Truscan
Extract

Summarise a page without guessing

Give the model the article, not the navigation and the cookie banner.

Summary quality is mostly input quality. A model handed raw HTML spends its attention on menus and disclaimers. Handed the article, it summarises the article.

  • Boilerplate is removed before the text reaches the model, which cuts token cost as well as noise.
  • A word count arrives with each page, so you can route long pages to a different strategy.
  • Ask to omit the body and you still get the excerpt, which is sometimes all a list view needs.

The excerpt field is a short opening extract, which is often enough on its own for a list view.