diff --git a/archon-poc/README.md b/archon-poc/README.md
new file mode 100644
index 0000000..d3bc4ed
--- /dev/null
+++ b/archon-poc/README.md
@@ -0,0 +1,62 @@
+# Archon Architecture Proof-of-Concept
+
+A three-layer architecture separating identity, runtime logic, and intelligence.
+
+## Architecture Overview
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│  Layer 1: Hermes Profile (Thin)                              │
+│  - Identity & routing only                                   │
+│  - No intelligence or reasoning logic                        │
+│  - Pure configuration (< 50 lines)                           │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  Layer 2: Claw Code Runtime                                  │
+│  - All business logic lives here                             │
+│  - Tool registry and execution                               │
+│  - Message routing and orchestration                         │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  Layer 3: Gemma 4 via Ollama                                 │
+│  - The actual intelligence                                   │
+│  - Local inference at localhost:11434                        │
+│  - Model: gemma3:4b (available) / gemma4:4b (target)         │
+└─────────────────────────────────────────────────────────────┘
+```
+
+## Files
+
+| File | Purpose |
+|------|---------|
+| `profile.yaml` | Thin Hermes layer - identity only |
+| `runtime/harness.py` | Claw Code runtime stub |
+| `runtime/tool_registry.py` | Tool definitions and registry |
+| `ollama_client.py` | Gemma 4 interface layer |
+| `test_integration.py` | End-to-end integration test |
+
+## Quick Start
+
+```bash
+# Run integration test
+python3 test_integration.py
+
+# Test Ollama connection
+python3 ollama_client.py "Hello, Gemma!"
+```
+
+## Design Principles
+
+1. **Profile is THIN** - Only identity, no logic
+2. **Runtime owns everything** - All intelligence orchestration
+3. **Local inference** - No external API dependencies
+4. **Testable** - Full integration test coverage
+
+## Epic Reference
+
+Implements Epic #370: Archon Architecture
+Dispatch #371: Proof-of-Concept Implementation
diff --git a/archon-poc/__pycache__/ollama_client.cpython-312.pyc b/archon-poc/__pycache__/ollama_client.cpython-312.pyc
new file mode 100644
index 0000000..50d557a
Binary files /dev/null and b/archon-poc/__pycache__/ollama_client.cpython-312.pyc differ
diff --git a/archon-poc/ollama_client.py b/archon-poc/ollama_client.py
new file mode 100644
index 0000000..d09f6a3
--- /dev/null
+++ b/archon-poc/ollama_client.py
@@ -0,0 +1,226 @@
+#!/usr/bin/env python3
+"""
+Ollama Client - Layer 3 Interface
+
+Connects to local Ollama instance for Gemma inference.
+This is the intelligence layer of the Archon architecture.
+"""
+
+import sys
+import json
+import requests
+from typing import Dict, Any, List, Optional, Generator
+from dataclasses import dataclass
+
+
+@dataclass
+class OllamaConfig:
+    base_url: str = "http://localhost:11434"
+    default_model: str = "gemma3:4b"
+    timeout: int = 120
+
+
+class OllamaClient:
+    """
+    Client for Ollama API - the intelligence layer.
+    
+    Communicates with local Ollama instance to run
+    Gemma models for inference.
+    """
+    
+    def __init__(self, base_url: str = "http://localhost:11434"):
+        self.config = OllamaConfig(base_url=base_url)
+        self.session = requests.Session()
+    
+    def health_check(self) -> bool:
+        """Check if Ollama is reachable."""
+        try:
+            response = self.session.get(
+                f"{self.config.base_url}/api/tags",
+                timeout=5
+            )
+            return response.status_code == 200
+        except requests.RequestException:
+            return False
+    
+    def list_models(self) -> List[str]:
+        """List available models."""
