research-orchestrator

Multi-agent research is coordinated by breaking down complex queries into structured workflows for parallel teams, using a lead-agent and subagent framework to manage execution and synthesize diverse findings into a coherent final answer.

Depth-first research workflows prioritize deep vertical exploration of specific subtopics, while breadth-first workflows distribute subagents to cover diverse horizontal perspectives before aggregating and synthesizing all findings across the research teams.

To set up a multi-agent research project, create a structured research plan specifying the required lead-agent and subagents, define task decomposition strategies, and establish output aggregation rules to synthesize the final results.

Yes, multi-agent orchestration supports analyzing new product design by decomposing the query into parallel research workflows, enabling multiple subagents to gather diverse perspectives and synthesize comprehensive analysis findings.

Synthesizing findings across parallel teams requires a structured approach to task definition and output aggregation, meaning complex queries without clear decomposition strategies may result in fragmented or incoherent final research synthesis.

Use multi-agent orchestration instead of standard research when facing complex queries that require parallel team coordination, diverse perspective gathering, and comprehensive synthesis that a single agent cannot efficiently achieve.