Dispatch Parallel Agents
Decompose a multi-task problem into independent units and dispatch subagents in parallel. Use when facing 2+ independent tasks with no shared state or sequential dependencies. Each agent gets a self-contained prompt with full context.
Decompose the work into independent tasks. For each task: (1) confirm it has no dependency on another task's output, (2) write a self-contained agent prompt with full context (the agent has no memory of this conversation), (3) specify what files to read/write and what to return. Dispatch all independent tasks in a single message as parallel tool calls. Wait for all to complete. Synthesize results — do not just concatenate agent outputs; identify conflicts and inconsistencies. Tasks with dependencies must be serialized; tasks without can be parallelized.