Model choice controls and managed defaults

The idea

Owners pick models per tier or fix the chains. People set floors or pins down to one step. Managed defaults move to new compatible models, and explicit pins stay fixed.

The problem
People need control over models without breaking routing or keeping stale defaults.
Why it matters
Only edited values are saved, overrides record who set them, and defaults stay current.
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.

Where it lives in the product

  • 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.

  • Own accounts, execution profiles and payers

    Hosted work uses API keys. A subscription runs only through the vendor's product or the person's own runner. Executor, model, host and payer inherit from the platform through org, role and worker.

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.