By LucivHuang
A Spec-Driven Development (SDD), tech-stack-agnostic, resumable, extensible, semi-automated AI workflow orchestration plugin covering the full journey from requirement specifications to code delivery.
Identify ambiguous areas in feature specification, ask up to 5 clarification questions, encode answers back into specification.
Scan project and interactively generate constitution.md and project-context.json
Execute implementation plan by processing and executing all tasks defined in tasks.md
Execute implementation planning workflow using plan template, generate design artifacts.
Post-workflow refinement command — adjust artifacts, update code, extract and deposit memory by value
Scans ambiguous areas in feature specifications, generates prioritized question queue (≤5) by taxonomy, each question with recommended answer and integration plan
Designs feature architecture by analyzing existing code patterns and conventions, providing comprehensive implementation blueprint with specific file creation/modification checklist, component design, data flow, and build sequence
Deep analysis of existing codebase, tracking implementation paths, mapping architectural layers, understanding patterns and abstractions, documenting dependencies, providing code-level guidance for new feature development
Executes code implementation according to task list in tasks.md, following project conventions and verified code patterns, completing and marking progress task by task
Technology-agnostic code review expert, reviews implementation code based on project specifications and verified patterns, outputs graded issue list and scoring report
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
中文 | English
一个规格驱动开发(SDD)、技术栈无关、支持断点续做、可扩展、半自动,覆盖需求规格到代码交付的 AI 全流程编排插件
runBefore/runAfter 在任意核心阶段前/后插入自定义扩展(如需求获取、E2E 测试、部署等)| 原则 | 具体体现 |
|---|---|
| Spec 是校验协议,不是永久真理 | AI 写 spec 给人审,确认理解一致后推进;不追求 spec 作为系统的 source of truth |
| AI 写、人审、人不维护 | 所有文档产物由 AI 生成和读取,人只在暂停点审查决策,不手动编辑产物 |
| 半自动执行 | 一条命令启动,AI 自动推进,关键决策点(需求澄清、方案选择、任务确认)暂停等人 |
| 一次性工作流 | 每个 feature 分支一个独立生命周期,完成后清理阶段产物;新需求新工作流,不搞累积式 spec |
| 文档 | 说明 |
|---|---|
| 快速开始 | 安装、环境要求、第一次运行 |
| 配置参考 | workflow.json 完整配置说明 |
| Hook Points 扩展 | 在核心阶段前/后插入自定义扩展 |
| 对比与设计哲学 | 了解 CodeSpec 与其他工具的区别,以及它的设计哲学 |
# 添加插件市场
/plugin marketplace add https://github.com/LucivHuang/CodeSpec.git
# 安装插件:
/plugin install codespec@codespec-marketplace
# 运行(首次会自动引导项目初始化)
/codespec:workflow 新增用户个人资料页面功能
环境要求:bash + jq;git 可选但推荐。
详见 快速开始指南。
| 命令 | 用途 |
|---|---|
/codespec:workflow | 完整工作流(一键走完全流程) |
/codespec:specify | 生成功能规格 |
/codespec:clarify | 需求澄清 |
/codespec:plan | 方案设计 |
/codespec:tasks | 任务拆解 |
/codespec:implement | 执行实现 |
/codespec:review | 代码审查 |
/codespec:refine | 完成后微调 |
/codespec:constitution | 项目规范初始化/更新 |
.codespec/
├── config/workflow.json # 用户创建 — 工作流配置(可选)
├── memory/ # 首次初始化 — 项目记忆文件
└── specs/<feature-name>/
├── spec.md # 阶段 1-2 — 功能规格说明
├── plan.md # 阶段 3 — 方案设计
├── tasks.md # 阶段 4 — 任务清单
├── review-report.md # 阶段 6 — 代码审查报告
└── extensions/ # 可选 — 扩展阶段产出
MIT
npx claudepluginhub lucivhuang/codespec --plugin codespecComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Complete developer toolkit for Claude Code