diff --git a/src/fluxer/migrate_message.py b/src/fluxer/migrate_message.py
index 37d5c06..8f74cce 100644
--- a/src/fluxer/migrate_message.py
+++ b/src/fluxer/migrate_message.py
@@ -1,4 +1,5 @@
 import asyncio
+import discord
 import logging
 import re
 from typing import Callable, Awaitable, Dict, Any
@@ -74,6 +75,16 @@ async def migrate_messages(context: MigrationContext, source_channel_id: int, ta
             if not context.is_running:
                 break
                 
+            # Skip system messages like "pinned a message", etc.
+            # We treat thread_starter_message (type 21) as our thread marker.
+            if msg.type == discord.MessageType.thread_starter_message:
+                content = f"> <<< THREAD: **{msg.channel.name}** >>>"
+            elif msg.type not in [discord.MessageType.default, discord.MessageType.reply]:
+                continue
+            else:
+                # Get clean content
+                content = msg.clean_content
+                
             # Process attachments
             files = []
             attachments_to_process = list(msg.attachments)
@@ -85,13 +96,13 @@ async def migrate_messages(context: MigrationContext, source_channel_id: int, ta
                 is_forwarded = msg.flags.forwarded
             
             # If forwarded, the content and attachments might be in message_snapshots (discord.py 2.5+)
-            content = msg.clean_content
+            # Note: If content was set by thread_starter_message, we don't overwrite it.
             if is_forwarded:
                 logger.debug(f"Detected forwarded message: ID={msg.id}, Flags={msg.flags.value}")
                 if hasattr(msg, 'message_snapshots') and msg.message_snapshots:
                     # For now we handle the first snapshot
                     snapshot = msg.message_snapshots[0]
-                    if not content:
+                    if not content: # Only update content if it wasn't already set (e.g., by thread_starter_message)
                         content = snapshot.content
                         if hasattr(msg, 'guild') and msg.guild:
                             content = clean_mentions(content, msg.guild)
@@ -132,13 +143,6 @@ async def migrate_messages(context: MigrationContext, source_channel_id: int, ta
                     thread = msg.thread
                     logger.info(f"Detected thread '{thread.name}' on message {msg.id}")
                     
-                    # Send Start Marker
-                    stats["threads"] += 1
-                    await context.fluxer_writer.send_marker(
-                        channel_id=target_channel_id,
-                        content=f"> <<< THREAD: **{thread.name}** >>>"
-                    )
-                    
                     # Migrate thread messages
                     # We don't pass a progress callback here to avoid confusing the UI
                     # but we do want to track count if possible.
diff --git a/src/fluxer/writer.py b/src/fluxer/writer.py
index a4feb0d..64bd9a1 100644
--- a/src/fluxer/writer.py
+++ b/src/fluxer/writer.py
@@ -1,7 +1,8 @@
 import asyncio
+import io
 import logging
 from typing import Optional, List, Dict, Any
-from fluxer import Bot, Webhook, Forbidden
+from fluxer import Bot, Webhook, Forbidden, File
 
 logger = logging.getLogger(__name__)
 
@@ -245,6 +246,11 @@ class FluxerWriter:
             
         final_content = prefix + display_content if display_content else prefix
 
+        # Convert files to fluxer.File objects
+        fluxer_files = None
+        if files:
+            fluxer_files = [File(io.BytesIO(f["data"]), filename=f["filename"]) for f in files]
+
         try:
             # Current limitation: fluxer.py execute_webhook doesn't support 'message_reference' yet.
             # So if we have a reply, we MUST use the bot's direct send method.
@@ -253,7 +259,7 @@ class FluxerWriter:
                     content=final_content,
                     username=f"{author_name} (discord)",
                     avatar_url=author_avatar_url,
-                    files=files,
+                    files=fluxer_files,
                     wait=True
                 )
                 return str(msg.id) if msg else None
@@ -274,7 +280,7 @@ class FluxerWriter:
                 msg_data = await self.client.send_message(
                     channel_id=channel_id,
                     content=final_bot_content,
-                    files=files,
+                    files=fluxer_files,
                     message_reference=message_reference
                 )
                 return str(msg_data["id"]) if msg_data else None
@@ -290,11 +296,16 @@ class FluxerWriter:
         Sends a simple marker message (e.g., thread start/end) using the bot directly.
         """
         assert self.client is not None
+        
+        fluxer_files = None
+        if files:
+            fluxer_files = [File(io.BytesIO(f["data"]), filename=f["filename"]) for f in files]
+            
         try:
             msg_data = await self.client.send_message(
                 channel_id=channel_id,
                 content=content,
-                files=files
+                files=fluxer_files
             )
             return str(msg_data["id"]) if msg_data else None
         except Exception as e:
diff --git a/src/stoat/migrate_message.py b/src/stoat/migrate_message.py
index c2842af..3d97146 100644
--- a/src/stoat/migrate_message.py
+++ b/src/stoat/migrate_message.py
@@ -1,4 +1,5 @@
 import asyncio
+import discord
 import logging
 import re
 from typing import Callable, Awaitable, Dict, Any
@@ -73,6 +74,17 @@ async def migrate_messages(context: MigrationContext, source_channel_id: int, ta
             if not context.is_running:
                 break
                 
+            # Skip system messages like "pinned a message", etc.
+            # We treat thread_starter_message (type 21) as our thread marker.
+            content = "" # Initialize content
+            if msg.type == discord.MessageType.thread_starter_message:
+                content = f"> <<< THREAD: **{msg.channel.name}** >>>"
+            elif msg.type not in [discord.MessageType.default, discord.MessageType.reply]:
+                continue
+            else:
+                # Get clean content
+                content = msg.clean_content
+                
             # Process attachments
             files = []
             attachments_to_process = list(msg.attachments)
@@ -82,8 +94,6 @@ async def migrate_messages(context: MigrationContext, source_channel_id: int, ta
             if hasattr(msg.flags, 'forwarded'):
                 is_forwarded = msg.flags.forwarded
             
-            # Get clean content
-            content = msg.clean_content
             if is_forwarded:
                 logger.debug(f"Detected forwarded message: ID={msg.id}, Flags={msg.flags.value}")
                 if hasattr(msg, 'message_snapshots') and msg.message_snapshots:
@@ -128,13 +138,6 @@ async def migrate_messages(context: MigrationContext, source_channel_id: int, ta
                     thread = msg.thread
                     logger.info(f"Detected thread '{thread.name}' on message {msg.id}")
                     
-                    # Send Start Marker
-                    stats["threads"] += 1
-                    await context.stoat_writer.send_marker(
-                        channel_id=target_channel_id,
-                        content=f"> <<< THREAD: **{thread.name}** >>>"
-                    )
-                    
                     # Migrate thread messages recursively
                     thread_stats = await migrate_messages(
                         context=context,
diff --git a/src/ui/app.py b/src/ui/app.py
index 9068fa3..c0eca69 100644
--- a/src/ui/app.py
+++ b/src/ui/app.py
@@ -1076,7 +1076,8 @@ class MigrationCLI:
 
             # 2. Select Target Channel
             with console.status(f"[yellow]Fetching {platform_name} channels...[/yellow]"):
-                f_channels = await self.engine.writer.get_channels()
+                full_f_channels = await self.engine.writer.get_channels()
+                f_channels = [c for c in full_f_channels if c.get('name') not in ["reaper_logs", "reaper-logs"]]
             if not f_channels:
                 console.print(f"[yellow]No channels found in {platform_name} community.[/yellow]")
                 return