Nine-stage lifecycle; the builder never judges itself

The idea

Each item records which stages apply (context, architecture, design, spec, tests, build, standards, verification, delivery), why, who decided and which gate enforces each. A verdict from the builder's own context counts as Unverified.

The problem
Stages like design or tests could be skipped with no reason recorded, and builders graded their own work.
Why it matters
A required stage passes only on this round's evidence. Missing evidence holds once and then marks the receipt Unverified.
Where it stands
In production as of

Recorded evidence

  • Production · Read-only compiled probe

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

  • Stage exit checklists and loop limits

    Each stage has an exit checklist. Code judges the clear items and System One the unclear ones. Repeated holds go to a person, and a loop guard stops the run when the same result repeats.

  • Work review instead of code review

    Other employees verify a pushed commit in a fresh checkout, check and release the candidate, and watch production with rollback. Code writes 'ready' only when every stamp is current. A rule set picks each repository's review mode from its indicators.

  • System One decision cascade

    Bounded choices (route, next action, failure class, done, risk, intent) are typed questions. A rule answers first, then a typed arm, then a small LLM, each with a confidence.

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.