dispatching-parallel-agents

Debug multiple independent test failures in parallel by splitting them into separate focused agent tasks. Each agent receives a scoped objective to investigate an unrelated failure, allowing simultaneous root cause analysis without waiting for sequential fixes.

Task decomposition for parallel debugging splits independent failures into isolated workstreams. Use it when multiple unrelated test files, subsystems, or bug domains fail and can be investigated without shared state between them.

Scope agent prompts for parallel root cause analysis by giving each agent a narrow objective, constraints, and expected output. This ensures focused investigation per failure and produces structured results for integration.

Yes, you can dispatch parallel agents for test failures across different subsystems when those failures are independent. The agents investigate separate bug domains simultaneously, provided no shared state exists between the tasks.

Verify combined fixes after dispatching parallel agents by comparing results, checking for conflicts, and validating the integrated changes. This post-fix integration step ensures that independently resolved issues do not introduce regressions.

You should not use parallel agents for debugging when failures share state or depend on each other. If fixing one bug alters the conditions of another, sequential debugging is required to avoid conflicting results.