소개
Multi-Agent Brainstorming is a specialized Claude Code skill designed to transform single-agent designs into robust, production-ready architectures through a simulated peer-review process. By assigning distinct, non-overlapping roles—such as the Skeptic, Constraint Guardian, and User Advocate—it systematically identifies hidden assumptions, security vulnerabilities, and usability issues. This sequential workflow ensures that every technical decision is stress-tested and documented in a mandatory Decision Log, providing high confidence and reducing the risk of failure in complex software projects.
