Specs and the decision log

The idea

Work can start from a block spec that people and agents edit. It has notes on blocks, marks for unread changes, an audit, tickets per requirement, a trace from each requirement to its evidence, and a decision log.

The problem
Planning ended in chat history, and later nobody knew why one approach won.
Why it matters
Each requirement traces to a verdict, and the next plan starts from earlier decisions.
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.

  • Past decisions in the worker's context

    When a worker plans and acts, its context lists the matching accepted decisions, the owner's earlier choices and the accepted lessons, within a size limit, so the worker does not ask again.

  • Platform proof for each requirement

    Each part of a requirement needs proof that the platform made. A saved document gets an automatic check of its exact revision and bytes. Semantic criteria need a recorded model review.

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.