An agent will often download images and files during a data extraction. By default, the agent saves these files in the internal database. However, you can configure the agent to save all files to disk by unchecking the Embed files in database box on the Internal Database configuration screen. When writing files to disk, the agent will store the full path to each file in the internal database.
If you are exporting data and there are corresponding files in the internal database, then these files will also export to the external database. If you uncheck the Embed files in database box, then only the file paths will export to the external database. If you are exporting data to a file format that doesn't support embedded files, such as Excel 2003, then only the file paths will export to the external database and the files will be written to disk.
Typically, Content Grabber will use the name of a file when saving it to disk. However, since a website could potentially use the same name for two different files, Content Grabber will set the file name with a unique identifier (GUID) if the file already exists on disk.
You can also use a file name transformation script to name the files, and configure the script to use some of the other extracted data elements for the file name. For example, you could name a product image with the product name or product ID.