One effort allocation per item and stage

The idea

One decision per item and stage picks low, medium or high effort. That level sets the stage's turns, time, cost, rounds and consultations, and the worker can ask for more.

The problem
Fixed limits are too tight for hard work and wasteful for easy work.
Why it matters
Every limit comes from what is known about the work.
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.

  • Engineer pipeline with stage shares

    Work follows an engineer's pipeline, and each stage gets a share of the time and cost. Review is one plan review and one holistic diff review, within about 10% of spend.

  • Tier × effort routing from runtime signals

    Each call's model tier and effort are chosen separately. The inputs are the stage, the risk of the pending action, the change size, open findings, failures, repository size and the budget left. Floors tune from verified outcomes.

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.