Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

ideationideation 搜索

Agent Skill

ideation 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 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。

总安装

873

周安装

36

GitHub Stars

公开资料未说明

下载量

285
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:ideation(ideation 搜索)
来源仓库:https://github.com/caldiaworks/caldiaworks-marketplace
仓库路径:skills/ideation
安装命令:
npx skills add https://github.com/caldiaworks/caldiaworks-marketplace --skill ideation
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/caldiaworks/caldiaworks-marketplace --skill ideation

简介

ideation 将粗糙想法转化为结构化文档,通过对话逐步澄清需求与可行性。

  • 适用于产品构思、功能定义与任务规划前期阶段。
  • 通过 npx skills add 命令从 caldiaworks 市场安装,拒绝跳过评估直接实现。
  • 输出为可追踪的 idea 文档,可作为 USDM 或其他需求的输入源。
  • ideation 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Ideation: Ideas Into Structured Documents

Help turn rough ideas into structured, validated idea documents through natural collaborative dialogue. Start by understanding the current project context, then ask questions one at a time to refine the idea. Once the idea is clear, present approaches with feasibility evaluation and get user approval.

Do NOT skip to implementation. The purpose of this skill is to produce a well-thought-out idea document that can feed into requirements definition (e.g., USDM) or be tracked as a task. Even if the idea seems simple, unexamined assumptions cause the most wasted work.

Process

Complete these steps in order:

Step 1: Explore Context

Investigate all information sources referenced in the user's input before proceeding.

1a: External References

Scan the user's input for external references — app names, technology names, URLs, specifications, standards, protocols, services, or documents. If any are found:

  • URLs: Fetch and read the linked content to understand scope, API surface, constraints, or specifications
  • Named technologies/services: Search for official documentation, capabilities, limitations, and integration patterns
  • Existing apps or products: Research their features, behavior, and design to understand what the user is referencing
  • Standards/protocols: Look up the specification to understand requirements and constraints

Do NOT proceed to Step 2 until all external references are investigated. Misunderstanding an external reference propagates errors into every subsequent step.

1b: Project Context

Review the current project to understand what already exists:

  • Check project files, documentation, and recent commits
  • Identify existing solutions, patterns, and conventions
  • Note constraints, dependencies, and technical boundaries

This grounds the ideation in reality rather than starting from a blank slate.

Step 2: Ask Clarifying Questions

Understand the idea through focused dialogue:

  • Ask one question at a time — do not overwhelm with multiple questions
  • Prefer multiple choice questions when feasible — easier to answer than open-ended
  • Focus on understanding: purpose, target users, constraints, success criteria
  • Be ready to revisit earlier questions if new information changes the picture

Continue until you have a clear understanding of what the user wants and why.

Step 3: Propose Approaches

Present 2-3 different approaches with trade-offs:

  • Lead with your recommended approach and explain why
  • Present options conversationally with reasoning
  • Each approach must include a feasibility evaluation:

- Differentiation: How does this differ from existing solutions? - Technical risks and constraints: What could go wrong technically? - Pre-mortem: "If this idea fails, what would be the cause?"

  • Apply YAGNI — remove unnecessary features from all approaches

Step 4: Present and Validate

Once the user selects an approach (or a combination):

  • Present the idea in sections, scaled to complexity
  • Ask after each section whether it looks right so far
  • Be ready to go back and revise if something does not make sense

Step 5: Write Idea Document

After the user approves the idea, generate the document.

Save to: .docs/ideas/YYYY-MM-DD-<topic>.md

The document must contain these sections:

SectionContent
WhatWhat it does, the problem it solves
WhyRationale, user benefit, business value
ScopeIn scope / out of scope boundaries
StakeholdersTarget users, affected systems
ApproachesThe explored alternatives with trade-offs and feasibility evaluation
DecisionThe selected approach and rationale

See templates/idea-document.md for the output template.

These sections are structured to serve as input for requirements definition tools (e.g., USDM):

  • What maps to Requirements
  • Why maps to Reasons
  • Scope maps to Descriptions (context, constraints)
  • Stakeholders informs scope confirmation

Step 6: Output Selection

After writing the idea document, ask the user which output they want:

OptionAction
A: Markdown only (default)Save the idea document to .docs/ideas/ — done
B: Markdown + GitHub IssueSave the document and create a GitHub Issue with the content
C: Markdown + requirementsSave the document and invoke a requirements definition skill (e.g., usdm)

Key Principles

  • One question at a time — do not bundle multiple questions in one message
  • Multiple choice preferred — easier to answer than open-ended when feasible
  • YAGNI ruthlessly — remove unnecessary features from all approaches
  • Explore alternatives — always propose 2-3 approaches before settling
  • Incremental validation — present sections, get approval, then move on
  • Be flexible — go back and clarify when something does not make sense
  • Feasibility over optimism — every approach needs honest risk assessment

Credits

This skill is based on the brainstorming skill from obra/superpowers (MIT License).

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.45%
按下载量换算110

Claude

30.73%
按下载量换算88

Cursor

19.6%
按下载量换算56

Gemini CLI

9.21%
按下载量换算26

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills