Currently (near as I can tell) some of TextExtracts' processing is done in public functions, but a lot (getting text based on a page title, tidying output) takes place behind private functions that are only accessible through the API. So for other extensions to take full advantage of what TextExtracts can do, they have to go through the API, which per mediawiki.org is discouraged. It would be nice if TextExtracts' processing were opened up so other extensions could rely on it without going through the API.
One use case is the Description2 extension, which generates description meta tags from article content. As suggested on the extension's talk page, TextExtracts integration would substantially improve the page descriptions that the extension generates.