Relatively sure it's not. I'm using pi and I can't find any place it's doing any such thing. My current theory is that it's behavior trained into the model either accidentally or intentionally, probably during post-training in some harness that _does_ give hints of that sort?

You could also increase compaction.reserveTokens in ~/.pi/agent/settings.json to make pi compact more often. Also you can use /compact <instructions> to specify what facts should be included in the summary.

One hard part is that from a handoff file you only have information for 1 session, so you have to manually determine next week which facts are still true. At Substrate we solve this problem using our memory: we have 1 memory outside of an agent and track the source of every fact. https://trysubstrate.co/l/a2hn

after thinking about it more, in the attention there is some positional encoding, so I think it's just that tokens are attending to this as well.