diff --git a/.claude/tools/agent_cache_wrapper.py b/.claude/tools/agent_cache_wrapper.py
new file mode 100644
index 0000000..94f69ee
--- /dev/null
+++ b/.claude/tools/agent_cache_wrapper.py
@@ -0,0 +1,115 @@
+# .claude/tools/agent_cache_wrapper.py
+"""
+Cache wrapper for AI agents
+Use this in your agent workflows to add caching
+"""
+
+import os
+import sys
+from pathlib import Path
+
+# Add the tools directory to Python path
+tools_dir = Path(__file__).parent
+sys.path.insert(0, str(tools_dir))
+
+from local_cache_client import (
+    get_cache,
+    cached_ai_query,
+    store_ai_response,
+    print_cache_stats,
+)
+
+
+class AgentCacheWrapper:
+    """Wrapper for agent AI calls with caching support."""
+
+    def __init__(self, agent_type: str, project: str = None):  # type: ignore
+        self.agent_type = agent_type
+        self.project = project or os.getenv("AI_CACHE_PROJECT", "job_forge")
+        self.cache = get_cache()
+
+        print(f"🤖 {agent_type.title()} Agent initialized with caching")
+
+    def query_with_cache(self, prompt: str, make_ai_call_func=None) -> str:
+        """
+        Query with cache support.
+
+        Args:
+            prompt: The prompt to send
+            make_ai_call_func: Function to call if cache miss (should return AI response)
+
+        Returns:
+            AI response (from cache or fresh API call)
+        """
+        # Try cache first
+        cached_response, was_hit = cached_ai_query(
+            prompt, self.agent_type, self.project
+        )
+
+        if was_hit:
+            return cached_response  # type: ignore
+
+        # Cache miss - make AI call
+        if make_ai_call_func:
+            print(f"🤖 Making fresh AI call for {self.agent_type}...")
+            ai_response = make_ai_call_func(prompt)
+
+            # Store in cache for next time
+            if ai_response:
+                store_ai_response(prompt, ai_response, self.agent_type, self.project)
+
+            return ai_response
+        else:
+            print(f"⚠️ No AI call function provided for cache miss")
+            return None  # type: ignore
+
+    def store_response(self, prompt: str, response: str):
+        """Manually store a response in cache."""
+        store_ai_response(prompt, response, self.agent_type, self.project)
+
+    def get_stats(self):
+        """Get cache statistics for this session."""
+        return self.cache.get_stats()
+
+
+# Convenience functions for each agent type
+def technical_lead_query(prompt: str, ai_call_func=None) -> str:
+    """Technical Lead agent with caching."""
+    wrapper = AgentCacheWrapper("technical_lead")
+    return wrapper.query_with_cache(prompt, ai_call_func)
+
+
+def qa_engineer_query(prompt: str, ai_call_func=None) -> str:
+    """QA Engineer agent with caching."""
+    wrapper = AgentCacheWrapper("qa_engineer")
+    return wrapper.query_with_cache(prompt, ai_call_func)
+
+
+def devops_engineer_query(prompt: str, ai_call_func=None) -> str:
+    """DevOps Engineer agent with caching."""
+    wrapper = AgentCacheWrapper("devops_engineer")
+    return wrapper.query_with_cache(prompt, ai_call_func)
+
+
+def fullstack_developer_query(prompt: str, ai_call_func=None) -> str:
+    """Full-Stack Developer agent with caching."""
+    wrapper = AgentCacheWrapper("fullstack_developer")
+    return wrapper.query_with_cache(prompt, ai_call_func)
+
+
+# Example usage and testing
+if __name__ == "__main__":
+    # Example AI call function (replace with your actual Claude Code integration)
+    def example_ai_call(prompt):
+        # This is where you'd call Claude Code or your AI service
+        # For testing, return a mock response
+        return f"Mock AI response for: {prompt[:50]}..."
+
+    # Test with Technical Lead
+    response = technical_lead_query(
+        "What is the current FastAPI project structure?", example_ai_call
+    )
+    print(f"Response: {response}")
+
+    # Print stats
+    print_cache_stats()
diff --git a/.claude/tools/local_cache_client.py b/.claude/tools/local_cache_client.py
new file mode 100644
index 0000000..4b5ed20
--- /dev/null
+++ b/.claude/tools/local_cache_client.py
@@ -0,0 +1,307 @@
+# .claude/tools/local_cache_client.py
+"""
+AI Cache Client for Local Development
+Integrates with n8n-based AI response caching system
+"""
+
+import requests
+import json
+import os
+import hashlib
+import time
+from typing import Optional, Dict, Any
+from datetime import datetime
+
+
+class AICacheClient:
+    """Client for interacting with AI Cache MCP service."""
+
+    def __init__(self, base_url: str = None, enabled: bool = True):  # type: ignore
+        # Default to your n8n webhook URL
+        self.base_url = base_url or os.getenv(
+            "AI_CACHE_URL", "https://n8n.hotserv.cloud/webhook"
+        )
+        self.enabled = (
+            enabled and os.getenv("AI_CACHE_ENABLED", "true").lower() == "true"
+        )
+        self.timeout = int(os.getenv("AI_CACHE_TIMEOUT", "15"))
+
+        # Stats tracking
+        self.session_hits = 0
+        self.session_misses = 0
+        self.session_start = time.time()
+        self.connection_failed = False
+
+        if self.enabled:
+            print(f"🧠 AI Cache enabled: {self.base_url}")
+            self._test_connection()
+        else:
+            print("⚠️ AI Cache disabled")
+
+    def _test_connection(self):
+        """Test if the cache service is accessible."""
+        try:
+            response = requests.get(
+                f"{self.base_url}/ai-cache-stats", 
+                timeout=3  # Quick test
+            )
+            if response.status_code == 200:
+                print("✅ Cache service is accessible")
+            else:
+                print(f"⚠️ Cache service returned HTTP {response.status_code}")
+                self.connection_failed = True
+        except Exception as e:
+            print(f"❌ Cache service unreachable: {str(e)[:50]}...")
+            self.connection_failed = True
+
+    def _normalize_prompt(self, prompt: str) -> str:
+        """Normalize prompt for consistent matching."""
+        return prompt.strip().lower().replace("\n", " ").replace("  ", " ")
+
+    def lookup_cache(
+        self, prompt: str, agent_type: str, project: str = "job_forge"
+    ) -> Optional[str]:
+        """Look up a cached AI response."""
+        if not self.enabled or self.connection_failed:
+            return None
+
+        try:
+            start_time = time.time()
+
+            response = requests.post(
+                f"{self.base_url}/ai-cache-lookup",
+                json={"prompt": prompt, "agent_type": agent_type, "project": project},
+                timeout=self.timeout,
+            )
+
+            lookup_time = (time.time() - start_time) * 1000
+
+            if response.status_code == 200:
+                try:
+                    # Debug: print raw response
+                    raw_text = response.text
+                    print(f"🔍 Debug - Raw response: '{raw_text[:100]}...'")
+                    
+                    if not raw_text.strip():
+                        print(f"❌ Cache MISS [{agent_type}] - Empty response | Lookup: {lookup_time:.0f}ms")
+                        self.session_misses += 1
+                        return None
+                        
+                    data = response.json()
+                    if data.get("found"):
+                        similarity = data.get("similarity", 1.0)
+                        hit_count = data.get("hit_count", 1)
+
+                        print(
+                            f"✅ Cache HIT! [{agent_type}] Similarity: {similarity:.2f} | Used: {hit_count}x | Lookup: {lookup_time:.0f}ms"
+                        )
+                        self.session_hits += 1
+                        return data.get("response")
+                    else:
+                        print(f"❌ Cache MISS [{agent_type}] | Lookup: {lookup_time:.0f}ms")
+                        self.session_misses += 1
+                        return None
+                except json.JSONDecodeError as e:
+                    print(f"🚨 JSON decode error: {str(e)} | Response: '{response.text[:50]}'")
+                    self.session_misses += 1
+                    return None
+            else:
+                print(f"⚠️ Cache lookup failed: HTTP {response.status_code}")
+                return None
+
+        except requests.exceptions.Timeout:
+            print(f"⏱️ Cache lookup timeout ({self.timeout}s)")
+            return None
+        except Exception as e:
+            print(f"🚨 Cache error: {str(e)}")
+            return None
+
+    def store_cache(
+        self,
+        prompt: str,
+        response: str,
+        agent_type: str,
+        ai_service: str = "claude",
+        model: str = "claude-sonnet-4",
+        project: str = "job_forge",
+    ) -> bool:
+        """Store an AI response in cache."""
