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clear审计通过

spec-driven-dev规范驱动的开发

Agent Skill

spec-driven-dev 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

282

周安装

12

GitHub Stars

公开资料未说明

下载量

99
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

3

许可证

MIT

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:spec-driven-dev(规范驱动的开发)
来源仓库:https://github.com/lorenzogirardi/ai-ecom-demo
仓库路径:skills/spec-driven-dev
安装命令:
npx skills add https://github.com/lorenzogirardi/ai-ecom-demo --skill spec-driven-dev
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/lorenzogirardi/ai-ecom-demo --skill spec-driven-dev

简介

用于查找、检索和筛选相关信息,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。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装 spec-driven-dev 技能。
  • 当前无额外说明,需参考原始 SKILL.md 获取完整功能细节。

SKILL.md

ABOUTME: Spec-driven development framework for ecommerce project

ABOUTME: Manages lifecycle: /spec.plan -> /spec.refine -> /spec.tasks -> /spec.run

Spec-Driven Development

Iterative feature development ensuring zero ambiguity before execution.

Quick Reference

CommandPurpose
/spec.plan <intent>Create spec from feature description
/spec.refine [section]Improve spec with research
/spec.clarify <response>Answer clarification questions
/spec.tasksBreak spec into executable tasks
/spec.run [task#]Execute tasks with TDD

Core Principle

Iterate until clarity: No task execution begins until ALL questions are resolved. Claude must be able to execute without interruptions.


Phase 1: /spec.plan - Create Specification

Trigger: /spec.plan <description> or "I want to build/add X"

Workflow

  1. Check specs/ folder - create if missing
  2. Generate spec file: specs/{feature-slug}.md
  3. Fill initial sections from user intent
  4. Generate clarifying questions
  5. STOP and present questions

Output

Created: specs/feature-name.md (DRAFT)

Questions requiring clarification:
1. [Question about scope]
2. [Question about behavior]

Use /spec.clarify to answer.

Phase 2: /spec.refine - Research & Improve

Trigger: /spec.refine [section]

Workflow

  1. Load active DRAFT spec
  2. Search codebase for similar patterns
  3. Update Technical Strategy
  4. Re-evaluate clarity
  5. If questions remain: STOP and present

Phase 3: /spec.clarify - Answer Questions

Trigger: /spec.clarify <response>

Example

User: /spec.clarify Q1: OAuth2 with Google only. Q2: Admin can also delete.

Updated specs/auth-system.md:
- Added OAuth2/Google to Technical Strategy
- Updated permissions

Remaining questions: None
Spec ready. Use /spec.tasks

Phase 4: /spec.tasks - Task Breakdown

Trigger: /spec.tasks

Prerequisites

  • Active spec must be DRAFT or APPROVED
  • "Open Questions" section must be empty
  • If questions exist: STOP → /spec.clarify

Task Granularity

Tasks should be high-level logical units:

  • "Implement authentication middleware"
  • "Create user model and repository"
  • "Add API endpoints for user CRUD"

TDD cycle happens WITHIN each task during /spec.run.


Phase 5: /spec.run - Execute Tasks

Trigger: /spec.run [task#]

Prerequisites

  • Task file must exist: specs/{feature}.tasks.md
  • If no task file: STOP → /spec.tasks

Execution Rules

  • TDD for each task: Red → Green → Refactor → Commit
  • Invoke language skills: /typescript for.ts files
  • Respect hooks: Pre-commit must pass
  • Mark completed in task file

File Structure

specs/
├── README.md                   # Project config
├── user-wishlist.md            # Spec (APPROVED)
├── user-wishlist.tasks.md      # Task breakdown
├── payment-stripe.md           # Spec (DRAFT)
└── ...

Spec Status Flow

DRAFT -> APPROVED -> IN_PROGRESS -> COMPLETED
          |              |
          v              v
      (questions?)   (blocked?)
          |              |
          v              v
        DRAFT      IN_PROGRESS

Ecommerce-Specific Scopes

When creating specs, consider these modules:

ModuleFilesConsiderations
Authapps/backend/src/modules/auth/JWT, sessions, rate limits
Catalogapps/backend/src/modules/catalog/Redis caching
CartFrontend hooks + BackendSession persistence
Ordersapps/backend/src/modules/orders/Transactions
Checkoutapps/frontend/src/app/checkout/Payment flow

Templates

See references/templates.md for:

  • Spec file template
  • Task file template
  • specs/README.md template

Session Resume

On context compaction:

  1. Check specs/ for files with status IN_PROGRESS
  2. Check .tasks.md files for unchecked items
  3. Report: "Found in-progress spec: X with Y tasks remaining"
  4. Ask: "Continue with /spec.run?"

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

Gemini CLI

28.66%
按下载量换算28

Antigravity

24.56%
按下载量换算24

Claude Code

15.81%
按下载量换算16

Codex

12.54%
按下载量换算12

windsurf

7.75%
按下载量换算8

trae

3.14%
按下载量换算3

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。

来源信息

继续浏览同类 Skills