Game Studio Agent Architecture — complete setup (Phases 1-7)

48 coordinated Claude Code subagents for indie game development:
- 3 leadership agents (creative-director, technical-director, producer)
- 10 department leads (game-designer, lead-programmer, art-director, etc.)
- 23 specialist agents (gameplay, engine, AI, networking, UI, tools, etc.)
- 12 engine-specific agents (Godot, Unity, Unreal with sub-specialists)

Infrastructure:
- 34 skills (slash commands) for workflows, reviews, and team orchestration
- 8 hooks for commit validation, asset checks, session management
- 11 path-scoped rules enforcing domain-specific standards
- 28 templates for design docs, reports, and collaborative protocols

Key features:
- User-driven collaboration protocol (Question → Options → Decision → Draft → Approval)
- Engine version awareness with knowledge-gap detection (Godot 4.6 pinned)
- Phase gate system for development milestone validation
- CLAUDE.md kept under 80 lines with extracted doc imports

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

.claude/docs/hooks-reference/post-sprint-retrospective.md

@@ -0,0 +1,78 @@
+# Hook: post-sprint-retrospective
+
+## Trigger
+
+Manual trigger at the end of each sprint (typically invoked by the producer
+agent or the human developer).
+
+## Purpose
+
+Automatically generates a retrospective starting point by analyzing the sprint
+data: what was planned vs completed, velocity changes, bug trends, and common
+blockers. This is not a git hook but a workflow hook invoked through the
+`producer` agent.
+
+## Implementation
+
+This is a workflow hook, not a git hook. It is invoked by running:
+
+```
+@producer Generate sprint retrospective for Sprint [N]
+```
+
+The producer agent should:
+
+1. **Read the sprint plan** from `production/sprints/sprint-[N].md`
+2. **Calculate metrics**:
+   - Tasks planned vs completed
+   - Story points planned vs completed (if used)
+   - Carryover items from previous sprint
+   - New tasks added mid-sprint
+   - Average task completion time
+3. **Analyze patterns**:
+   - Most common blockers
+   - Which agent/area had the most incomplete work
+   - Which estimates were most inaccurate
+4. **Generate the retrospective**:
+
+```markdown
+# Sprint [N] Retrospective
+
+## Metrics
+| Metric | Value |
+|--------|-------|
+| Tasks Planned | [N] |
+| Tasks Completed | [N] |
+| Completion Rate | [X%] |
+| Carryover from Previous | [N] |
+| New Tasks Added | [N] |
+| Bugs Found | [N] |
+| Bugs Fixed | [N] |
+
+## Velocity Trend
+[Sprint N-2]: [X] | [Sprint N-1]: [Y] | [Sprint N]: [Z]
+Trend: [Improving / Stable / Declining]
+
+## What Went Well
+- [Automatically detected: tasks completed ahead of estimate]
+- [Facilitator adds team observations]
+
+## What Went Poorly
+- [Automatically detected: tasks that were carried over or cut]
+- [Automatically detected: areas with significant estimate overruns]
+- [Facilitator adds team observations]
+
+## Blockers
+| Blocker | Frequency | Resolution Time | Prevention |
+|---------|-----------|----------------|-----------|
+
+## Action Items for Next Sprint
+| # | Action | Owner | Priority |
+|---|--------|-------|----------|
+
+## Estimation Accuracy
+| Area | Avg Planned | Avg Actual | Accuracy |
+|------|------------|-----------|----------|
+```
+
+5. **Save** to `production/sprints/sprint-[N]-retro.md`