What problem does it solve? Coordinating multiple AI agents on a shared codebase requires structured supervision: dispatching tasks, waiting on worker completion or escalations, gating decisions, and decomposing work across agents. This Skill provides the Orca orchestration workflow so an agent can act as a coordinator over supervised Orca workers instead of improvising ad-hoc handoffs. ## Core Features & Use Cases - Supervised worker coordination: Send threaded messages, use blocking ask/reply flows, dispatch tasks, and wait on worker_done or escalation events. - Task DAGs and decision gates: Decompose work across agents, define task dependencies, and run coordinator loops with conditional action gates. - Version-matched guidance: Loads the exact orchestration reference served by the installed Orca binary via ORCA skills get orchestration, so instructions never drift from the runtime. - Use Case: A user asks to split a feature across three agents in separate worktrees, monitor their progress, and gate the merge on review approval — the coordinator dispatches tasks, waits on worker_done events, and applies the decision gate. ## Quick Start Ask the agent to coordinate multiple Orca workers on a task DAG, then have it run the resolved Orca executable's skills get orchestration command to load the version-matched guide before dispatching any work.