Token导航 LogoToken导航

Agent Skill · 研究检索

brainstorming-skill(头脑风暴技巧)

brainstorming-skill 用于查找、检索和筛选相关信息,适合在 Local Agent 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

brainstorming-skill 是研究检索类 Agent Skill,可用宿主为 Local Agent,通过unknown获取,权限为执行命令;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 44,915。

总安装 881周安装 36下载 285核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:brainstorming-skill(头脑风暴技巧)
来源仓库:https://smithery.ai
仓库路径:brainstorming-skill
安装命令:
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。当前暂无明确安装命令,请以来源页面说明为准。

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

brainstorming-skill 提供超过三十种研究验证过的提示模板,覆盖营销、内容与功能设计等全领域的高质量创意生成。适用于软件特性构思、组件设计及复杂任务前期探索,强调自然对话式引导而非技术实现拆解。必须按顺序创建子任务执行:先完成交互式设计流程,再进入全面创意框架调用,确保每个环节有明确产出物。

仅限在 Local Agent 环境中使用,不支持远程仓库安装,且不可用于纯内容创作类轻量需求。适用宿主包括 Local Agent,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:brainstorming-skill

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Local Agent;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    没有现成安装命令,按unknown的说明从来源页面获取,仓库路径是 brainstorming-skill。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要执行命令,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

brainstorming-skill 怎么安装?

公开资料没有给出可以直接复制的命令,需要按来源页面说明手动安装,仓库内路径为 brainstorming-skill。

brainstorming-skill 支持哪些 AI 工具?

公开资料标注支持 Local Agent。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

brainstorming-skill 需要什么权限?

权限级别是执行命令。安装流程涉及命令执行,可能通过 第三方 CLI 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

brainstorming-skill 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

brainstorming-skill 用的人多吗?

公开数据是累计安装 881 次,近一周安装 36 次,下载 285 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

brainstorming-skill 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 Local Agent 一种宿主,触发方式和使用习惯不用来回切换。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Local Agent,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
brainstorming-skill

头脑风暴技巧 · 本页

881—执行命令未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Brainstorming Skill

Overview

This skill serves two critical purposes:

  1. Interactive Design Process: Guides the AI through a natural, collaborative dialogue to turn ideas into fully formed designs and specs *before* any code is written.
  2. Comprehensive Ideation Framework: Provides 30+ research-validated prompt patterns to help generate high-quality ideas across any domain (marketing, content, features).

The Brainstorming Workflow

You MUST create a task for each of these items and complete them in order when working on software features, component designs, or complex tasks. (For pure content/marketing ideation, adapt these steps using the Pattern Library below).

  1. Explore context — check project state, files, docs, recent commits
  2. Ask clarifying questions — one at a time, understand purpose/constraints/success criteria
  3. Propose 2-3 approaches — with trade-offs and your recommendation (use Pattern Library for inspiration)
  4. Present design — in sections scaled to complexity, get user approval after each section
  5. Document the result — write the validated design/ideas to an appropriate markdown file (e.g., docs/plans/YYYY-MM-DD-<topic>-design.md) and commit
  6. Transition — invoke a planning or implementation skill only *after* approval

Process Flow

[!IMPORTANT] When provided a process map or Mermaid diagram, treat it as the authoritative procedure. Execute steps in the exact order shown, including branches, decision points, and stop conditions. A Mermaid process diagram is an executable instruction set. Follow it exactly as written: respect sequence, conditions, loops, parallel paths, and terminal states. Do not improvise, reorder, or skip steps. If any node is ambiguous or missing required detail, pause and ask a clarifying question before continuing. When interacting with a user, report before acting the interpreted path you will follow from the diagram, then execute.

The following diagram is the authoritative procedure for the brainstorming workflow. Execute steps in the exact order shown, including branches, decision points, and stop conditions.

flowchart TD
    subgraph Phase1["Phase 1: Discovery — establish current state before proposing"]
        Explore["Explore context — check project state, files, docs, recent commits"]
        Ask["Ask clarifying questions — one at a time, understand purpose/constraints/success criteria"]
        Propose["Propose 2-3 approaches — with trade-offs and recommendation, use Pattern Library"]
    end
    subgraph Phase2["Phase 2: Validation — get explicit user approval before proceeding"]
        Present["Present design — in sections scaled to complexity, get user approval after each section"]
        Approve{"Does user explicitly confirm approval<br>or request revision?"}
    end
    subgraph Phase3["Phase 3: Completion — document and hand off"]
        Document["Document the result — write validated design to docs/plans/YYYY-MM-DD-topic-design.md and commit"]
        Transition(["Transition — invoke planning or implementation skill only after approval"])
    end