+        try:
+            response = self.session.get(
+                f"{self.config.base_url}/api/tags",
+                timeout=self.config.timeout
+            )
+            response.raise_for_status()
+            data = response.json()
+            return [m["name"] for m in data.get("models", [])]
+        except requests.RequestException as e:
+            raise OllamaError(f"Failed to list models: {e}")
+    
+    def generate(
+        self,
+        prompt: str,
+        model: Optional[str] = None,
+        system: Optional[str] = None,
+        context: Optional[list] = None,
+        options: Optional[Dict[str, Any]] = None
+    ) -> Dict[str, Any]:
+        """
+        Generate a response from the model.
+        
+        Args:
+            prompt: The user prompt
+            model: Model name (default: gemma3:4b)
+            system: System prompt
+            context: Previous context for conversation
+            options: Additional generation options
+        
+        Returns:
+            Response dict from Ollama
+        """
+        model = model or self.config.default_model
+        
+        payload = {
+            "model": model,
+            "prompt": prompt,
+            "stream": False
+        }
+        
+        if system:
+            payload["system"] = system
+        if context:
+            payload["context"] = context
+        if options:
+            payload["options"] = options
+        
+        try:
+            response = self.session.post(
+                f"{self.config.base_url}/api/generate",
+                json=payload,
+                timeout=self.config.timeout
+            )
+            response.raise_for_status()
+            return response.json()
+        except requests.RequestException as e:
+            raise OllamaError(f"Generation failed: {e}")
+    
+    def generate_stream(
+        self,
+        prompt: str,
+        model: Optional[str] = None,
+        system: Optional[str] = None
+    ) -> Generator[str, None, None]:
+        """
+        Stream generate responses.
+        
+        Yields response chunks as they arrive.
+        """
+        model = model or self.config.default_model
+        
+        payload = {
+            "model": model,
+            "prompt": prompt,
+            "stream": True
+        }
+        
+        if system:
+            payload["system"] = system
+        
+        try:
+            response = self.session.post(
+                f"{self.config.base_url}/api/generate",
+                json=payload,
+                stream=True,
+                timeout=self.config.timeout
+            )
+            response.raise_for_status()
+            
+            for line in response.iter_lines():
+                if line:
+                    try:
+                        data = json.loads(line)
+                        if "response" in data:
+                            yield data["response"]
+                    except json.JSONDecodeError:
+                        continue
+                        
+        except requests.RequestException as e:
+            raise OllamaError(f"Streaming failed: {e}")
+    
+    def chat(
+        self,
+        messages: list,
+        model: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Chat completion with message history.
+        
+        Args:
+            messages: List of {role, content} dicts
+            model: Model name
+        """
+        model = model or self.config.default_model
+        
+        payload = {
+            "model": model,
+            "messages": messages,
+            "stream": False
+        }
+        
+        try:
+            response = self.session.post(
+                f"{self.config.base_url}/api/chat",
+                json=payload,
+                timeout=self.config.timeout
+            )
+            response.raise_for_status()
+            return response.json()
+        except requests.RequestException as e:
+            raise OllamaError(f"Chat failed: {e}")
+
+
+class OllamaError(Exception):
+    """Ollama API error."""
+    pass
+
+
+def main():
+    """CLI entry point."""
+    client = OllamaClient()
+    
+    # Health check
+    print("Checking Ollama connection...")
+    if not client.health_check():
+        print("ERROR: Ollama not reachable at localhost:11434")
+        sys.exit(1)
+    print("✓ Ollama is running\n")
+    
+    # List models
+    print("Available models:")
+    for model in client.list_models():
+        print(f"  - {model}")
+    print()
+    
+    # Generate if prompt provided
+    if len(sys.argv) > 1:
+        prompt = " ".join(sys.argv[1:])
+        print(f"Prompt: {prompt}")
+        print("-" * 40)
+        
+        try:
+            response = client.generate(
+                prompt=prompt,
+                system="You are a helpful assistant. Be concise."
