feat(docs): add architecture deep-dive documentation v3.4.0

Add comprehensive technical documentation explaining Claude Code internals:

- NEW: guide/architecture.md (~800 lines) - Complete architecture deep-dive
  - Master Loop (while tool_call), Tool Arsenal, Context Management
  - Sub-Agent Architecture, Permission System, MCP Integration
  - Edit Tool mechanics, Session persistence, Philosophy comparison
  - 5 ASCII diagrams, source confidence tiers (Tier 1/2/3)

- guide/ultimate-guide.md: Add Section 2.7 "Under the Hood" summary
- guide/cheatsheet.md: Add "Under the Hood (Quick Facts)" section
- machine-readable/reference.yaml: Add architecture index + deep_dive refs
- Update all navigation links across repository

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

machine-readable/llms.txt

@@ -8,7 +8,8 @@ This repository provides everything needed to go from Claude Code beginner to po
 
 1. **Complete Guide** (`guide/ultimate-guide.md`) - 4700+ lines covering all aspects of Claude Code
 2. **Cheatsheet** (`guide/cheatsheet.md`) - 1-page printable daily reference
-3. **Audit Prompt** (`tools/audit-prompt.md`) - Self-contained prompt to analyze your Claude Code setup against best practices
+3. **Architecture Internals** (`guide/architecture.md`) - How Claude Code works under the hood (master loop, tools, context)
+4. **Audit Prompt** (`tools/audit-prompt.md`) - Self-contained prompt to analyze your Claude Code setup against best practices
 
 ## Target Audience
 
@@ -56,6 +57,7 @@ claude-code-ultimate-guide/
 ├── guide/                              # Core documentation
 │   ├── ultimate-guide.md               # Complete guide (main content)
 │   ├── cheatsheet.md                   # 1-page reference
+│   ├── architecture.md                 # How Claude Code works internally
 │   └── adoption-approaches.md          # Implementation strategy
 ├── tools/                              # Interactive utilities
 │   ├── audit-prompt.md                 # Setup audit tool
@@ -77,8 +79,16 @@ claude-code-ultimate-guide/
 └── quiz/                               # Interactive knowledge quiz
 ```
 
-## Guide Structure (10 Sections)
+## Guide Structure (10 Sections + Architecture)
 
+**Architecture Deep Dive** (`guide/architecture.md`):
+- Master Loop: `while(tool_call)` - no DAGs, no classifiers, no RAG
+- 8 Core Tools: Bash, Read, Edit, Write, Grep, Glob, Task, TodoWrite
+- Context: ~200K tokens, auto-compact at 75-92%
+- Sub-agents: isolated context, max depth=1
+- Philosophy: "less scaffolding, more model"
+
+**Main Guide Sections** (`guide/ultimate-guide.md`):
 1. **Quick Start** - Installation, first workflow, essential commands
 2. **Core Concepts** - Context management, Plan Mode, Rewind, Mental Model
 3. **Memory & Settings** - CLAUDE.md files, .claude/ folder, precedence rules

machine-readable/reference.yaml

@@ -3,14 +3,24 @@
 # Source: guide/ultimate-guide.md
 # Purpose: Condensed index for LLMs to quickly answer user questions about Claude Code
 
-version: "3.3.0"
+version: "3.4.0"
 updated: "2026-01"
 
 # ════════════════════════════════════════════════════════════════
 # DEEP DIVE - Line numbers in guide/ultimate-guide.md
 # Usage: Read file at line N for detailed documentation on topic
+# For architecture internals, see guide/architecture.md
 # ════════════════════════════════════════════════════════════════
 deep_dive:
+  # Architecture internals (guide/architecture.md)
+  architecture_master_loop: "guide/architecture.md:60"
+  architecture_tools: "guide/architecture.md:130"
+  architecture_context: "guide/architecture.md:200"
+  architecture_subagents: "guide/architecture.md:280"
+  architecture_permissions: "guide/architecture.md:350"
+  architecture_mcp: "guide/architecture.md:450"
+  architecture_philosophy: "guide/architecture.md:580"
+  # Main guide (guide/ultimate-guide.md)
   installation: 185
   first_workflow: 266
   essential_commands: 315
@@ -197,6 +207,19 @@ mcp:
   check: "/mcp"
   config: ".claude/mcp.json or ~/.claude.json"
 
+# ════════════════════════════════════════════════════════════════
+# ARCHITECTURE INTERNALS - see guide/architecture.md
+# ════════════════════════════════════════════════════════════════
+architecture:
+  master_loop: "while(tool_call) - no DAG, no classifier, no RAG"
+  core_tools: "Bash, Read, Edit, Write, Grep, Glob, Task, TodoWrite"
+  context_budget: "~200K tokens, auto-compact at 75-92%"
+  subagents: "Task tool spawns isolated agents (depth=1 max)"
+  philosophy: "less scaffolding, more model - trust Claude's reasoning"
+  mcp_protocol: "JSON-RPC 2.0, treated as native tools"
+  permissions: "interactive prompts + allow/deny rules + hooks"
+  deep_dive: "guide/architecture.md"
+
 # ════════════════════════════════════════════════════════════════
 # THINK LEVELS - see deep_dive.trinity_pattern
 # ════════════════════════════════════════════════════════════════