claude-code-game-studios

What this skill does

```markdown
---
name: claude-code-game-studios
description: Turn Claude Code into a full 49-agent game dev studio with 72 workflow skills, automated hooks, and a real studio hierarchy for Godot, Unity, and Unreal projects.
triggers:
  - "set up claude code game studios"
  - "use ai agents for game development"
  - "set up game dev studio with claude"
  - "add game studio agents to my project"
  - "how do I use claude code for game dev"
  - "set up godot unity unreal ai workflow"
  - "49 agents game studio claude"
  - "game development ai coordination system"
---

# Claude Code Game Studios

> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.

Claude Code Game Studios transforms a single Claude Code session into a structured game development studio with 49 specialized AI agents, 72 workflow slash commands, 12 automated safety hooks, and 11 path-scoped coding rules — mirroring a real studio hierarchy across design, programming, art, audio, narrative, QA, and production.

---

## Installation

### New Project

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

Then type `/start` inside Claude Code.

### Add to an Existing Project

```bash
# Copy only the studio infrastructure (no src/ or assets/ overwrite)
git clone https://github.com/Donchitos/Claude-Code-Game-Studios.git _studio_tmp
cp -r _studio_tmp/.claude ./
cp _studio_tmp/CLAUDE.md ./CLAUDE.md
rm -rf _studio_tmp

claude
# Then run:
/adopt        # Detects existing code and maps it into the studio structure
```

### Prerequisites

```bash
# Required
npm install -g @anthropic-ai/claude-code

# Recommended (hooks fail gracefully without these)
brew install jq          # macOS
sudo apt install jq      # Linux
# Python 3 — usually pre-installed
```

---

## Studio Hierarchy

Agents are tiered by responsibility and model:

```
Tier 1 — Directors (Opus)
  creative-director · technical-director · producer

Tier 2 — Department Leads (Sonnet)
  game-designer · lead-programmer · art-director
  audio-director · narrative-director · qa-lead
  release-manager · localization-lead

Tier 3 — Specialists (Sonnet/Haiku)
  gameplay-programmer · engine-programmer · ai-programmer
  network-programmer · tools-programmer · ui-programmer
  systems-designer · level-designer · economy-designer
  technical-artist · sound-designer · writer
  world-builder · ux-designer · prototyper
  performance-analyst · devops-engineer · analytics-engineer
  security-engineer · qa-tester · accessibility-specialist
  live-ops-designer · community-manager
```

### Engine Specialist Sets

| Engine | Lead Agent | Sub-Specialists |
|---|---|---|
| Godot 4 | `godot-specialist` | GDScript, Shaders, GDExtension |
| Unity | `unity-specialist` | DOTS/ECS, Shaders/VFX, Addressables, UI Toolkit |
| Unreal Engine 5 | `unreal-specialist` | GAS, Blueprints, Replication, UMG/CommonUI |

---

## Key Slash Commands

Type `/` in Claude Code to access all 72 skills. The most important ones:

### Onboarding

```
/start                    # Guided entry — asks where you are, routes you correctly
/help                     # Shows 7-phase pipeline and available skills
/project-stage-detect     # Analyzes an existing project and suggests next steps
/setup-engine godot 4.6   # Configures engine-specific agents and rules
/adopt                    # Integrates existing code/assets into the studio structure
```

### Game Design

```
/brainstorm               # Explore game ideas from scratch with creative-director
/map-systems              # Visualize system dependencies across the game
/design-system <name>     # Deep-dive design for a specific system (combat, economy, etc.)
/quick-design             # Rapid design doc for a small feature
/review-all-gdds          # Audit consistency across all Game Design Documents
/propagate-design-change  # Coordinate a design change across affected departments
```

### Stories & Sprints

```
/create-epics             # Break down the GDD into epics
/create-stories           # Decompose an epic into user stories
/dev-story <id>           # Start development on a story (branches, scaffolding, context)
/story-done <id>          # Close a story (tests, docs, PR checklist)
/sprint-plan              # Build a sprint from the backlog
/sprint-status            # Show current sprint health
/estimate <story-id>      # Get effort estimates with rationale
```

### Code & Architecture

```
/create-architecture      # Generate technical architecture for the project
/architecture-decision    # Record an ADR (Architecture Decision Record)
/architecture-review      # Audit current architecture against the design
/code-review              # Review a file or diff with lead-programmer
/tech-debt                # Identify and prioritize technical debt
/perf-profile             # Analyze performance bottlenecks
```

### QA & Testing

```
/qa-plan                  # Generate a QA plan for a feature or milestone
/smoke-check              # Fast sanity check — critical paths only
/regression-suite         # Full regression test generation
/test-setup               # Scaffold test infrastructure for the project
/test-evidence-review     # Review test results and flag gaps
/skill-test               # Test a skill after modifying it
/skill-improve            # Improve a skill based on test results
```

### Team Orchestration

These coordinate multiple agents simultaneously on a single feature:

```
/team-combat              # Spawns: gameplay-programmer + systems-designer + qa-tester
/team-narrative           # Spawns: writer + world-builder + narrative-director
/team-ui                  # Spawns: ui-programmer + ux-designer + accessibility-specialist
/team-release             # Spawns: release-manager + devops-engineer + qa-lead
/team-polish              # Spawns: performance-analyst + technical-artist + qa-tester
/team-audio               # Spawns: audio-director + sound-designer
/team-level               # Spawns: level-designer + world-builder + technical-artist
/team-live-ops            # Spawns: live-ops-designer + analytics-engineer + community-manager
/team-qa                  # Spawns: qa-lead + qa-tester + performance-analyst
```

### Release

```
/release-checklist        # Full pre-release gate check
/launch-checklist         # Day-of-launch verification
/changelog                # Generate changelog from git history
/patch-notes              # Player-facing patch notes from commits
/hotfix <issue>           # Scoped emergency fix workflow
```

---

## Project Structure

```
CLAUDE.md                         # Master studio configuration
.claude/
  settings.json                   # Hooks, permissions, safety rules
  agents/                         # 49 agent definitions (.md + YAML frontmatter)
  skills/                         # 72 slash commands (one subdirectory per skill)
  hooks/                          # 12 hook scripts (bash)
  rules/                          # 11 path-scoped coding standards
  statusline.sh                   # Shows context%, model, stage, epic breadcrumb
  docs/
    workflow-catalog.yaml         # 7-phase pipeline (read by /help)
    templates/                    # 39 document templates (GDDs, ADRs, specs, etc.)
src/                              # Game source code
assets/                           # Art, audio, VFX, shaders, data
design/                           # GDDs, narrative docs, level designs
docs/                             # Technical docs and ADRs
tests/                            # Unit, integration, performance, playtest suites
tools/                            # Build and pipeline tools
prototypes/                       # Isolated throwaway prototypes
production/                       # Sprint plans, milestones, release tracking
```

---

## How Agents Are Defined

Each agent is a Markdown file in `.claude/agents/` with YAML frontmatter:

```markdown
---
name: gameplay-programmer
tier: 3
model: claude-sonnet
lead: lead-programmer
do