Hierarchical Agent Orchestration — Manager & Worker Architecture

I will build a hierarchical agent orchestration system — a manager agent that decomposes complex tasks, assigns sub-tasks to specialised worker agents, monitors worker progress, handles worker failures by reassigning tasks, and integrates worker outputs into a coherent final result. Hierarchical orchestration scales to more complex tasks than flat agent architectures — the manager agent's decomposition and assignment capability allows the system to handle task complexity and scale that a single orchestrator directing all agents simultaneously cannot manage effectively.

The build covers manager agent with task decomposition and assignment logic, worker agent pool with specialised configurations, progress monitoring and timeout handling, worker failure detection and task reassignment, output integration, and a hierarchical execution trace for debugging.