fix: Replace layout text objects instead of updating

Delete old title/subtitle objects and create new ones with custom text.
This fixes issues with:
- Split title objects (e.g., "The" + "Tutorial" as separate objects)
- Text truncation
- Overlapping text

Increased render wait time to 2 seconds for better stability.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <[email protected]>

mcp_server_comprehensive.py

@@ -424,19 +424,30 @@ async def create_thumbnail(inputs: Dict[str, Any]) -> Dict[str, Any]:
                 objects = objects_result.get("objects", [])
                 text_objects = [obj for obj in objects if obj.get("type") == "text"]
 
-                # Update all text objects that contain title-like text or are large
+                # Find large text objects (titles) and delete them
+                title_objects = []
                 for obj in text_objects:
-                    obj_id = obj.get("id")
-                    obj_text = obj.get("text", "")
                     font_size = obj.get("fontSize", 0)
-
-                    # If it's a large text object (likely a title)
-                    if font_size > 100 or "title" in obj_id.lower():
-                        await execute_api_call(page, "updateObject", obj_id, {
-                            "text": title,
-                            "isFromLayout": False
-                        })
-                        results.append({"step": "update_title_text", "objectId": obj_id})
+                    if font_size > 100 or "title" in obj.get("id", "").lower():
+                        title_objects.append(obj)
+
+                # Delete all old title objects
+                for obj in title_objects:
+                    await execute_api_call(page, "deleteObject", obj.get("id"))
+                    results.append({"step": "delete_old_title", "objectId": obj.get("id")})
+
+                # Add new title text
+                if title_objects:
+                    first_title = title_objects[0]
+                    await execute_api_call(page, "addText", title, {
+                        "x": first_title.get("x", 73),
+                        "y": first_title.get("y", 65),
+                        "fontSize": first_title.get("fontSize", 190),
+                        "fontFamily": first_title.get("fontFamily", "Inter"),
+                        "fill": first_title.get("fill", "#000000"),
+                        "fontWeight": "bold"
+                    })
+                    results.append({"step": "add_new_title"})
 
         # Update subtitle if specified
         if subtitle:
@@ -446,19 +457,34 @@ async def create_thumbnail(inputs: Dict[str, Any]) -> Dict[str, Any]:
                 objects = objects_result.get("objects", [])
                 text_objects = [obj for obj in objects if obj.get("type") == "text"]
 
-                # Find subtitle (smaller text, contains "subtitle" or "description")
+                # Find subtitle objects (smaller text with "subtitle" in text/id)
+                subtitle_objects = []
                 for obj in text_objects:
-                    obj_id = obj.get("id")
                     obj_text = obj.get("text", "").lower()
                     font_size = obj.get("fontSize", 0)
-
-                    # Smaller text or contains subtitle/description keywords
-                    if (font_size < 100 and "subtitle" in obj_text) or "subtitle" in obj_id.lower() or "description" in obj_id.lower():
-                        await execute_api_call(page, "updateObject", obj_id, {
-                            "text": subtitle,
-                            "isFromLayout": False
-                        })
-                        results.append({"step": "update_subtitle_text", "objectId": obj_id})
+                    obj_id = obj.get("id", "").lower()
+
+                    if "subtitle" in obj_text or "subtitle" in obj_id or "description" in obj_id:
+                        subtitle_objects.append(obj)
+
+                # Delete old subtitle objects
+                for obj in subtitle_objects:
+                    await execute_api_call(page, "deleteObject", obj.get("id"))
+                    results.append({"step": "delete_old_subtitle", "objectId": obj.get("id")})
+
+                # Add new subtitle text
+                if subtitle_objects:
+                    first_subtitle = subtitle_objects[0]
+                    await execute_api_call(page, "addText", subtitle, {
+                        "x": first_subtitle.get("x", 84),
+                        "y": first_subtitle.get("y", 428),
+                        "fontSize": first_subtitle.get("fontSize", 68),
+                        "fontFamily": first_subtitle.get("fontFamily", "Inter"),
+                        "fill": first_subtitle.get("fill", "#FFFFFF"),
+                        "hasBackground": first_subtitle.get("hasBackground", True),
+                        "backgroundColor": first_subtitle.get("backgroundColor", "#000000")
+                    })
+                    results.append({"step": "add_new_subtitle"})
 
         # Add Huggy if specified
         if huggy_id:
@@ -470,8 +496,8 @@ async def create_thumbnail(inputs: Dict[str, Any]) -> Dict[str, Any]:
             result = await execute_api_call(page, "addObject", obj)
             results.append({"step": "add_custom_object", "result": result})
 
-        # Wait for rendering to complete
-        await asyncio.sleep(1)
+        # Wait for rendering to complete (longer wait after adding new objects)
+        await asyncio.sleep(2)
 
         # Export the final thumbnail
         export_result = await execute_api_call(page, "exportCanvas", "png")