Token导航 LogoToken导航

Agent Skill · 研究检索

feature-requirements(功能要求)

feature-requirements 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。

feature-requirements 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;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 42,402。

总安装 1,077周安装 44GitHub Stars 32下载 348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。
GitHub
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:feature-requirements(功能要求)
来源仓库:https://github.com/steveclarke/dotfiles
仓库路径:skills/feature-requirements
安装命令:
npx skills add https://github.com/steveclarke/dotfiles --skill feature-requirements
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/steveclarke/dotfiles --skill feature-requirements

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。

feature-requirements 提供结构化命令引导创建功能需求文档,涵盖需求发现、范围界定和生成统一文档。适用于 Codex、Claude、Cursor、Gemini CLI,适合在功能开发初期梳理需求时使用。依据项目内的功能开发指南和愿景文档进行指导,确保符合既定流程。

安装前需确认权限范围和维护状态,注意可能触发文件读写和本地文档访问。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。

来源仓库内路径:skills/feature-requirements

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 Codex、Claude、Cursor、Gemini CLI;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

    可以直接复制的命令是 npx skills add https://github.com/steveclarke/dotfiles --skill feature-requireme…;命令里的占位符换成自己的路径或密钥再执行。

  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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

feature-requirements 怎么安装?

在终端执行 npx skills add https://github.com/steveclarke/dotfiles --skill feature-requirements。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

feature-requirements 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。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。

feature-requirements 需要什么权限?

权限级别是只读。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。

feature-requirements 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。

feature-requirements 用的人多吗?

公开数据是累计安装 1,077 次,近一周安装 44 次,下载 348 次,GitHub 32 星。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。

feature-requirements 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

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 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 Codex,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。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 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
feature-requirements

功能要求 · 本页

1,07732只读未明确通过

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

Requirements

Overview

Structured command to help create a requirements document by guiding through requirements discovery, practical scoping, and generating a single, consolidated requirements document with clear, practical language.

Create a requirements document following the established feature development process. First, read the feature development guide at @~/.local/share/dotfiles/ai/guides/feature-development-process.md and the vision document at @vision.md to understand the feature scope and documentation approach.

Note: If the project has a local copy of the guide at @project/guides/feature-development-process.md, you may reference that instead for project-specific modifications.

IMPORTANT: Start by creating a TODO list to track the 3 phases of requirements creation. Mark the first phase as "in_progress" and others as "pending".

Requirements Document Structure

Single Consolidated Document (requirements.md) includes:

User Stories & Business Needs:

  • User stories structured by role
  • Business capabilities and user workflows
  • Core system capabilities the system must provide

Business Data & Rules:

  • Business entities and data requirements (entities, relationships, business-level attributes and constraints)
  • Business logic and rules that govern system behavior
  • User experience requirements

Quality Attributes & Constraints:

  • Performance, security, reliability, and scalability requirements
  • Compliance and privacy controls
  • Technical constraints and limitations

Integration & Success:

  • Integration touchpoints with existing systems
  • Development and testing approaches
  • Success criteria and acceptance criteria
  • Constraints and assumptions

Requirement Identification Format

All requirements must use systematic ID prefixes for easy cross-referencing in specs, plans, and discussions.

PrefixUsageNumbering
US-X.X.XUser StoriesSection.Subsection.Item
REQ-X.X.XRequirementsSection.Subsection.Item
SC-XSuccess CriteriaSequential number

Format Examples:

**US-1.1.1**: As a staff member, I want to view custom field data...

**REQ-2.1.1**: The system must render form controls dynamically based on...

**SC-1**: Staff can view custom field values on detail pages.

Numbering Rules:

  • Numbers follow document section hierarchy (section 2.1 → REQ-2.1.x)
  • User stories in section 1.1 use US-1.1.1, US-1.1.2, etc.
  • Requirements in section 3.2 use REQ-3.2.1, REQ-3.2.2, etc.
  • Success criteria use simple sequential numbers (SC-1 through SC-N)

Existing System Patterns to Consider

When gathering requirements, check if existing system capabilities apply:

  • Search: Full-text search capabilities
  • File Attachments: Document/file upload and management
  • Audit Trail: Change tracking and history
  • Multi-tenancy: Data isolation per organization/tenant
  • Permissions: Role-based access control
  • Tagging: Classification and categorization systems
  • Custom Fields: Extensible data fields

Ask early: "Which of these existing capabilities might apply to this feature?"

