What is the Decision Log?

The Decision Log is a mandatory artifact created by the skill that records every decision made, alternatives considered, and how objections were resolved for future auditing.

Can I skip specific agents in the review process?

The skill is designed with a strict operating model where all roles—Skeptic, Constraint Guardian, and User Advocate—are invoked to ensure a robust, risk-mitigated final design.

What is the benefit of sequential review over parallel brainstorming?

Sequential review prevents 'idea swarm' chaos by forcing each agent to focus on a specific, non-overlapping mandate, ensuring thoroughness and preventing premature convergence.

When should I use this multi-agent skill?

Use it when a design or idea requires higher confidence, risk reduction, or a formal review before moving into the implementation phase.

Multi-Agent Brainstorming

Name: Multi-Agent Brainstorming
Author: claudiodearaujo

byclaudiodearaujo

•

Productivity & Workflow

Orchestrates a structured, multi-role peer review process to validate system designs and minimize project risks.

The Multi-Agent Brainstorming skill transforms single-agent ideas into robust, battle-tested architectures by simulating a formal peer-review process within Claude Code. By assigning specific, non-overlapping roles—such as the Skeptic, Constraint Guardian, and User Advocate—it identifies hidden assumptions, edge cases, and YAGNI violations before implementation begins. This gated workflow ensures that every design choice is explicitly logged and reviewed, preventing premature convergence and 'idea swarm' chaos while maintaining a rigorous audit trail of decisions through a mandatory Decision Log.

Key Features

01Conflict resolution and final disposition via a dedicated Arbiter agent

02Sequential review loop to prevent cognitive overload and chaotic brainstorming

03Gated exit criteria ensuring all risks are addressed before proceeding

041 GitHub stars

05Structured role-based agents including Skeptic, Constraint Guardian, and User Advocate

06Mandatory Decision Log for auditing choices, alternatives, and objections

Use Cases

01Validating complex system architectures before writing any production code

02Ensuring user experience alignment through dedicated User Advocate simulation

03Identifying potential security vulnerabilities or performance bottlenecks in new features

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/sistema-de-narra-o-de-livro multi-agent-brainstorming

For use in Claude.ai and ChatGPT

Download Skill