Here's the solution I kept:
http://intellitect.com/transitioning-between-sprintsiterations-with-tfs/
In short, you use Iteration Paths to determine the current sprint.
Then you create a query that returns all work items under the current iteration (which identifies your current sprint).
You use a Work Item Query Policy to ensure that all work items associated with the check-in are part of the result of the query (and thus part of the current iteration/sprint).