Process: Work Through These 3 Phases Sequentially

Phase 1: Requirements Discovery

Ask focused questions to gather ALL requirements (don't categorize yet):

  • What specific capabilities does the system need to provide?
  • What business entities does it need to manage (main concepts, relationships, business-level attributes)?
  • What data does it need to store and what are the business constraints?
  • What business rules need to be enforced?
  • What are the different audiences (staff, members, public, admins)?
  • How do requirements differ across these audiences?
  • What should each audience be able to see and do?
  • How do users interact with the system?
  • What happens when there's no data? (Empty states for staff vs members)
  • How is the data presented? (Feed-style, table, cards, etc.)
  • What makes this feature usable for all technical skill levels?
  • What are the security and permission requirements?
  • What performance expectations exist?
  • What integrations are needed with existing systems?
  • What compliance or privacy requirements apply?

ONE QUESTION AT A TIME - Wait for answers before proceeding.

Phase 2: Practical Focus & Future Planning

For each discovered requirement, ask:

  • Is this immediately practical for the current scope?
  • Should this go in current requirements or future.md for later phases?
  • Can we state this in clear, simple language?
  • Is this appropriately scoped for the current phase?
  • Is this testable and measurable?

Create future.md for requirements that are:

  • Great ideas but beyond current scope
  • Complex features requiring Phase 2/3 implementation
  • Ideas worth preserving for future consideration
  • Advanced features that build on the initial foundation

Keep in current requirements only what is:

  • Immediately practical and achievable
  • Essential for core functionality
  • Written in clear, simple language

Phase 3: Document Generation

  • Apply ID prefixes to all user stories (US-), requirements (REQ-), and success criteria (SC-) following the Requirement Identification Format above
  • Create a single consolidated requirements.md document with the following structure:

1. User Stories & Business Needs - structured by role with clear business capabilities 2. Core System Capabilities - what the system must provide to users 3. Business Data Requirements - entities, relationships, business-level attributes 4. Business Rules & Logic - rules and constraints that govern behavior 5. User Experience Requirements - how the system should feel and behave 6. Quality Attributes & Constraints - performance, security, reliability, scalability 7. Integration Touchpoints - connections with existing systems 8. Development & Testing Approach - how the system will be built and validated 9. Success Criteria - clear acceptance criteria and definition of success 10. Constraints & Assumptions - scope boundaries and foundational assumptions

  • Create future.md for deferred requirements (organized by user stories and technical phases)
  • Update or create discussion-summary.md:

- Check if discussion-summary.md exists in the feature directory - If exists: Add a "Requirements Phase" section with: - Technical discussions and research from requirements phase - "Resolved Questions from Vision" subsection (table showing what was open and how resolved) - Update Key Decisions Log with new decisions (continuing numbering from vision) - Update Technical Context with newly referenced files - If not exists: Create it with full structure including Vision Phase placeholder

  • Include table of contents with proper links in all documents
  • Include systematic section numbering (1.1, 1.2, 2.1, 2.2, etc.) for precise cross-referencing
  • Cross-reference the vision document for business context
  • Use clear, practical language throughout
  • Follow the exact structure from our feature development process
  • When documenting key architectural decisions (separate page vs modal, feed vs table, etc.), capture the rationale (linkability, performance, UX, etc.)

Guidelines

  • Ask ONE question at a time and wait for response
  • Update your TODO list as you complete each phase
  • Keep language clear and conversational, not corporate-speak
  • Focus on immediately useful requirements, avoid over-engineering
  • Organize requirements in logical, readable structure that serves the whole team
  • Reference vision document and feature development process
  • Stay practical and focus on readable, actionable documents
  • Stay at Requirements Level: Focus on WHAT and WHY, not HOW

- Good: "System must track whether event is draft or published" - Avoid: "Use boolean flag for is_draft field" - Good: "Events belong to one organization" - Avoid: "Include organization_id foreign key" - Save technical implementation details (data types, table structures, field names) for the specification phase

  • Expect Iteration: Requirements often get refined as users think through implications. This is normal and healthy. Be ready to update requirements based on new insights.
  • When requirements change significantly (optional→required, add→remove features), update ALL related sections consistently

Getting Started Process

  1. Create your 3-phase TODO list immediately
  2. Ask user: "What feature directory are you working in? Please provide the full path (e.g., project/features/FT033-feature-name)"
  3. Ask user: "Please tag your vision document with @vision.md so I can understand the feature scope"
  4. Validate the directory exists and confirm where you'll create the requirements document
  5. Review the vision document to understand feature context
  6. Begin Phase 1 with the first requirements discovery question
  7. Keep questions focused and wait for answers before proceeding to next questions

平台分布

Codex

37.12%
按下载量换算129

Claude

26.42%
按下载量换算92

Cursor

20.11%
按下载量换算70

Gemini CLI

9.45%
按下载量换算33

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。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