tasks

Automated task orchestration runs an autonomous execution loop that repeatedly processes ready tasks until all are complete. It creates, executes, and tracks work items without stopping, eliminating manual task management overhead.

Autonomous task management handles blockers by dynamically spawning subtasks to resolve the issue. It re-evaluates task readiness to unblock work and ensures continuous workflow execution without manual intervention.

You can coordinate sub-agents in task automation by waiting for results via TaskOutput before continuing. This sub-agent coordination ensures deterministic progress and prevents the workflow from advancing prematurely.

The best way to start autonomous task processing is to create an initial task and run the processing loop. Executing the command starts the autonomous loop immediately, processing ready tasks and tracking them until completion.

Autonomous task orchestration operates without external dependencies. It enforces safe, continuous execution internally through blocker resolution, deployment polling guidance, and non-stop looping until all tasks are completed.

You should avoid an autonomous execution loop if your workflow requires manual intervention between steps or cannot handle dynamic subtask spawning. The loop is designed for non-stop, deterministic progress without pausing for human input.