brainstorming

Guides collaborative design exploration and spec writing before any implementation begins.

Updated Aug 20, 2026
One-click install
npx skills add https://github.com/Pega88/agy-superpowers --skill brainstorming-pega88
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brainstorming
Source: https://github.com/Pega88/agy-superpowers/tree/main/.agents/plugins/superpowers/skills/brainstorming
Command: npx skills add https://github.com/Pega88/agy-superpowers --skill brainstorming-pega88

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? It prevents wasted implementation effort by forcing requirements, design, and user approval to happen before any code is written, catching wrong assumptions early. ## Core Features & Use Cases - Three-Path Classification: Routes every request into spike, bounded, or architectural tracks with appropriately scaled process and a mandatory approval gate. - Structured Design Dialogue: Asks clarifying questions one at a time, proposes 2-3 approaches with trade-offs, and presents sectioned designs for incremental approval. - Spec Authoring & Review: Writes validated design docs to dated spec files, runs a self-review for placeholders and contradictions, and can dispatch a spec-reviewer subagent. - Visual Companion Server: Serves browser-based mockups, diagrams, and clickable option screens via a local WebSocket server when questions are better shown than told. - Use Case: Before building a new feature, classify the request, refine the idea through dialogue, get explicit design approval, then hand off a committed spec to the writing-plans skill. ## Quick Start Ask the agent to brainstorm and design the new feature you have in mind before writing any code.

Frequently Asked Questions about brainstorming

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

FAQPage Schema
How do I brainstorm a feature design with an AI coding agent?

Invoke the brainstorming workflow before implementation: it classifies your request as spike, bounded, or architectural, asks clarifying questions one at a time, proposes approaches with trade-offs, and requires your explicit approval before any code is written.

What is the difference between spike, bounded, and architectural tasks?

A spike answers a feasibility question with throwaway work. Bounded covers well-scoped changes to existing code with a short in-chat design. Architectural covers new projects or subsystem restructuring and requires a full written spec and implementation plan.

Can I see visual mockups during design brainstorming?

Yes. The visual companion starts a local server that serves HTML mockups, diagrams, and clickable option screens to your browser. It is offered only when a question is genuinely visual, and your clicks are recorded as events the agent reads.

Does brainstorming work for small one-file changes?

Yes, via the bounded path. The design may be just two sentences in chat, but the approval gate still applies: the agent presents the short design and waits for your explicit yes before implementing.

What happens after the design spec is approved?

The spec is written to docs/superpowers/specs with a dated filename, self-reviewed for placeholders and contradictions, committed to git, and then handed off to the writing-plans skill to create the implementation plan.