team

To execute parallel tasks concurrently, you can orchestrate multiple workers by spawning N agents and routing independent workstreams to them. This distributes tasks via MCP tools to maximize throughput and reduce total cycle time.

Multi-agent orchestration is the coordination of N parallel workers to execute independent tasks within a framework. It routes specific task types, like frontend or backend work, to appropriate agent roles via MCP tools while monitoring progress and verifying results.

You route frontend and backend tasks by invoking a team skill that decomposes plans into parallelizable workstreams. The orchestration logic automatically routes frontend tasks to designer agents and backend tasks to architect agents using MCP tools.

Yes, you can use parallel workers within the oh-my-ccg framework to execute independent tasks. The skill handles end-to-end coordination, task distribution, lifecycle management, and safety checks to ensure correctness during concurrent execution.

You should use parallel execution when you have a set of independent tasks identified from a plan. Orchestrating multiple workers concurrently maximizes throughput and significantly reduces total project delivery cycle time while preserving correctness.

To spawn multiple workers, invoke the team skill with a task description and worker count, such as specifying 4 workers to spawn UI component tasks. The framework handles progress monitoring and verification to ensure quality.