When capturing an image URI from Safari the result isn’t a JPEG file but a webarchive (which I consider a bug) While this works fine in EF opening it in Safari won’t work when offline (just stupid of it).
Working around that by saving the image to disk and importing the file means loosing meta such as the source URI.
Since there will always be situations where the user would choose a different format than EF I’d suggest adding metadata copying between entries.
This way I could capture the image from Safari, see that it’s not ideal, re-capture it as a file, bring over the meta, delete the webarchive.
For now I’ll try adjusting http://c-command.com/scripts/eaglefiler/duplicate-records