diff --git a/plugins/platforms/simplex/adapter.py b/plugins/platforms/simplex/adapter.py
index 9c3d22a429fa..116874efc1e8 100644
--- a/plugins/platforms/simplex/adapter.py
+++ b/plugins/platforms/simplex/adapter.py
@@ -501,21 +501,24 @@ async def send(
         reply_to: Optional[str] = None,
         metadata: Optional[Dict[str, Any]] = None,
     ) -> SendResult:
-        """Send a text message to a contact or group."""
-        corr_id = self._make_corr_id()
+        """Send a text message to a contact or group.
 
+        Long content is split into ``MAX_MESSAGE_LENGTH``-sized chunks via the
+        base :meth:`truncate_message` helper and sent in order; short content
+        is delivered as a single message.
+        """
         if chat_id.startswith("group:"):
-            group_id = chat_id[6:]
-            cmd_str = f"#[{group_id}] {content}"
+            prefix = f"#[{chat_id[6:]}] "
         else:
-            cmd_str = f"@[{chat_id}] {content}"
+            prefix = f"@[{chat_id}] "
 
-        payload = {
-            "corrId": corr_id,
-            "cmd": cmd_str,
-        }
+        for chunk in self.truncate_message(content, MAX_MESSAGE_LENGTH):
+            payload = {
+                "corrId": self._make_corr_id(),
+                "cmd": f"{prefix}{chunk}",
+            }
+            await self._send_ws(payload)
 
-        await self._send_ws(payload)
         return SendResult(success=True)
 
     async def send_typing(self, chat_id: str, metadata=None) -> None:
@@ -640,19 +643,20 @@ async def _standalone_send(
 
     try:
         if chat_id.startswith("group:"):
-            group_id = chat_id[6:]
-            cmd_str = f"#[{group_id}] {message}"
+            prefix = f"#[{chat_id[6:]}] "
         else:
-            cmd_str = f"@[{chat_id}] {message}"
+            prefix = f"@[{chat_id}] "
 
-        payload = {
-            "corrId": f"hermes-snd-{int(time.time() * 1000)}",
-            "cmd": cmd_str,
-        }
+        chunks = BasePlatformAdapter.truncate_message(message, MAX_MESSAGE_LENGTH)
 
         async with _wsclient.connect(ws_url, open_timeout=10, close_timeout=5) as ws:
-            await ws.send(json.dumps(payload))
-            # Give the daemon a moment to process the command before closing.
+            for i, chunk in enumerate(chunks):
+                payload = {
+                    "corrId": f"hermes-snd-{int(time.time() * 1000)}-{i}",
+                    "cmd": f"{prefix}{chunk}",
+                }
+                await ws.send(json.dumps(payload))
+            # Give the daemon a moment to process the command(s) before closing.
             await asyncio.sleep(0.5)
 
         return {"success": True, "platform": "simplex", "chat_id": chat_id}
diff --git a/tests/gateway/test_simplex_plugin.py b/tests/gateway/test_simplex_plugin.py
index 1048168aa6e6..ed59c0612390 100644
--- a/tests/gateway/test_simplex_plugin.py
+++ b/tests/gateway/test_simplex_plugin.py
@@ -8,6 +8,7 @@
 from __future__ import annotations
 
 import json
+import re
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
@@ -234,6 +235,65 @@ async def test_send_group():
     assert result.success is True
 
 
+@pytest.mark.asyncio
+async def test_send_short_content_single_send():
+    """Content within the limit is delivered as exactly one WS send."""
+    from gateway.config import PlatformConfig
+    cfg = PlatformConfig(enabled=True, extra={"ws_url": "ws://localhost:5225"})
+    adapter = SimplexAdapter(cfg)
+
+    mock_ws = AsyncMock()
+    adapter._ws = mock_ws
+
+    result = await adapter.send("contact-42", "short message")
+    mock_ws.send.assert_called_once()
+    assert result.success is True
+
+
+@pytest.mark.asyncio
+async def test_send_long_content_chunks_into_ordered_sends():
+    """Content longer than the max is split into multiple ordered sends,
+    each chunk staying within the advertised limit."""
+    from gateway.config import PlatformConfig
+    cfg = PlatformConfig(enabled=True, extra={"ws_url": "ws://localhost:5225"})
+    adapter = SimplexAdapter(cfg)
+
+    mock_ws = AsyncMock()
+    adapter._ws = mock_ws
+
+    max_len = _simplex.MAX_MESSAGE_LENGTH
+    # Plain text (no code fences) longer than the limit -> several chunks.
+    long_text = "word " * (max_len // 2)
+    assert len(long_text) > max_len
+
+    result = await adapter.send("contact-42", long_text)
+    assert result.success is True
+    assert mock_ws.send.call_count > 1
+
+    total = mock_ws.send.call_count
+    bodies = []
+    for i, call in enumerate(mock_ws.send.call_args_list):
+        payload = json.loads(call[0][0])
+        # Every chunk is addressed to the same contact, in order.
+        assert payload["cmd"].startswith("@[contact-42] ")
+        chunk = payload["cmd"][len("@[contact-42] "):]
+        # Each chunk body must respect the per-message limit.
+        assert len(chunk) <= max_len
+        # Multi-part indicator appended by the base helper (whatever its exact
+        # spacing): a trailing "(n/N)" marker must be present, in send order.
+        assert re.search(r"\(\d+/\d+\)\s*$", chunk)
+        assert re.search(rf"\({i + 1}/{total}\)\s*$", chunk)
+        # Drop the trailing part-indicator to recover the original body text.
+        bodies.append(re.sub(r"\s*\(\d+/\d+\)\s*$", "", chunk))
+
+    # Content coverage: reassembling the indicator-stripped chunk bodies must
+    # account for every word of the original message. truncate_message splits
+    # on word boundaries and lstrips inter-chunk whitespace, so word-for-word
+    # reassembly is reliable even if exact byte concatenation is not.
+    reassembled = " ".join(bodies).split()
+    assert reassembled == long_text.split()
+
+
 @pytest.mark.asyncio
 async def test_send_when_ws_not_connected_does_not_crash():
     from gateway.config import PlatformConfig