diff --git a/koboldcpp.py b/koboldcpp.py
index 52156d45f..080d7ea93 100755
--- a/koboldcpp.py
+++ b/koboldcpp.py
@@ -9,7 +9,6 @@
 # scenarios and everything Kobold and KoboldAI Lite have to offer.
 
 import os
-import queue
 try:
     os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID" # try set GPU to PCI order first thing
 except Exception:
@@ -482,18 +481,11 @@ class MCPStdioClient:
         self.lock = threading.Lock()
         self.stderr_buffer = []
         self.stderr_limit = 20
-        self.stderr_alive = True
-        self.stdout_alive = True
-        self.stdout_queue = queue.Queue()
+        self.alive = True
         self.stderr_thread = threading.Thread(
             target=self._read_stderr,
             daemon=True
         )
-        self.stdout_thread = threading.Thread(
-            target=self._read_stdout,
-            daemon=True
-        )
-        self.stdout_thread.start()
         self.stderr_thread.start()
     def _read_stderr(self):
         try:
@@ -505,51 +497,23 @@ class MCPStdioClient:
                 if len(self.stderr_buffer) > self.stderr_limit:
                     self.stderr_buffer.pop(0)
         finally:
-            self.stderr_alive = False
-    def _read_stdout(self):
-        try:
-            for line in self.process.stdout:
-                if not line:
-                    break
-                line = line.rstrip("\n")
-                if line:
-                    try:
-                        data = json.loads(line)
-                        if isinstance(data, dict) and isinstance(data.get("id"), (int, str)):
-                            self.stdout_queue.put(data)
-                    except json.JSONDecodeError:
-                        # Ignore lines that aren't valid JSON
-                        pass
-        finally:
-            self.stdout_alive = False
-    def _empty_queue(self):
-        # Drain any stale responses belonging to orphans or other IDs
-        while not self.stdout_queue.empty():
-            try:
-                self.stdout_queue.get_nowait()
-            except queue.Empty:
-                break
+            self.alive = False
 
     def send(self, message: dict, await_response=True) -> dict: # Send JSON-RPC request and wait for one response.
         line = json.dumps(message)
-        response = None
-        with self.lock: #idea: since we use a lock and empty queue before/after, we guarantee that any ID belongs to us, single request in flight only
+        with self.lock:
             if self.process.stdin.closed:
                 raise RuntimeError("MCP server stdin is closed")
-            self._empty_queue()
             self.process.stdin.write(line + "\n")
             self.process.stdin.flush()
             if not await_response:
                 return None
-            try:
-                response = self.stdout_queue.get(timeout=30) # short 30s timeout in case stalled
-            except queue.Empty:
-                response = None
+            response = self.process.stdout.readline()
         if not response:
             errmsg = "\n".join(self.stderr_buffer[-10:])
             print(f"[MCP Server Error!]\n{errmsg}")
-            raise RuntimeError("MCP server stdio response failed")
-        return response
+            raise RuntimeError("MCP server closed stdout")
+        return json.loads(response)
     def notify(self, message: dict) -> None: # Send JSON-RPC notification (no response expected).
         line = json.dumps(message)
         with self.lock: