To search for all the cards that have a non-empty 'sentence' field use the search
sentence:_*
or
sentence:_* tag:JLPT1 for just the for JLPT1 words which have example sentences.
In fact, since only the Core6000 words have sentences, just use tag:CORE instead of sentence:_*
To use the deck as a sentence deck, you will have to, of course, make a new card model, but that is easily done. It depends on what kind of sentence deck you want - reading sentences, producing, listening and understanding?
If you want to sort the cards/sentences as discussed in this thread
http://forum.koohii.com/showthread.php?tid=5091
so that you can work from sentences with a few, more common kanji and gradually add new kanji,
use the fields 'ko2001-index' or 'rtkliteko2001-index' fields, provided by cangy in the original spreadsheets. See the above thread for more details, but basically, in the browser sort on one of the index fields, from lowest to highest. Then use the shared plugin called reset creation times to resort the tag:CORE cards in the order of the chosen index field. If you have them in the right order and give them a range of dates earlier than all the other cards in the deck (eg 2009) they should appear as new cards in the right order. Do backup well first.
I'm not sure how to answer this better, since I'm not sure what kind of sentence deck you want, but I hope it is of some help.
Edited: 2011-06-29, 8:13 am