跳到正文
cccskills
免费在 GitHub 公开

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md4.3 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Brainstorming Ideas Into Designs

Overview

Help turn ideas into fully formed designs and specs through natural collaborative dialogue.

Start by understanding the current project context, then ask questions one at a time to refine the idea. Once you understand what you're building, present the design in small sections (200-300 words), checking after each section whether it looks right so far.

The Process

Understanding the idea:

Note: This skill is typically invoked by agents like nimbus-researcher when requirements are unclear. You can also invoke it directly when:

  • Starting a new component and need to explore design options
  • Facing architectural decisions with multiple valid approaches
  • Requirements are ambiguous and need clarification

For complete component creation with OpenSpec integration, use /propose-component command instead.

  • You SHOULD begin by reviewing the current project state (files, docs, recent commits). This ensures the design fits existing patterns and doesn't duplicate work.
  • You SHOULD ask questions one at a time to refine the idea. This creates a focused conversation and helps clarify requirements incrementally.
  • You MAY prefer multiple choice questions when they clarify options quickly, but open-ended questions are fine too—choose based on conversation flow.
  • You SHOULD structure exploration into separate messages if multiple topics need investigation. This prevents overwhelming both parties.
  • You MUST focus on understanding: purpose, constraints, and success criteria.

Exploring approaches:

  • You SHOULD propose 2-3 different approaches with trade-offs. This explores the design space and prevents premature convergence on the first idea.
  • You SHOULD present options conversationally with your recommendation and reasoning. Explain why you're recommending one approach over others.
  • You MUST lead with your RECOMMENDED option and explain the trade-offs. Help the user understand your reasoning before they decide.

Presenting the design:

  • You SHOULD verify you understand the full scope before presenting (not just your initial interpretation). Ask clarifying questions until you can describe what's being built with confidence.
  • You MUST present design in reviewable sections (200-300 words each). This lets the user verify direction incrementally rather than discovering misalignment at the end.
  • You MUST ask after each section whether it looks right so far. If it doesn't, correct the misalignment before moving forward.
  • You MUST cover: architecture, components, data flow, error handling, and testing. Missing any of these creates gaps later.
  • You SHOULD be ready to go back and clarify if something doesn't make sense. This is normal and expected.

After the Design

Next Steps:

  • You have a validated design/plan document
  • User can choose their next step:
    • Create an OpenSpec proposal via /openspec:proposal for formal specification
    • Use the plan for implementation planning
    • Continue refining the design if needed
    • Save the plan for later reference
  • Ask the user what they'd like to do next

Key Principles

  • YAGNI ruthlessly: You MUST remove unnecessary features from all designs. Scope creep at design time is easier to fix than during implementation.

  • Explore alternatives: You SHOULD propose 2-3 approaches with trade-offs before settling on the recommended approach. This prevents the "first idea" bias and helps users understand the design space.

  • Incremental validation: You MUST present design in sections and ask after each one. Discovering fundamental misalignment after 30 minutes of work is wasteful.

  • Be flexible with clarification: You SHOULD be ready to go back and ask clarifying questions when something doesn't make sense. This is normal and expected, not a failure.

  • Focus on patterns: You MUST ensure designs align with established Nimbus patterns and standards (see ./openspec/AGENTS.md for project architecture). Consistency reduces learning curve for implementers.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Manage restore context files so skills can survive /clear and /compact. Use to write, delete, check, or list .agent-restore-context-* files.

中文简介正在准备中。当前显示原文说明。

commercetools/nimbus52026年10月17日 更新

Set up the restore-context hook so skills can survive /clear and /compact. Detects .claude/ vs .agents/ layout automatically.

中文简介正在准备中。当前显示原文说明。

commercetools/nimbus52026年10月17日 更新

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

中文简介正在准备中。当前显示原文说明。

commercetools/nimbus52026年10月17日 更新

Archive a completed change. Use when the user wants to finalize and archive a change after implementation is complete.

中文简介正在准备中。当前显示原文说明。

commercetools/nimbus52026年10月17日 更新

Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.

中文简介正在准备中。当前显示原文说明。

commercetools/nimbus52026年10月17日 更新

Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.

中文简介正在准备中。当前显示原文说明。

commercetools/nimbus52026年10月17日 更新

查看 commercetools 的全部技能

报告此技能的问题