Multi-Agent System Design

★ 6.7 · orchestration

multi-agent-designing is a Claude Code skill that helps design the decomposition of multi-agent tasks, including agent selection based on individual strengths, context handoff protocols, and result merging strategies for parallel builds. Rather than describing what each agent does, the skill focuses on how to split work across agents — which subtasks to assign, in what order to merge outputs, and how to pass context through the pipeline. It complements general orchestration skills by providing a structured approach to task planning and agent coordination. Developers and system architects building multi-agent pipelines will find it useful for thinking through assignment logic and merge sequencing.