agentdocs-orchestrator

To coordinate parallel agents, this skill decomposes complex requests into atomic tasks with explicit inputs and outputs, then synchronizes status across planning and runtime environments to execute them concurrently and aggregate results into a unified report.

Task decomposition for parallel execution is the process of breaking complex user requests down into atomic units with explicit inputs and outputs, enabling independent tasks to run via local simulation or Claude CLI-based sub-agents while managing task dependencies.

Managing task dependencies and runtime isolation involves coordinating parallel sub-agents through status synchronization that updates workflow documents, archives completed work, and provides guardrails for safe Claude CLI usage during execution.

Yes, you can use Claude CLI sub-agents for parallel execution by dropping this skill into your automation workflow and providing a complex request; the system auto-decomposes the request and orchestrates the sub-agents automatically.

The best way to aggregate results from parallel sub-agents is through final aggregation, which combines the outputs of synchronized, independent tasks into a cohesive, consumable report while archiving completed work.