Skip to main content
As many as you say. A review stage names its reviewers and agree, the number that must accept for the stage to pass. Leave agree out and every reviewer must accept. Below the number, the findings go to the stage that owns the fix, which runs again with them, as many times as its retry allows. Obversa does not merge on the pass; a person or the next stage does.

In Obversa

Two reviewers from two families read the same change at once, and one acceptance is enough here:
examples/teams/threshold-panel.ts (excerpt)
Set agree to the number of reviewers when one dissent must be enough to fail the step. Below that, the change can pass despite a dissent, and the dissent is still on the record.

Cross-model review

The panel’s reviewers are checked against the writer’s family before the run starts, the same way a single reviewer is. Should another model review the code? has the rule.

Next steps