master-orchestrator

Multi-step tasks are managed by decomposing them into parallelizable sub-tasks and coordinating specialized AI agents for execution. This orchestration uses a structured flow including requirement deep-diving, multi-dimensional decomposition, and sub-agent scheduling to ensure comprehensive analysis and efficient execution.

Multi-agent orchestration works by dispatching and managing specialized AI agents like researchers, coders, and planners to handle decomposed sub-tasks. It employs a 'diverge then converge' thinking model to explore solutions thoroughly before selecting the optimal path and integrating the results.

Yes, you can use multi-agent coordination for software architecture design. The orchestration decomposes the complex architecture request into specialized sub-tasks, coordinating different AI agents to handle components like research, coding, and strategic planning in parallel.

The best way to break down complex requests is through advanced task decomposition and multi-dimensional analysis. This process categorizes the requirements, schedules sub-agents for parallel execution, and integrates the results while resolving any conflicts to deliver a comprehensive report.

You should not use multi-agent orchestration for simple, linear tasks that lack multiple dependencies or require no parallel execution. The structured 6-step execution flow is designed for complex problems needing deep analysis, strategic planning, and creative generation across various specialized domains.