+            )
+            print(response.get("response", "No response"))
+            print("-" * 40)
+            print(f"Tokens: {response.get('eval_count', 'N/A')}")
+        except OllamaError as e:
+            print(f"Error: {e}")
+            sys.exit(1)
+    else:
+        print("Usage: ollama_client.py <prompt>")
+        print("Example: ollama_client.py 'What is the Archon architecture?'")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/archon-poc/profile.yaml b/archon-poc/profile.yaml
new file mode 100644
index 0000000..555b5aa
--- /dev/null
+++ b/archon-poc/profile.yaml
@@ -0,0 +1,25 @@
+name: Archon-POC
+display_name: Archon (Proof-of-Concept)
+model: gemma3:4b
+variant: archon
+provider: ollama
+ollama_host: http://localhost:11434
+
+creator: Ezra
+lineage: Archon / Timmy Time Nexus
+purpose: Three-layer architecture POC
+
+# THIN PROFILE - No logic, only identity and routing
+# All intelligence lives in Claw runtime + Gemma layer
+
+routing:
+  runtime: claw.harness
+  intelligence: ollama.gemma
+
+constraints:
+  max_tokens: 4096
+  temperature: 0.7
+
+tagging:
+  required: true
+  tag: "#archon-poc"
diff --git a/archon-poc/runtime/__pycache__/harness.cpython-312.pyc b/archon-poc/runtime/__pycache__/harness.cpython-312.pyc
new file mode 100644
index 0000000..940f546
Binary files /dev/null and b/archon-poc/runtime/__pycache__/harness.cpython-312.pyc differ
diff --git a/archon-poc/runtime/__pycache__/tool_registry.cpython-312.pyc b/archon-poc/runtime/__pycache__/tool_registry.cpython-312.pyc
new file mode 100644
index 0000000..3173914
Binary files /dev/null and b/archon-poc/runtime/__pycache__/tool_registry.cpython-312.pyc differ
diff --git a/archon-poc/runtime/harness.py b/archon-poc/runtime/harness.py
new file mode 100644
index 0000000..0784ae6
--- /dev/null
+++ b/archon-poc/runtime/harness.py
@@ -0,0 +1,177 @@
+#!/usr/bin/env python3
+"""
+Claw Code Runtime - Layer 2 of Archon Architecture
+
+This harness contains all business logic, orchestrating between
+the thin Hermes profile (Layer 1) and Gemma intelligence (Layer 3).
+"""
+
+import sys
+import json
+from typing import Dict, Any, Optional, List
+from pathlib import Path
+
+# Add parent to path for imports
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from runtime.tool_registry import ToolRegistry
+
+
+class ClawHarness:
+    """
+    The Claw Runtime - where all logic lives.
+    
+    Responsibilities:
+    - Message routing and orchestration
+    - Tool execution management
+    - Context window management
+    - Conversation state tracking
+    """
+    
+    def __init__(self, ollama_host: str = "http://localhost:11434"):
+        self.ollama_host = ollama_host
+        self.tools = ToolRegistry()
+        self.conversation_history: List[Dict[str, Any]] = []
+        self.session_id: Optional[str] = None
+        
+    def process_message(self, message: str, context: Optional[Dict] = None) -> Dict[str, Any]:
+        """
+        Main entry point for processing messages.
+        
+        Flow:
+        1. Parse intent
+        2. Determine if tools needed
+        3. Route to Gemma or execute tools
+        4. Return formatted response
+        """
+        context = context or {}
+        
+        # Log to conversation history
+        self.conversation_history.append({
+            "role": "user",
+            "content": message,
+            "timestamp": self._get_timestamp()
+        })
+        
+        # Check for tool invocation
+        tool_call = self.tools.parse_tool_call(message)
+        if tool_call:
+            result = self._execute_tool(tool_call)
+            return self._format_response(
+                content=result,
+                tool_used=tool_call["name"],
+                metadata={"tool_result": True}
+            )
+        
+        # Route to intelligence layer (Gemma)
+        return self._route_to_intelligence(message, context)
+    
+    def _route_to_intelligence(self, message: str, context: Dict) -> Dict[str, Any]:
+        """Route message to Gemma via Ollama."""
+        from ollama_client import OllamaClient
+        
+        client = OllamaClient(base_url=self.ollama_host)
+        
+        # Build prompt with context
+        prompt = self._build_prompt(message, context)
+        
+        # Get response from Gemma
+        response = client.generate(
+            model="gemma3:4b",
+            prompt=prompt,
+            system=self._get_system_prompt()
+        )
+        
+        # Log response
+        self.conversation_history.append({
+            "role": "assistant",
+            "content": response.get("response", ""),
+            "timestamp": self._get_timestamp()
+        })
+        
+        return self._format_response(
+            content=response.get("response", ""),
+            metadata={
+                "model": "gemma3:4b",
+                "tokens_used": response.get("eval_count", 0)
+            }
+        )
+    
+    def _execute_tool(self, tool_call: Dict) -> str:
+        """Execute a tool and return result."""
+        return self.tools.execute(tool_call)
+    
+    def _build_prompt(self, message: str, context: Dict) -> str:
+        """Build context-aware prompt for Gemma."""
+        history = "\n".join([
+            f"{msg['role']}: {msg['content']}"
+            for msg in self.conversation_history[-5:]  # Last 5 messages
+        ])
+        
+        return f"""Previous conversation:
+{history}
+
+User: {message}
+
+Assistant:"""
+    
+    def _get_system_prompt(self) -> str:
+        """Get system prompt for Gemma."""
+        return """You are the Archon POC, a helpful AI assistant.
+Be concise but thorough. Tag your response with #archon-poc."""
+    
+    def _format_response(self, content: str, tool_used: Optional[str] = None, 
+                        metadata: Optional[Dict] = None) -> Dict[str, Any]:
+        """Format response for return to Layer 1."""
+        response = {
+            "content": content,
+            "status": "success",
+            "layer": "claw_runtime",
+            "tag": "#archon-poc"
+        }
+        
+        if tool_used:
+            response["tool_used"] = tool_used
+        if metadata:
+            response["metadata"] = metadata
+            
+        return response
+    
+    def _get_timestamp(self) -> str:
+        """Get current timestamp."""
+        from datetime import datetime
+        return datetime.now().isoformat()
+
+
+def main():
+    """CLI entry point."""
+    harness = ClawHarness()
+    
+    if len(sys.argv) > 1:
+        message = " ".join(sys.argv[1:])
+        result = harness.process_message(message)
+        print(json.dumps(result, indent=2))
+    else:
+        # Interactive mode
+        print("Archon Harness - Interactive Mode")
+        print("Type 'exit' to quit\n")
+        
+        while True:
+            try:
+                message = input("> ")
+                if message.lower() in ("exit", "quit"):
+                    break
+                    
+                result = harness.process_message(message)
+                print(f"\n{result['content']}\n")
+                
+            except KeyboardInterrupt:
+                break
+            except EOFError:
+                break
+        
+        print("\nGoodbye!")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/archon-poc/runtime/tool_registry.py b/archon-poc/runtime/tool_registry.py
new file mode 100644
index 0000000..7e8b931
--- /dev/null
+++ b/archon-poc/runtime/tool_registry.py
@@ -0,0 +1,194 @@
+#!/usr/bin/env python3
+"""
+Tool Registry - Layer 2 Component
+
+Defines and manages tools available to the Claw runtime.
+Tools are executed locally, not sent to the intelligence layer.
+"""
+
+import re
+import json
+import subprocess
+from typing import Dict, Any, List, Callable, Optional
+from dataclasses import dataclass
+from datetime import datetime
+
+
+@dataclass
+class Tool:
+    name: str
+    description: str
+    parameters: Dict[str, Any]
+    handler: Callable
+
+
+class ToolRegistry:
+    """
+    Registry of available tools for the Claw runtime.
+    
+    Tools are pattern-matched from user messages and executed
+    before routing to the intelligence layer.
+    """
+    
+    def __init__(self):
+        self.tools: Dict[str, Tool] = {}
+        self._register_builtin_tools()
+    
+    def _register_builtin_tools(self):
+        """Register built-in tools."""
+        self.register(Tool(
+            name="time",
+            description="Get current time",
+            parameters={},
+            handler=self._get_time
+        ))
+        
+        self.register(Tool(
+            name="status",
+            description="Get system status",
+            parameters={},
+            handler=self._get_status
+        ))
+        
+        self.register(Tool(
+            name="echo",
+            description="Echo a message back",
+            parameters={"message": "string"},
+            handler=self._echo
+        ))
+        
+        self.register(Tool(
+            name="ollama_list",
+            description="List available Ollama models",
+            parameters={},
+            handler=self._ollama_list
+        ))
+    
+    def register(self, tool: Tool):
+        """Register a new tool."""
+        self.tools[tool.name] = tool
+    
+    def parse_tool_call(self, message: str) -> Optional[Dict[str, Any]]:
+        """
+        Parse a message for tool invocation.
+        
+        Patterns:
+        - /tool_name
+        - /tool_name param=value
+        - @tool_name
+        """
+        # Match /tool_name or @tool_name patterns
+        match = re.match(r'^[/@](\w+)(?:\s+(.+))?$', message.strip())
+        if not match:
+            return None
+        
+        tool_name = match.group(1)
+        args_str = match.group(2) or ""
+        
+        if tool_name not in self.tools:
+            return None
+        
+        # Parse parameters
+        params = {}
+        if args_str:
+            # Simple key=value parsing
+            for pair in args_str.split():
+                if '=' in pair:
+                    key, value = pair.split('=', 1)
+                    params[key] = value
+                else:
+                    # Positional argument as 'message'
+                    params["message"] = args_str
+        
+        return {
+            "name": tool_name,
+            "parameters": params
+        }
+    
+    def execute(self, tool_call: Dict[str, Any]) -> str:
+        """Execute a tool call."""
+        tool_name = tool_call["name"]
+        params = tool_call.get("parameters", {})
+        
+        if tool_name not in self.tools:
+            return f"Error: Unknown tool '{tool_name}'"
+        
+        tool = self.tools[tool_name]
+        
+        try:
+            result = tool.handler(**params)
+            return f"[Tool: {tool_name}]\n{result}"
+        except Exception as e:
+            return f"[Tool Error: {tool_name}]\n{str(e)}"
+    
+    def list_tools(self) -> List[str]:
+        """List all available tools."""
+        return [
+            f"{name}: {tool.description}"
+            for name, tool in self.tools.items()
+        ]
+    
+    # --- Tool Handlers ---
+    
+    def _get_time(self) -> str:
+        """Get current time."""
+        return datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+    
+    def _get_status(self) -> str:
+        """Get system status."""
+        return json.dumps({
+            "runtime": "claw",
+            "version": "1.0.0-poc",
+            "status": "operational",
+            "tools_available": len(self.tools)
+        }, indent=2)
+    
+    def _echo(self, message: str = "") -> str:
+        """Echo a message."""
+        return message
+    
+    def _ollama_list(self) -> str:
+        """List Ollama models."""
+        try:
+            result = subprocess.run(
+                ["curl", "-s", "http://localhost:11434/api/tags"],
+                capture_output=True,
+                text=True
+            )
+            data = json.loads(result.stdout)
+            models = [m["name"] for m in data.get("models", [])]
+            return f"Available models:\n" + "\n".join(f"  - {m}" for m in models)
+        except Exception as e:
+            return f"Error listing models: {e}"
+
+
+def main():
+    """CLI for testing tool registry."""
+    registry = ToolRegistry()
+    
+    print("Available tools:")
+    for tool_info in registry.list_tools():
+        print(f"  {tool_info}")
+    
+    print("\nTest parsing:")
+    test_messages = [
+        "/time",
+        "/status",
+        "/echo Hello world",
+        "/ollama_list",
+        "Regular message without tool"
+    ]
+    
+    for msg in test_messages:
+        parsed = registry.parse_tool_call(msg)
+        if parsed:
+            result = registry.execute(parsed)
+            print(f"\n> {msg}")
+            print(result)
+        else:
+            print(f"\n> {msg}")
+            print("(No tool call detected)")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/archon-poc/test_integration.py b/archon-poc/test_integration.py
new file mode 100644
index 0000000..9bb9f46
--- /dev/null
+++ b/archon-poc/test_integration.py
@@ -0,0 +1,214 @@
+#!/usr/bin/env python3
+"""
+Integration Test - Archon Architecture
+
+End-to-end test verifying the full three-layer flow:
+1. Thin Hermes Profile (Layer 1)
+2. Claw Code Runtime (Layer 2)  
+3. Gemma via Ollama (Layer 3)
+"""
+
+import sys
+import yaml
+import json
+import unittest
+from pathlib import Path
+from typing import Dict, Any
+
+# Add archon-poc to path
+sys.path.insert(0, str(Path(__file__).parent))
+
+from ollama_client import OllamaClient, OllamaError
+from runtime.harness import ClawHarness
+from runtime.tool_registry import ToolRegistry
+
+
+class TestArchonArchitecture(unittest.TestCase):
+    """Integration tests for Archon three-layer architecture."""
+    
+    @classmethod
+    def setUpClass(cls):
+        """Set up test fixtures."""
+        cls.base_path = Path(__file__).parent
+        
+    # === Layer 1: Profile Tests ===
+    
+    def test_01_profile_exists(self):
+        """Verify profile.yaml exists and is valid."""
+        profile_path = self.base_path / "profile.yaml"
+        self.assertTrue(profile_path.exists(), "profile.yaml must exist")
+        
+        with open(profile_path) as f:
+            profile = yaml.safe_load(f)
+        
+        self.assertIn("name", profile)
+        self.assertIn("model", profile)
+        self.assertEqual(profile["model"], "gemma3:4b")
+        
+    def test_02_profile_is_thin(self):
+        """Verify profile is thin (< 50 lines, no logic)."""
+        profile_path = self.base_path / "profile.yaml"
+        lines = profile_path.read_text().splitlines()
+        
+        self.assertLess(len(lines), 50, "Profile must be < 50 lines")
+        
+        # Check for routing section (should exist)
+        content = "\n".join(lines)
+        self.assertIn("routing:", content)
+        
+    # === Layer 2: Runtime Tests ===
+    
+    def test_03_harness_exists(self):
+        """Verify harness.py exists and imports."""
+        harness_path = self.base_path / "runtime" / "harness.py"
+        self.assertTrue(harness_path.exists())
+        
+        # Already imported in setUp, if we got here it works
+        harness = ClawHarness()
+        self.assertIsNotNone(harness)
+        
+    def test_04_tool_registry(self):
+        """Verify tool registry works."""
+        registry = ToolRegistry()
+        
+        # Test tool listing
+        tools = registry.list_tools()
+        self.assertGreater(len(tools), 0, "Must have at least one tool")
+        
+        # Test tool parsing
+        parsed = registry.parse_tool_call("/time")
+        self.assertIsNotNone(parsed)
+        self.assertEqual(parsed["name"], "time")
+        
+        # Test tool execution
+        result = registry.execute(parsed)
+        self.assertIn("Tool: time", result)
+        
+    def test_05_harness_message_processing(self):
+        """Test harness can process messages."""
+        harness = ClawHarness()
+        
+        # Test tool invocation path
+        result = harness.process_message("/status")
+        self.assertEqual(result["status"], "success")
+        self.assertEqual(result["layer"], "claw_runtime")
+        
+    # === Layer 3: Ollama Tests ===
+    
+    def test_06_ollama_connection(self):
+        """Verify Ollama is reachable."""
+        client = OllamaClient()
+        self.assertTrue(client.health_check(), "Ollama must be running")
+        
+    def test_07_ollama_models(self):
+        """Verify gemma model is available."""
+        client = OllamaClient()
+        models = client.list_models()
+        
+        # Check for gemma3:4b (available) or gemma4:4b (target)
+        gemma_models = [m for m in models if "gemma" in m.lower()]
+        self.assertGreater(len(gemma_models), 0, "Need at least one Gemma model")
+        
+    def test_08_ollama_generation(self):
+        """Test basic generation through Ollama."""
+        client = OllamaClient()
+        
+        response = client.generate(
+            prompt="Say 'Archon test passed' and nothing else.",
+            system="You are a test assistant. Be brief."
+        )
+        
+        self.assertIn("response", response)
+        self.assertIsInstance(response["response"], str)
+        self.assertGreater(len(response["response"]), 0)
+        
+    # === End-to-End Tests ===
+    
+    def test_09_full_flow_tool_path(self):
+        """Test full flow: Profile -> Runtime -> Tool -> Response."""
+        harness = ClawHarness()
+        
+        # Simulate what the thin profile would do - just route to harness
+        result = harness.process_message("/echo Integration test")
+        
+        self.assertEqual(result["status"], "success")
+        self.assertIn("content", result)
+        self.assertIn("Integration test", result["content"])
+        self.assertEqual(result["layer"], "claw_runtime")
+        self.assertEqual(result["tag"], "#archon-poc")
+        
+    def test_10_full_flow_intelligence_path(self):
+        """Test full flow: Profile -> Runtime -> Ollama -> Response."""
+        harness = ClawHarness()
+        
+        # This routes through Ollama
+        result = harness.process_message("Say 'Archon intelligence layer active'")
+        
+        self.assertEqual(result["status"], "success")
+        self.assertIn("content", result)
+        self.assertIn("layer", result)
+        self.assertIn("metadata", result)
+        
+    def test_11_conversation_history(self):
+        """Test conversation history tracking."""
+        harness = ClawHarness()
+        
+        # Send multiple messages
+        harness.process_message("Message 1")
+        harness.process_message("Message 2")
+        
+        # Check history
+        self.assertEqual(len(harness.conversation_history), 4)  # 2 user + 2 assistant
+        
+    # === Architecture Compliance Tests ===
+    
+    def test_12_all_files_exist(self):
+        """Verify all required files exist."""
+        required_files = [
+            "README.md",
+            "profile.yaml",
+            "runtime/harness.py",
+            "runtime/tool_registry.py",
+            "ollama_client.py",
+            "test_integration.py"
+        ]
+        
+        for file_path in required_files:
+            full_path = self.base_path / file_path
+            self.assertTrue(
+                full_path.exists(),
+                f"Required file missing: {file_path}"
+            )
+
+
+def run_tests():
+    """Run all integration tests."""
+    print("=" * 60)
+    print("ARCHON ARCHITECTURE - INTEGRATION TEST SUITE")
+    print("=" * 60)
+    print()
+    
+    # Create test suite
+    loader = unittest.TestLoader()
+    suite = loader.loadTestsFromTestCase(TestArchonArchitecture)
+    
+    # Run with verbose output
+    runner = unittest.TextTestRunner(verbosity=2)
+    result = runner.run(suite)
+    
+    print()
+    print("=" * 60)
+    if result.wasSuccessful():
+        print("✓ ALL TESTS PASSED")
+        print("Archon Architecture POC is working correctly!")
+        print("=" * 60)
+        return 0
+    else:
+        print("✗ SOME TESTS FAILED")
+        print("See output above for details")
+        print("=" * 60)
+        return 1
+
+
+if __name__ == "__main__":
+    sys.exit(run_tests())