By Jerrylalala
AI-powered development tools. 51 agents, 4 commands, 66 skills, 1 MCP server for code review, research, design, and workflow automation.
Visually compares live UI implementation against Figma designs and provides detailed feedback on discrepancies. Use after writing or modifying HTML/CSS/React components to verify design fidelity.
Iteratively refines UI design through N screenshot-analyze-improve cycles. Use PROACTIVELY when design changes aren't coming together after 1-2 attempts, or when user requests iterative refinement.
Detects and fixes visual differences between a web implementation and its Figma design. Use iteratively when syncing implementation to match Figma specs.
Creates or updates README files following Ankane-style template for Ruby gems. Use when writing gem documentation with imperative voice, concise prose, and standard section ordering.
中文技术文档写作专家。覆盖:技术博客/设计文档/需求文档/架构说明的中文写作。⚠️ 安全提示:生成内容在发布前需人工审查,避免泄露内部 API 路径、安全配置等敏感信息。
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.
Build applications where agents are first-class citizens. Use this skill when designing autonomous agents, creating MCP tools, implementing self-modifying systems, or building apps where features are outcomes achieved by agents operating in a loop.
Run comprehensive agent-native architecture review with scored principles
This skill should be used when writing Ruby gems following Andrew Kane's proven patterns and philosophy. It applies when creating new Ruby gems, refactoring existing gems, designing gem APIs, or when clean, minimal, production-ready Ruby library code is needed. Triggers on requests like "create a gem", "write a Ruby library", "design a gem API", or mentions of Andrew Kane's style.
1: [P/P+][C][G][R] 探索需求,生成需求文档
External network access
Connects to servers outside your machine
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 claimnpx claudepluginhub jerrylalala/compound-engineering --plugin compound-engineeringBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
No model invocation
Executes directly as bash, bypassing the AI model
No model invocation
Executes directly as bash, bypassing the AI model
中文增强版 Claude Code 复合工程插件 — 真实 Agent Teams、分层验证、UI 设计联动,让每一个工程单元都比上一个更容易。
每一个工程单元都应该让后续单元更容易——而不是更难。
复合工程反转了技术债务的积累方向。80% 在规划和审查,20% 在执行:
Resume → 0: Ideate/Ideas → 1: Brainstorm → 2: Plan → 3: Work → 4: Review → 5: Compound → Repeat
| # | 命令 | 一句话 | 常用参数 |
|---|---|---|---|
| — | /ce:resume | 回来了,告诉我现在啥状态 | — |
| 0 | /ce:ideate | 不知道做什么,AI 帮我想方向 | — |
| 0.5 | /ce:ideas | 管理想法停车场,帮我挑一个开始做 | — |
| 1 | /ce:brainstorm | 知道大概要做什么,帮我想清楚 | [P] [P+] [C] [G] [R] |
| 2 | /ce:plan | 想清楚了,拆成可执行的步骤 | [T] |
| 3 | /ce:work | 按计划动手干 | [T] [T+] [V] [V+] [R] |
| 4 | /ce:review | 干完了,帮我审查代码 | [C] [G] [T] |
| 5 | /ce:compound | 审完了,把经验记下来 | — |
ideate = 让 AI 分析项目现状,生成改进方向建议(输出写入 IDEAS.md) ideas = 管理你已有的想法(无参数=选一个开始做,有参数=加新想法)
你的状态是什么?