+        if not self.enabled or not response or len(response.strip()) < 10:
+            return False
+
+        try:
+            start_time = time.time()
+
+            result = requests.post(
+                f"{self.base_url}/ai-cache-store",
+                json={
+                    "prompt": prompt,
+                    "response": response,
+                    "ai_service": ai_service,
+                    "model": model,
+                    "agent_type": agent_type,
+                    "project": project,
+                },
+                timeout=self.timeout,
+            )
+
+            store_time = (time.time() - start_time) * 1000
+
+            if result.status_code == 200:
+                data = result.json()
+                if data.get("success"):
+                    print(
+                        f"💾 Response cached [{agent_type}] | Store: {store_time:.0f}ms"
+                    )
+                    return True
+                else:
+                    print(
+                        f"📄 Already cached [{agent_type}] | Store: {store_time:.0f}ms"
+                    )
+                    return False
+            else:
+                print(f"⚠️ Cache store failed: HTTP {result.status_code}")
+                return False
+
+        except requests.exceptions.Timeout:
+            print(f"⏱️ Cache store timeout ({self.timeout}s)")
+            return False
+        except Exception as e:
+            print(f"🚨 Cache store error: {str(e)}")
+            return False
+
+    def get_stats(self) -> Dict[str, Any]:
+        """Get cache statistics."""
+        try:
+            response = requests.get(
+                f"{self.base_url}/ai-cache-stats", timeout=self.timeout
+            )
+
+            if response.status_code == 200:
+                stats = response.json()
+
+                # Add session stats
+                session_time = time.time() - self.session_start
+                session_total = self.session_hits + self.session_misses
+                session_hit_rate = (
+                    (self.session_hits / session_total * 100)
+                    if session_total > 0
+                    else 0
+                )
+
+                stats["session_stats"] = {
+                    "hits": self.session_hits,
+                    "misses": self.session_misses,
+                    "total": session_total,
+                    "hit_rate_percentage": round(session_hit_rate, 1),
+                    "duration_minutes": round(session_time / 60, 1),
+                }
+
+                return stats
+            else:
+                return {"error": f"Failed to get stats: {response.status_code}"}
+
+        except Exception as e:
+            return {"error": f"Stats error: {str(e)}"}
+
+    def print_session_summary(self):
+        """Print session cache performance summary."""
+        total = self.session_hits + self.session_misses
+        if total == 0:
+            return
+
+        hit_rate = (self.session_hits / total) * 100
+        session_time = (time.time() - self.session_start) / 60
+
+        print(f"\n📊 Cache Session Summary:")
+        print(
+            f"   Hits: {self.session_hits} | Misses: {self.session_misses} | Hit Rate: {hit_rate:.1f}%"
+        )
+        print(f"   Session Time: {session_time:.1f} minutes")
+
+        if hit_rate > 60:
+            print(f"   🎉 Excellent cache performance!")
+        elif hit_rate > 30:
+            print(f"   👍 Good cache performance")
+        else:
+            print(f"   📈 Cache is learning your patterns...")
+
+
+# Global cache instance
+_cache_instance = None
+
+
+def get_cache() -> AICacheClient:
+    """Get or create global cache instance."""
+    global _cache_instance
+    if _cache_instance is None:
+        _cache_instance = AICacheClient()
+    return _cache_instance
+
+
+def cached_ai_query(
+    prompt: str, agent_type: str, project: str = "job_forge"
+) -> tuple[Optional[str], bool]:
+    """
+    Helper function for cached AI queries.
+    Returns: (cached_response, was_cache_hit)
+    """
+    cache = get_cache()
+    cached_response = cache.lookup_cache(prompt, agent_type, project)
+
+    if cached_response:
+        return cached_response, True
+    else:
+        return None, False
+
+
+def store_ai_response(
+    prompt: str, response: str, agent_type: str, project: str = "job_forge"
+):
+    """Helper function to store AI responses."""
+    cache = get_cache()
+    cache.store_cache(prompt, response, agent_type, project=project)
+
+
+def print_cache_stats():
+    """Print current cache statistics."""
+    cache = get_cache()
+    stats = cache.get_stats()
+
+    if "error" in stats:
+        print(f"❌ {stats['error']}")
+        return
+
+    summary = stats.get("summary", {})
+    session = stats.get("session_stats", {})
+
+    print(f"\n📈 AI Cache Statistics:")
+    print(f"   Overall Hit Rate: {summary.get('hit_rate_percentage', 0)}%")
+    print(f"   Total Saved: ${summary.get('total_cost_saved_usd', 0):.2f}")
+    print(f"   API Calls Saved: {summary.get('api_calls_saved', 0)}")
+
+    if session:
+        print(
+            f"   This Session: {session['hits']}/{session['total']} hits ({session['hit_rate_percentage']}%)"
+        )
+
+
+# Example usage for testing
+if __name__ == "__main__":
+    # Test the cache
+    cache = get_cache()
+
+    # Test lookup
+    result = cache.lookup_cache("What is the database schema?", "technical_lead")
+    print(f"Lookup result: {result}")
+
+    # Test store
+    cache.store_cache(
+        "What is the database schema?",
+        "PostgreSQL with users and applications tables",
+        "technical_lead",
+    )
+
+    # Print stats
+    print_cache_stats()
+    cache.print_session_summary()