feat(P0/P1/P2): Add E2E agent prober, version pinning, prometheus fixes

gateway-bot/http_api.py

@@ -3106,3 +3106,46 @@ async def health():
         "agents_count": len(AGENT_REGISTRY),
         "timestamp": datetime.utcnow().isoformat(),
     }
+
+
+@router.post("/debug/agent_ping")
+async def debug_agent_ping(request: dict = None):
+    """
+    E2E probe endpoint - tests full agent pipeline.
+    Used by agent-e2e-prober for monitoring.
+    Returns success only if router responds.
+    """
+    import time
+    start = time.time()
+    
+    try:
+        # Test 1: Check router connectivity
+        router_url = os.getenv("ROUTER_URL", "http://router:8000")
+        async with httpx.AsyncClient(timeout=10.0) as client:
+            router_resp = await client.get(f"{router_url}/health")
+            router_ok = router_resp.status_code == 200
+        
+        # Test 2: Check memory service connectivity
+        memory_url = os.getenv("MEMORY_SERVICE_URL", "http://memory-service:8000")
+        async with httpx.AsyncClient(timeout=10.0) as client:
+            memory_resp = await client.get(f"{memory_url}/health")
+            memory_ok = memory_resp.status_code == 200
+        
+        latency = time.time() - start
+        
+        return {
+            "success": router_ok and memory_ok,
+            "latency_seconds": round(latency, 3),
+            "checks": {
+                "router": router_ok,
+                "memory_service": memory_ok,
+            },
+            "timestamp": datetime.utcnow().isoformat(),
+        }
+    except Exception as e:
+        return {
+            "success": False,
+            "error": str(e)[:100],
+            "latency_seconds": round(time.time() - start, 3),
+            "timestamp": datetime.utcnow().isoformat(),
+        }