│
├── 刚打开项目,不记得上次干到哪了
│ └── /ce:resume → 看完摘要后选下一步
│
├── 不知道做什么,想找方向
│ ├── 完全没想法 → /ce:ideate → 生成建议到 IDEAS.md
│ └── 之前记了一些 → /ce:ideas → 选一个
│ └── 选好后 → brainstorm → plan → work → review
│
├── 知道要做什么功能
│ └── /ce:brainstorm 我要做XXX → plan → work → review
│
├── 需求已经很清楚,不需要讨论
│ └── /ce:plan 做XXX功能 → work → review
│
├── 修 Bug
│ ├── 简单 bug → 直接改 → /ce:review
│ └── 复杂 bug → brainstorm 分析 → plan → work → review
│
├── 代码写完了,要审查
│ └── /ce:review [C]
│
└── 刚解决了一个有价值的问题
└── /ce:compound
| 命令 | 用途 |
|---|---|
/ce:doctor | 健康检查:检测 CLI、MCP、认证状态(skill 入口) |
/ce:pr | 创建 PR 并询问是否合并(skill 入口) |
/ce:sync-upstream | 检测上游仓库更新(skill 入口) |
→ 打开交互式工作流图 — 点击每个阶段,展开参数详情 + 使用示例
/plugin marketplace add Jerrylalala/compound-engineering
/plugin install compound-engineering
claude --plugin-dir "/path/to/compound-engineering/plugins/compound-engineering"
CLI 转换命令依赖 Bun。发布 @jerry-jian/compound-plugin 前,先使用已发布的上游 CLI 包并显式指向本公共仓库:
# Windows PowerShell
$env:COMPOUND_PLUGIN_GITHUB_SOURCE="https://github.com/Jerrylalala/compound-engineering"
bunx @every-env/compound-plugin install compound-engineering --to codex --includeSkills
# macOS/Linux
COMPOUND_PLUGIN_GITHUB_SOURCE=https://github.com/Jerrylalala/compound-engineering \
bunx @every-env/compound-plugin install compound-engineering --to gemini --output .
当前 npm 上已发布的是 @every-env/compound-plugin。@jerry-jian/compound-plugin 发布前,请使用上面的环境变量方式指向本公共仓库。
ce:work [T] 使用 Claude Code 原生 Agent Teams,不是角色扮演:
ce:work [V] 按任务类型自动选择验证层,任务完成必须通过全部激活层:
| Layer | 内容 | 触发条件 |
|---|---|---|
| -1.5 | 环境指纹:自动检测启动命令 + 测试命令(npm/pytest/cargo/go test) | V/V+ 启用时 |
| 0 | CLI 构建/语法检查(3 次修复循环) | 始终 |
| 1 | API / 数据库验证 | 后端变更 |
| 2 | 多路由验证:浏览器 UI + TEST_COMMAND + CLI + API + 设计图对比 | 按项目类型自动选择 |
| 3 | 独立验收确认 | 始终 |
[V+] 额外支持零配置 Electron 桌面应用。Layer 2 自动检测项目类型:前端跑 Playwright,后端跑 TEST_COMMAND + curl,CLI 跑参数验证,有设计图时自动对比。失败后最多 3 轮自动修复(诊断修复 → 缩小范围 → 仅诊断报告)。
ce:brainstorm [P] 显式启动 party-mode,用 3 个核心视角快速验证;[P+] 启动 12-14 个视角深度发散。两种模式都会保留 emoji + 角色名 + 相互质疑/补充的讨论体验,并在退出后自动触发探索者+挑战者结构化收敛。
Party Mode 结果会进入 requirements 文档:P1-P4 优先级、共识/分歧/风险、复用机会,以及 Reuse / Build Boundary(哪些复用成熟能力、哪些只写胶水代码、哪些必须自研)。
/ce:review [C][G] 同时调用 Codex 和 Gemini 进行交叉验证,双方共同发现的问题优先级自动提升。
Claude 收到外部 AI 审核结果后,会先出裁决表(✅ 认同 / ⚠️ 调整实施 / ❌ 不适用)供用户确认,再动手修复。Claude 是最终裁判,不照单全收——外部 AI 只看局部,Claude 掌握全局上下文。
本插件内置 UI 设计工作流,支持从设计到实现的完整链路:
.pen 设计文件,生成/修改设计稿,无需切换工具Pencil 设计 → frontend-design 生成实现 → figma-design-sync 对比验证 → 迭代修复
→ Pencil MCP 完整设计联动文档(工具列表 · 工作流 · 使用示例)
Personalized coding tutorials that use your actual codebase for examples with spaced repetition quizzes
Harness-native ECC plugin for engineering teams - 67 agents, 277 skills, 93 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Comprehensive 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.
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.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Complete developer toolkit for Claude Code