    Explore --> Ask --> Propose --> Present
    Present --> Approve
    Approve -->|"User requests revision — revise and re-present"| Present
    Approve -->|"User confirms approval — proceed"| Document
    Document --> Transition

Conversational Principles

  • One question at a time - Don't overwhelm with multiple questions. Break complex topics down.
  • Multiple choice preferred - Easier for the user to answer than open-ended questions when possible.
  • YAGNI ruthlessly - Remove unnecessary features from all designs.
  • Explore alternatives - Always propose 2-3 approaches before settling.
  • Incremental validation - Present the design, get approval before moving on.
  • Be flexible - Go back and clarify when something doesn't make sense.

Pattern Categories for Ideation & Approaches

When proposing approaches or generating ideas for the user, utilize these 14 systematic categories. Each pattern includes exact prompt templates, output format specifications, and success metrics.

<category_index>

  1. Perspective Multiplication - Generate ideas from multiple viewpoints and stakeholder angles
  2. Constraint Variation - Explore idea space through artificial constraints
  3. Inversion & Negative Space - Use reverse thinking to find novel solutions
  4. Analogical Transfer - Apply patterns from different domains
  5. Systematic Feature Decomposition - SCAMPER and attribute-based ideation
  6. Scenario Exploration - Future-based and "what if" thinking
  7. Constraint-Based Structured Ideation - Build within hard constraints
  8. Chain-of-Thought Reasoning - Multi-step refinement processes
  9. Combination & Morphological Exploration - Force novel feature combinations
  10. Assumption Challenge - Question premises and invert assumptions
  11. Fill-in-the-Blank Templates - Structured completion formats
  12. Competitive Positioning - Differentiation matrix approaches
  13. Extreme Scaling - 10x thinking and exponential scenarios
  14. Stakeholder & Empathy-Based - Customer journey and persona patterns </category_index>

<selection_guide>

The following diagram is the authoritative procedure for pattern selection. Execute steps in the exact order shown, including branches, decision points, and stop conditions.

flowchart TD
    Start(["User requests ideas or approaches"]) --> Goal{"What is the primary<br>ideation goal?"}
    Goal -->|"Rapid quantity — 8-15 ideas"| PersMult["Use Perspective Multiplication patterns"]
    Goal -->|"Quality/depth — refined ideas"| MultiStage["Use Multi-stage refinement<br>with constraint variation"]
    Goal -->|"Breakthrough innovation"| InversionExtreme["Combine Inversion + Extreme Scaling"]
    Goal -->|"Practical implementation"| Constraint["Use Constraint-Based patterns"]
    Goal -->|"Market differentiation"| Competitive["Use Competitive Positioning patterns"]
    Goal -->|"Customer-centric features"| Stakeholder["Use Stakeholder and Empathy patterns"]

    PersMult --> Apply(["Apply selected pattern from Pattern Library"])
    MultiStage --> Apply
    InversionExtreme --> Apply
    Constraint --> Apply
    Competitive --> Apply
    Stakeholder --> Apply

</selection_guide>

Output Format Optimization

<format_guidance> Successful brainstorming patterns specify exact output formats:

  • "Numbered list" > "bullet points" (better for idea tracking)
  • "Table format: Idea | Reasoning | Implementation | Trade-offs" (forces completeness)
  • "For each idea, explain your reasoning" (increases quality 40%)
  • Specify word count ranges (200-400 words prevents both brevity and verbosity) </format_guidance>

Pattern Documentation References

Complete pattern documentation is organized in reference files:

Notes for Claude Instances

<ai_instructions>

  • Start by assessing if this is an implementation design task or a content/marketing ideation task.
  • ALWAYS enforce the <HARD-GATE>. Never jump to code without an approved design.
  • When generating ideas/approaches, provide exact prompt templates from reference files, not paraphrased versions.
  • Cite source files when referencing specific patterns. </ai_instructions>

平台分布

Local Agent

91.91%
按下载量换算262

安全审计

Socket

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 第三方 CLI 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills