brainstorm

Facilitate collaborative requirements discovery for ambiguous software development tasks.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/BaSui01/agentflow --skill brainstorm-basui01
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brainstorm
Source: https://github.com/BaSui01/agentflow/tree/main/.agents/skills/brainstorm
Command: npx skills add https://github.com/BaSui01/agentflow --skill brainstorm-basui01

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams and AI collaborators turn vague or conflicting development requests into a concrete, testable PRD and an agreed-upon MVP so engineering can proceed with minimal rework.

Core Features & Use Cases

  • Task-first PRD seeding: Immediately creates task directories and seeds prd.md with discovered facts, assumptions, open questions, and acceptance criteria.
  • Research-driven option generation: Inspects the repository and external conventions to propose 2–3 concrete implementation approaches with trade-offs.
  • Structured Q&A convergence: Runs a one-question-per-message loop to resolve blocking and preference decisions, record ADR-lite decisions, and produce a stepwise implementation plan.

Quick Start

Use the brainstorm skill to create and seed a PRD for a vague development request, run repo research, and propose 2–3 actionable approaches to decide an MVP.

Frequently Asked Questions about brainstorm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I scope vague software development requests into a clear MVP?

To scope vague development requests into a clear MVP, use collaborative requirements discovery to inspect the codebase, generate task directories, seed a PRD with acceptance criteria, and converge on a testable implementation plan.

What is the best way to generate a PRD for ambiguous new features?

Generating a PRD for ambiguous new features requires researching the repository to propose 2-3 concrete implementation approaches with trade-offs, then running a one-question-at-a-time decision loop to record outcomes and ADR-lite decisions.

How does requirements discovery work for evaluating multiple implementation paths?

Requirements discovery for multiple implementation paths works by inspecting the repository and external conventions to propose actionable approaches, then driving a structured Q&A convergence loop to resolve blocking decisions and establish a stepwise plan.

Can I use this scoping process to resolve conflicting development assumptions?

Yes, you can use this scoping process to resolve conflicting development assumptions by seeding a prd.md document with discovered facts and open questions, then systematically resolving them through a structured, one-question-per-message decision loop.

What are the limitations of using a one-question-at-a-time decision loop for requirements gathering?

The limitation of using a one-question-at-a-time decision loop for requirements gathering is that it requires sequential, interactive collaboration to resolve blocking decisions, making it less suitable for fully autonomous or batch-processed scoping workflows.