Adaptive per-turn context

The idea

Typed decisions say which optional sections a turn needs and how to retrieve. Code plans the slots under a ceiling, with a reduction ladder that never cuts more than before, and limits the queries.

The problem
Every turn built every section and ran the same retrieval.
Why it matters
Small turns get smaller, while the prefix and the required sections never change.
Where it stands
In production as of

A step-by-step replay of this idea is not published yet. The library adds one when a recorded run shows it.

  • Shadow first, promote on measured outcomes

    A new decision point starts in shadow. It is scored offline on labelled cases and live on outcomes, then promoted or demoted on that evidence. It acts only on the answer keys measured safe.

  • Hybrid memory recall

    Recall ranks memories with full-text and trigram search, plus vectors where embeddings exist. It says which arms ran and why vectors are off, and it gives the worker its readable memory as files.

Turn the numbers into shipped work.

Crewshal runs these choices for you: a persistent AI worker with memory and rules, on your Claude and Codex subscriptions.