team-testing

To automate multi-agent test generation and execution, you orchestrate coordinator, strategist, generator, executor, and analyst workers to run staged L1/L2/L3 testing pipelines. This achieves progressive coverage targets through automated pipeline selection, task dispatch, and sessioned result storage.

A Generator-Critic loop in test automation works by iteratively generating, running, and revising tests through configurable max rounds. The GC loop control applies coverage gating to close the loop on failures, ensuring test quality improves progressively until targets are met.

Yes, you can use multi-agent orchestration for staged L1/L2/L3 testing workflows in CI. The orchestration automates pipeline selection, task creation, dispatch, and coverage gating, producing session directories and team_msg logs for reproducible test sessions.

The best way to manage reproducible test sessions and coverage history is by persisting session directories alongside team_msg logging and shared wisdom. This approach stores coverage history and telemetry for audit and trend analysis across test automation runs.

You should not use multi-agent test orchestration for software repositories when manual test coordination is sufficient or when staged L1/L2/L3 pipelines are unnecessary. Without coverage gating or GC loop requirements, the overhead of spawning multiple worker agents outweighs the benefits.

Yes, team-testing supports automated quality reporting after pull requests. It runs a standard pipeline to generate L1 unit tests, executes them, initiates a Generator-Critic loop if needed, and produces a final L1 coverage report and quality summary.