头脑风暴

工程开发 社区

技能解读

当前原文的自动摘译,尚未人工审校。

工艺流程

探索背景→提出问题→提出方法→获得批准→记录计划

1. 理解这个想法

首先检查任何现有的上下文(文件、文档、之前的笔记、最近的工作)预先收集所有澄清问题 - 使用ask_questions、ask或questions工具(如果有);否则纯文本 重点关注:目的、目的、限制、成功标准、时间表

2. 澄清背景

时间表、预算、资源、技术限制 受众是谁?成功是什么样的?

3.探索方法

提出 2-3 种不同的权衡方法 提出您推荐的选项并解释原因

4. 获得批准

逐节提出计划,在继续之前获得批准 根据想法的复杂性扩展深度 涵盖与领域相关的维度:结构、关键决策、风险、后续步骤

流狐档案 运行与权限由流狐静态检查推断,不代表作者声明或实际测试结果。
领域
工程开发
作者声明的智能体
未找到明确声明;不据此推断已兼容或已测试
静态检查
88 / 100 · 启发式扫描,不代表运行安全
运行环境证据
未执行;当前没有隔离运行测试证据
作者 / 版本 / 许可证
@BrikerMan · 未声明许可证
上下文用量评级
低消耗
接入复杂程度
即装即用
是否需要外部接口密钥
不需要
兼容的系统
未声明(默认跨平台)
运行环境要求
未检出明确要求
检测到的文件与系统行为
  • 只读
  • 允许写入 / 修改
网络行为
仅限本地
安装命令数
无(仅作为资料)

需要注意: 未声明工具权限范围;实际权限由运行环境决定。

作者说明文件(原文) SKILL.md

# Brainstorming: From Idea to Plan

**Help turn ideas into fully formed plans through natural collaborative dialogue.**

Works for: product features, business ideas, creative projects, process improvements, technical designs, or any new initiative.

## Process

**Explore context → Ask questions → Propose approaches → Get approval → Document plan**

### 1. Understand the Idea

- Check any existing context first (files, docs, prior notes, recent work)
- Gather all clarifying questions upfront — use `ask_questions`, `ask`, or `questions` tool if available; plain text otherwise
- Focus on: purpose, who it's for, constraints, success criteria, timeline

### 2. Clarify Context

- Timeline, budget, resources, technical constraints
- Who is the audience? What does success look like?

### 3. Explore Approaches

- Propose 2-3 different approaches with trade-offs
- Lead with your recommended option and explain why

### 4. Get Approval

- Present plan section by section, get approval before moving on
- Scale depth to complexity of the idea
- Cover dimensions relevant to the domain: structure, key decisions, risks, next steps

### 5. Document the Plan

- Create a clear, actionable plan based on the agreed direction

<HARD-GATE>
Do NOT start implementing until the user has approved the plan. Even "simple" ideas need clarity first.
</HARD-GATE>

## Principles

- **Batch questions by dependency** — Ask independent questions together in one round; hold dependent ones for the next round after getting answers
- **Use question tools first** — If `ask_questions`, `ask`, or `questions` tools are available, use them; otherwise ask in plain text
- **Multiple choice preferred** — Easier to answer than open-ended when possible
- **Start broad, then narrow** — Understand the big picture first
- **No assumptions** — "Simple" ideas often have hidden complexity
- **YAGNI ruthlessly** — Remove unnecessary scope from all plans
- **Explore alternatives** — Always propose 2-3 approaches before settling
- **Incremental validation** — Present plan section by section, get approval before moving on
- **Be flexible** — Go back and clarify when something doesn't make sense

## Cross-References

- **Need research?** → Invoke `skilless.ai-research` to search the web, read pages, extract video transcripts, or conduct deep multi-source investigations
- **Ready to document the plan?** → Invoke `skilless.ai-writing` to produce a professional, structured design doc or report from the approved plan

讨论

基于 GitHub Discussions。登录 GitHub 即可参与讨论、点赞、订阅更新。