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/context-management.md

@@ -0,0 +1,26 @@
+# Context Management
+
+Context is the most critical resource in a Claude Code session. Manage it actively:
+
+- **Compact proactively** at ~65-70% context usage, not reactively when at the limit
+- **Use `/clear`** between unrelated tasks, or after 2+ failed correction attempts
+- **Context budgets by task type**:
+  - Light (read/review): ~3k tokens startup
+  - Medium (implement feature): ~8k tokens
+  - Heavy (multi-system refactor): ~15k tokens
+- **Preserve during compaction**: modified file list, active sprint tasks, architectural
+  decisions made this session, agent invocation outcomes, test results, unresolved blockers
+- When delegating to subagents, provide full context in the prompt -- subagents do not
+  inherit conversation history unless explicitly given it
+
+# Compaction Instructions
+
+When context is compacted, preserve the following in the summary:
+
+- List of files modified in this session and their purpose
+- Any architectural decisions made and their rationale
+- Active sprint tasks and their current status
+- Agent invocations and their outcomes (success/failure/blocked)
+- Test results (pass/fail counts, specific failures)
+- Unresolved blockers or questions awaiting user input
+- The current task and what step we are on