coordinator-mode

You orchestrate multi-agent task execution by decomposing complex tasks into smaller subtasks for concurrent processing, executing parallel reads, sequential writes, and merging worker results into a coherent answer.

Task decomposition is needed when a single agent cannot handle task complexity, such as integrating multiple data sources and generating a comprehensive analysis across different domains.

Parallel processing executes read operations concurrently across worker agents, while sequential writes ensure write operations occur in order, maintaining data consistency during distributed task orchestration.

Yes, multi-agent coordination is ideal for scenarios requiring integration of multiple data sources, distributing the processing across worker agents and synthesizing the results into a comprehensive analysis.

The best way to merge worker agent results is through coherent result synthesis, which combines outputs from distributed processing into a single, unified answer after parallel task execution completes.

A single agent lacks the capacity to handle complex multi-domain tasks efficiently, making distributed processing with task decomposition and parallel execution necessary for accurate result synthesis.