Fan out multiple subagents to fix a batch of independent code-review findings or plan tasks in an iOS repo without them stomping on each other. Use when you have several findings/tasks to fix with TDD and each one needs an xcodebuild test run to verify — naive parallelism means two agents editing the same working tree, or two xcodebuild test runs racing on the same simulator device (flaky installs, launch failures, derived-data collisions). Use when the user says "fix these findings in parallel", "knock out this task list with subagents", or after a code-review/plan-review pass produced ≥2 independent items.
