diff --git a/nanobot/agent/loop.py b/nanobot/agent/loop.py
index e7393e24b..3350c447b 100644
--- a/nanobot/agent/loop.py
+++ b/nanobot/agent/loop.py
@@ -351,6 +351,7 @@ class AgentLoop:
         on_progress: Callable[..., Awaitable[None]] | None = None,
         on_stream: Callable[[str], Awaitable[None]] | None = None,
         on_stream_end: Callable[..., Awaitable[None]] | None = None,
+        on_retry_wait: Callable[[str], Awaitable[None]] | None = None,
         *,
         session: Session | None = None,
         channel: str = "cli",
@@ -428,6 +429,7 @@ class AgentLoop:
             context_block_limit=self.context_block_limit,
             provider_retry_mode=self.provider_retry_mode,
             progress_callback=on_progress,
+            retry_wait_callback=on_retry_wait,
             checkpoint_callback=_checkpoint,
             injection_callback=_drain_pending,
         ))
@@ -738,6 +740,18 @@ class AgentLoop:
                 )
             )
 
+        async def _on_retry_wait(content: str) -> None:
+            meta = dict(msg.metadata or {})
+            meta["_retry_wait"] = True
+            await self.bus.publish_outbound(
+                OutboundMessage(
+                    channel=msg.channel,
+                    chat_id=msg.chat_id,
+                    content=content,
+                    metadata=meta,
+                )
+            )
+
         # Persist the triggering user message immediately, before running the
         # agent loop. If the process is killed mid-turn (OOM, SIGKILL, self-
         # restart, etc.), the existing runtime_checkpoint preserves the
@@ -756,6 +770,7 @@ class AgentLoop:
             on_progress=on_progress or _bus_progress,
             on_stream=on_stream,
             on_stream_end=on_stream_end,
+            on_retry_wait=_on_retry_wait,
             session=session,
             channel=msg.channel,
             chat_id=msg.chat_id,
diff --git a/nanobot/agent/runner.py b/nanobot/agent/runner.py
index 9b7d50316..d90c79fe6 100644
--- a/nanobot/agent/runner.py
+++ b/nanobot/agent/runner.py
@@ -71,6 +71,7 @@ class AgentRunSpec:
     context_block_limit: int | None = None
     provider_retry_mode: str = "standard"
     progress_callback: Any | None = None
+    retry_wait_callback: Any | None = None
     checkpoint_callback: Any | None = None
     injection_callback: Any | None = None
 
@@ -552,7 +553,7 @@ class AgentRunner:
             "tools": tools,
             "model": spec.model,
             "retry_mode": spec.provider_retry_mode,
-            "on_retry_wait": spec.progress_callback,
+            "on_retry_wait": spec.retry_wait_callback,
         }
         if spec.temperature is not None:
             kwargs["temperature"] = spec.temperature
diff --git a/nanobot/channels/manager.py b/nanobot/channels/manager.py
index 0e4821701..c0622a272 100644
--- a/nanobot/channels/manager.py
+++ b/nanobot/channels/manager.py
@@ -189,6 +189,9 @@ class ChannelManager:
                     if not msg.metadata.get("_tool_hint") and not self.config.channels.send_progress:
                         continue
 
+                if msg.metadata.get("_retry_wait"):
+                    continue
+
                 # Coalesce consecutive _stream_delta messages for the same (channel, chat_id)
                 # to reduce API calls and improve streaming latency
                 if msg.metadata.get("_stream_delta") and not msg.metadata.get("_stream_end"):