parallel-research

Multi-agent parallel research decomposes broad questions into independent sub-tasks, spawns concurrent AI researchers for each domain, and synthesizes findings into a unified Markdown report.

Breaking down complex research questions involves dividing the main inquiry into independent, focused sub-questions, allowing concurrent investigation across multiple domains before final synthesis into a comprehensive report.

Parallel research synthesis works by consolidating findings from concurrent AI researchers into a unified, comprehensive Markdown report after they independently address their assigned sub-tasks.

Yes, you need the Task tool for multi-agent research decomposition, as it is explicitly required for sub-agent instantiation during parallel execution and structured output synthesis.

Use parallel research decomposition when your research scope exceeds a single agent's context window or requires concurrent investigation across multiple domains, making a single agent insufficient for the task.

A limitation of multi-agent parallel research is that it requires independent sub-tasks for effective concurrent execution, meaning highly interdependent research questions may not decompose efficiently for parallel AI researchers.