Enforce limits on file size when converting to text

src/interface/web/app/components/chatInputArea/chatInputArea.tsx

@@ -237,6 +237,16 @@ export const ChatInputArea = forwardRef<HTMLTextAreaElement, ChatInputProps>((pr
             ? Array.from(nonImageFiles).concat(Array.from(attachedFiles || []))
             : Array.from(attachedFiles || []);
 
+        // Ensure files are below size limit (10 MB)
+        for (let i = 0; i < newFiles.length; i++) {
+            if (newFiles[i].size > 10 * 1024 * 1024) {
+                setWarning(
+                    `File ${newFiles[i].name} is too large. Please upload files smaller than 10 MB.`,
+                );
+                return;
+            }
+        }
+
         const dataTransfer = new DataTransfer();
         newFiles.forEach((file) => dataTransfer.items.add(file));
         setAttachedFiles(dataTransfer.files);

src/khoj/routers/api_content.py

@@ -384,10 +384,25 @@ async def convert_documents(
     files: List[UploadFile],
     client: Optional[str] = None,
 ):
+    MAX_FILE_SIZE_MB = 10  # 10MB limit
+    MAX_FILE_SIZE_BYTES = MAX_FILE_SIZE_MB * 1024 * 1024
+
     converted_files = []
     supported_files = ["org", "markdown", "pdf", "plaintext", "docx"]
 
     for file in files:
+        # Check file size first
+        file_size = 0
+        content = await file.read()
+        file_size = len(content)
+        await file.seek(0)  # Reset file pointer
+
+        if file_size > MAX_FILE_SIZE_BYTES:
+            logger.warning(
+                f"Skipped converting oversized file ({file_size / 1024 / 1024:.1f}MB) sent by {client} client: {file.filename}"
+            )
+            continue
+
         file_data = get_file_content(file)
         if file_data.file_type in supported_files:
             extracted_content = (