. auto mode.
  375. ## Design Philosophy
  376. This template is grounded in professional game development practices:
  377. - **Playable first** — a game you can play before documents about it
  378. - **MDA Framework** — Mechanics, Dynamics, Aesthetics analysis for game design
  379. - **Self-Determination Theory** — Autonomy, Competence, Relatedness for player motivation
  380. - **Flow State Design** — Challenge-skill balance for player engagement
  381. - **Bartle Player Types** — Audience targeting and validation
  382. - **Verification-Driven Development** — Tests first, then implementation
  383. ## Customization
  384. This is a **template**, not a locked framework. Everything is meant to be customized:
  385. - **Add/remove agents** — delete agent files you don't need, add new ones for your domains
  386. - **Edit agent prompts** — tune agent behavior, add project-specific knowledge
  387. - **Modify skills** — adjust workflows to match your team's process
  388. - **Add rules** — create new path-scoped rules for your project's directory structure
  389. - **Tune hooks** — adjust validation strictness, add new checks
  390. - **Pick your engine** — web, Godot, Unity or Unreal (see [Engines](#engines)); use the matching agent set, or none
  391. - **Tune process weight** — `modes.rigor` and its underlying knobs in `project.yaml` (see [Configuration](#configuration)). Override per-run with `--review solo` on any skill that runs a review gate.
  392. ## Platform Support
  393. Primary development on **Windows 10** with Git Bash. The hooks and scripts are also tested on **macOS** — its stock bash 3.2 and BSD tools, with and without `jq` — and on **Linux**. All hooks use POSIX-compatible patterns (`grep -E`, not `grep -P`) and include fallbacks for missing tools. The `notify.sh` hook uses PowerShell for Windows toast notifications and is a no-op elsewhere — desktop notifications on macOS/Linux are not yet wired. Please file issues for any platform-specific breakage.
  394. ## Community
  395. - **Discussions** — [GitHub Discussions](https://github.com/Donchitos/Claude-Code-Game-Studios/discussions) for questions, ideas, and showcasing what you've built
  396. - **Issues** — [Bug reports and feature requests](https://github.com/Donchitos/Claude-Code-Game-Studios/issues)
  397. ---
  398. ## Supporting This Project
  399. Claude Code Game Studios is free and open source. If it saves you time or helps you ship your game, consider supporting continued development:
  400. <p>
  401. <a href="https://www.buymeacoffee.com/donchitos3"><img src="https://img.shields.io/badge/Buy%20Me%20a%20Coffee-FFDD00?style=for-the-badge&logo=buy-me-a-coffee&logoColor=black" alt="Buy Me a Coffee"></a>
  402. &nbsp;
  403. <a href="https://github.com/sponsors/Donchitos"><img src="https://img.shields.io/badge/GitHub%20Sponsors-ea4aaa?style=for-the-badge&logo=githubsponsors&logoColor=white" alt="GitHub Sponsors"></a>
  404. </p>
  405. - **[Buy Me a Coffee](https://www.buymeacoffee.com/donchitos3)** — one-time support
  406. - **[GitHub Sponsors](https://github.com/sponsors/Donchitos)** — recurring support through GitHub
  407. Sponsorships help fund time spent maintaining skills, adding new agents, keeping up with Claude Code and engine API changes, and responding to community issues.
  408. ---
  409. *Built for Claude Code. Maintained and extended — contributions welcome via [GitHub Discussions](https://github.com/Donchitos/Claude-Code-Game-Studios/discussions).*
  410. ## License
  411. MIT License. See [LICENSE](LICENSE) for details.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.