team-verification-before-completion

Evidence-based task completion requires fresh verification output before marking work complete. You enforce this by applying a five-step gate (IDENTIFY, RUN, READ, VERIFY, CLAIM) that blocks premature completion claims until verification output passes.

Preventing premature completion claims involves triggering a formal verification gate across workflow stages like team-build, team-qa, and team-ship. This ensures explicit verification evidence is attached before any task completion status is accepted.

Yes, verification gates apply across ML metrics, data pipelines, query results, dashboard rendering, and schema validation. They require fresh verification output across these domains before declaring tasks done.

Implement a completion gate by executing the IDENTIFY, RUN, READ, VERIFY, CLAIM workflow. Before reporting task completion, trigger the gate and attach the evidence generated from the fresh verification output.

Task verification gates should not be used for intermediate progress updates or partial status reports. They are designed exclusively for moments when an agent claims work is done or signals task completion.

Evidence-based verification works with team-tdd, team-debug, team-build, team-qa, and team-ship workflows. It provides consistent quality control across all these software development stages by requiring fresh verification output.