Study P · Sessions & memory
History was a teacher all along: two canned examples replace the whole conversation
Synthetic history
Study Overview
The Hypothesis
If history isn't state (the view carries that) and isn't positional help (Study O), maybe it's worked precedent; and precedent can be faked.
Precedent, Faked
Two unit-tested examples (an ordinal insert and a move, on a tree the model never edits, ~900 tokens) restore stateless sessions to full-history accuracy on both models: worst paired difference vs K-view anywhere is 2–0 (n.s.), end-states back to 18–20/20, and gemini's examples arm goes 60/60 on the late placements that defined the M failure class; with 20/20 byte-perfect endings.
Content, Not Conversation
The system-prompt framing ties the fake-turns framing, so the value is content, not conversation. Flat ~2.1k input/step forever, half of keep-history's cost at 12 edits, no context ceiling.
The Revised Guidance
The session guidance is revised: fresh view per turn + two worked examples in the system prompt.