What a read-only step actually prints

A read-only step, a file read or a search, prints its one-line header the same way an edit does, but nothing further beneath it by default.

The actual result, a file's full contents or a long list of search matches, feeds straight back into the model's next decision rather than being dumped onto your screen in full, since a raw result like that is rarely something you personally need to read in the moment, only something the model needs to reason from.

This keeps the visible transcript readable during an investigation-heavy turn that might involve a dozen searches, rather than scrolling past pages of raw search output between each meaningful step.

This is a display choice specifically, not a change to what the model itself receives; aside from the size caps described in Why results are capped in size, which apply to every consumer of a result equally, nothing about the terminal staying terse means the model's own reasoning is working from less than what actually came back.

If you specifically want to see a file's contents yourself rather than only what the model does with it, opening the file directly in your own editor remains the straightforward way to do that; the transcript is not designed to double as a file viewer.