dispatching-parallel-agents

Debug multiple failing test files concurrently by dispatching independent debugging tasks across parallel agents. Group failures by root-cause domains, assign each agent a narrow target, and run investigations simultaneously to resolve unrelated failures faster.

Fix independent test failures by constraining each parallel agent's scope to a specific test file or subsystem. This prevents race conditions and overlapping edits, ensuring conflict-safe integration when merging concurrent investigations.

Use parallel agents for debugging when you have 2 or more unrelated failing test files, broken subsystems, or concurrent investigations that have no shared state or sequential dependencies. Sequential investigation is better for failures sharing root causes.

Yes, you can run concurrent investigations on broken subsystems without edit conflicts by grouping failures by independence and assigning focused agent scopes. Review each agent's results and check for overlapping edits during integration to ensure safety.

Validate CI test suites after merging parallel debugging changes by running the full test suite. This confirms all issues are resolved and checks that integrated changes from multiple agents do not introduce new test failures or regressions.