Claude Code makes some requests in the background: suggesting your next prompt, summarising what happened while you were away, and pulling out things worth remembering. These can now continue straight from the model's previous reply while the prompt cache (the server's short-lived saved copy of the conversation) is still fresh, instead of sending the whole conversation again. If the server refuses this, later attempts send the full history as before.
It is behind a remote switch that is off unless it is switched on remotely.
It could cut how much of your usage these background requests take.
tengu_expressive_pine Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.1.319: off
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.1.319. It isn't a statement about your account. What a flag value here can and cannot tell you
New in this build: tengu_expressive_pine