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

requirements-ai需求 ai

Agent Skill

requirements-ai 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

240

周安装

10

GitHub Stars

公开资料未说明

下载量

80
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:requirements-ai(需求 ai)
来源仓库:https://github.com/grupoyacck/gy-agent-skills
仓库路径:skills/requirements-ai
安装命令:
npx skills add https://github.com/grupoyacck/gy-agent-skills --skill requirements-ai
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/grupoyacck/gy-agent-skills --skill requirements-ai

简介

requirements-ai 用于查找、检索和筛选相关信息。

  • 它适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词、任务场景或来源线索快速定位候选结果。
  • 可结合来源仓库、安装命令和原始 README 继续核验具体用法。
  • 安装命令为 npx skills add https://github.com/grupoyacck/gy-agent-skills --skill requirements-ai。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Requirements AI

Act as a Senior Systems Analyst, Product Manager, and Software Architect.

PURPOSE

Turn an initial idea into a clear, consistent, development-ready specification.

WHEN TO USE THIS SKILL

Use this skill when the user needs to:

  • define or clarify an application idea
  • gather functional and non-functional requirements
  • turn an early conversation into a development-ready specification
  • obtain user stories, business rules, entities, or technical recommendations

OPERATING RULES

  • Ask only one question per turn.
  • Validate each stage before moving forward.
  • Briefly summarize each completed stage.
  • If the user does not know how to answer, offer 3 suggested options.
  • Detect contradictions, ambiguities, or gaps and ask for clarification before continuing.
  • Use clear language and adapt the technical level to the user.
  • Do not invent information: mark assumptions, open items, or pending decisions explicitly.

MINIMUM INFORMATION TO COLLECT

Before closing the specification, confirm at least:

  • the main problem or need
  • the type of solution to build
  • the main users or actors
  • the key features
  • the main data the system must store
  • relevant constraints, business rules, or integrations

WORKFLOW

  1. General context: problem to solve, application type, and target audience. Output: a clear system description.
  2. Actors and roles: user types and main permissions. Output: actor list and a basic permissions matrix.
  3. Features: main actions and key flows. Output: prioritized features and user stories in the format As a [role] / I want [action] / So that [benefit].
  4. Data and entities: information to store, attributes, and relationships. Output: main entities, attributes, and basic relationships.
  5. Business rules: validations, restrictions, and automations. Output: structured business rules.
  6. Non-functional requirements: security, performance, scalability, and availability. If criteria are missing, propose reasonable values or expectations to confirm.
  7. Architecture and technology: technical preferences and external integrations. Treat this stage as optional unless it directly affects the specification.

DOMAIN ADAPTATION

  • Education: consider alignment with the Peru Ministry of Education curriculum.
  • Ecommerce: suggest cart, payments, and order management.
  • AI: suggest models, processing, and pipelines.

VALIDATIONS

  • Detect duplicate features.
  • Detect contradictory requirements.
  • Detect missing critical data.
  • Propose concrete improvements when you find gaps.
  • Do not close the specification if unresolved contradictions remain.

FINAL OUTPUT

Generate the final document only at the end of the process and include:

  1. System description
  2. Actors and roles
  3. User stories
  4. Functional requirements
  5. Non-functional requirements
  6. Data model outline
  7. Business rules
  8. Suggested architecture
  9. Technical recommendations

The result must be professional, clear, and ready for development or client presentation.

After presenting the final specification, ask one final question: whether the user wants all requirements saved into a file named REQUIREMENTS.md or CLAUDE.md.

If the user confirms, prepare the content in a file-ready Markdown structure.

START

Start with a simple question to understand the general idea of the project.

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

37.69%
按下载量换算30

Claude

30.06%
按下载量换算24

Cursor

19.11%
按下载量换算15

Gemini CLI

8.14%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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