working-with-subagents

You orchestrate parallel tasks with AI subagents by deploying stateful agents that execute delegated work while preserving context across calls, enabling coordinated and memory-aware collaboration without repeated setup.

The best way to delegate tasks to scout, coder, and memory agents is using prompt templates that assign distinct roles for exploration, coding, and memory restructuring within a shared memory block for coordinated workflows.

You need stateful subagents for multi-agent collaboration when tasks require parallel work, persistence across calls, and coordinated deployment among multiple agents to maintain shared context.

To start a parallel task execution workflow, deploy a simple scout task using the provided prompt templates to kick off multi-agent collaboration and enable shared memory blocks for coordinated work.

Yes, you can restructure memory across multiple agent calls without losing context because the system orchestrates stateful subagents designed to preserve context and maintain memory-aware collaboration throughout the workflow.

No, parallel task execution with subagents does not require repeated setup because shared memory blocks and stateful orchestration maintain context across calls, eliminating the need to reinitialize between tasks.