brainstorm_skill

This skill helps you brainstorm ideas using Socratic questioning and multi-perspective analysis to refine concepts for implementation planning.
  • Python

9

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill mhylle/claude-skills-collection --skill brainstorm

  • SKILL.md13.5 KB

Overview

This skill delivers structured, interactive idea refinement using a Socratic questioning methodology combined with multi-perspective analysis and proactive research. It helps turn rough concepts into well-defined, implementable descriptions and identifies gaps, risks, and enhancement opportunities. Use it to prepare ideas for implementation planning or to decide whether architectural decisions need formal documentation.

How this skill works

On invocation the skill captures the core concept, goals, constraints, and project context, then runs iterative Socratic questioning until you signal readiness. It spawns targeted research agents (web research always, codebase research when code context is detected), applies Six Thinking Hats, SCAMPER, and a premortem, and then synthesizes findings into actionable insights. If clear architectural decisions arise, it prepares ADR content and can hand off to create-plan for implementation planning.

When to use it

  • You have a rough idea and want to surface hidden assumptions and gaps.
  • You need to evaluate options and choose between architectural or technical approaches.
  • You want a multi-perspective review (risks, benefits, creative alternatives).
  • You’re preparing an idea for implementation planning or ADR documentation.
  • You need focused research on patterns, best practices, and comparable projects.

Best practices

  • Provide an initial description with stated goals and any constraints to speed phase 1.
  • Answer Socratic questions candidly; the process relies on iterative clarification.
  • Flag any references to files, modules, or the codebase to enable codebase research.
  • Don’t rush: allow multiple Q/A rounds before moving to research and synthesis.
  • Use the synthesized output directly to drive create-plan or to create ADRs when decisions crystallize.

Example use cases

  • Refining a new feature idea before writing a proposal or ticket.
  • Exploring trade-offs between two implementation approaches for a shared service.
  • Finding missing requirements and edge cases for an MVP.
  • Preparing an architectural decision for documentation as an ADR.
  • Assessing reuse opportunities by comparing the idea to existing codebase patterns.

FAQ

I typically ask 2–4 questions per round and continue until you say you’re ready to proceed; there’s no fixed limit.

Will you research my codebase automatically?

Codebase research runs only when project context is detected (files, modules, references to the current system). Web research always runs.

Can you create ADRs from this session?

Yes. When significant decisions emerge I draft ADR content and will invoke the ADR flow if you confirm creation; ADRs default to Proposed unless you accept them.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational