Decomposes complex coding tasks into independent, parallelizable stories through a Socratic planning interview.
The plan-swarm skill facilitates structured planning for multi-agent AI coding sessions by breaking down high-level requirements into atomic, testable stories. It utilizes a Socratic interview approach to identify scope, dependencies, and risks while ensuring work is partitioned to avoid file conflicts. By storing these plans as structured decisions, it enables seamless orchestration for autonomous swarms, making it ideal for large-scale feature implementation or complex refactoring projects where multiple AI agents need to work in tandem.
主な機能
014 GitHub stars
02Dependency analysis to ensure parallelizable task execution
03Persistent storage of plans for multi-agent orchestration
04Socratic interview workflow for deep requirement gathering
05Automated codebase exploration and context retrieval
06Structured story decomposition with clear acceptance criteria
ユースケース
01Breaking down a large feature request into parallelizable sub-tasks
02Defining clear boundaries and risks for complex system refactoring
03Preparing structured roadmaps for autonomous multi-agent coding swarms
