Harness tools that run checks and keep evidence

The idea

QA test cases run in a headless browser on the item's own previews, with screenshots. Runbooks run as drills with approval and rollback. Device flashes are checked by a read-back hash.

The problem
QA, operations and hardware claims without runs or evidence cannot be trusted.
Why it matters
Each case or step has a status and stored evidence, and a person approves each risky action.
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.

  • Opinionated harness per work type

    Each work template carries steps, a quality bar, pitfalls, tool guidance and practitioner notes, plus a minimum model tier and effort. Its named industry checklists are settled by detectors or by attestations with evidence.

  • Visual evidence and regressions

    A visible change needs captures, and before-and-after pairs where asked. A service change needs a live check. A comparison fails on anything the change broke against the base.

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.