Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

app-maker

自动完成从需求分析、页面设计、数据库建模、代码生成到可视化调试和一键部署的全栈应用开发全流程。

MIT-0 · Free to use, modify, and redistribute. No attribution required.
0 · 10 · 0 current installs · 0 all-time installs
MIT-0
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill's stated purpose (full-stack app generation, multi-model LLM use, deployment) is consistent with the included scripts and documentation. However the top-level registry metadata in the prompt lists 'Required env vars: none' while SKILL.md, config.example.json, package.json and scripts reference API keys (Claude/Qwen/Gemini/GLM) and a ~/.config/app-maker/models.json config file. This mismatch between declared requirements and actual expectations is an incoherence that should be resolved before trusting automatic installation or execution.
Instruction Scope
SKILL.md instructs the agent/user to create and populate ~/.config/app-maker/models.json (containing API keys), run the provided python scripts (scripts/app_builder.py) and npx commands, and to deploy to external cloud platforms. Those actions align with the skill's purpose, but they involve writing/read­ing a config file in the user's home directory and launching networked operations (LLM calls, deployments). The Quickstart contains a hard-coded local path (C:\Users\52304\.openclaw\workspace) — likely leftover from packaging — which is sloppy and could confuse users.
Install Mechanism
No install spec is provided (instruction-only skill), which means nothing will be silently downloaded or extracted at install time. The risk surface is the included Python scripts which the user runs manually; that is lower-risk than arbitrary remote installs but still requires the user to execute local code.
!
Credentials
The skill expects multiple provider API keys (Anthropic/Claude, Qwen, Gemini, GLM). That is reasonable for a multi-model tool, but the package's top-level summary omitted required env vars. Also the recommended deployment/preview steps and package scripts can prompt users to supply other cloud credentials (Vercel, Alibaba) without declaring them up-front. The SKILL.md and config.example.json encourage storing API keys in a plaintext JSON file under ~/.config — this increases risk of accidental exfiltration or improper permissions if not handled carefully.
Persistence & Privilege
The skill does not request always:true and doesn't modify other skills. It writes its own config under ~/.config/app-maker which is normal for a CLI tool. Autonomous invocation is allowed by default (not a unique privilege here).
What to consider before installing
This skill appears to be what it advertises (a multi-model app generator), but there are noteworthy inconsistencies and privacy concerns you should address before running it: (1) The public metadata claims no required env vars, yet the docs and scripts require multiple LLM API keys — expect to provide CLAUDE/QWEN/GEMINI/GLM keys. (2) The skill asks you to create ~/.config/app-maker/models.json containing API keys in plaintext; prefer using environment variables or a secrets manager, and inspect file permissions. (3) Review scripts/app_builder.py (especially the _call_llm implementation) to confirm which remote endpoints the tool contacts and that your keys are only sent to intended providers. (4) Be cautious with deployment steps — deploying to Vercel/Alibaba may require additional cloud credentials which are not declared; don't provide more secrets than necessary. (5) The Quickstart includes a hardcoded user path (likely benign but sloppy) — check paths before running. If you cannot inspect the code, run it in an isolated/sandboxed environment and avoid storing production keys in the skill's config file. If the author can fix the metadata to list required env vars and document secure key handling, confidence would rise.

Like a lobster shell, security has layers — review code before you run it.

Current versionv1.0.1
Download zip
ai-assistantvk979exzjfhsws0csqpwg7tv4tx831q0bapp-developmentvk979exzjfhsws0csqpwg7tv4tx831q0bclaudevk979exzjfhsws0csqpwg7tv4tx831q0bcode-generationvk979exzjfhsws0csqpwg7tv4tx831q0bdatabase-designvk979exzjfhsws0csqpwg7tv4tx831q0bdeploymentvk979exzjfhsws0csqpwg7tv4tx831q0bfull-stackvk979exzjfhsws0csqpwg7tv4tx831q0blatestvk979exzjfhsws0csqpwg7tv4tx831q0bmulti-modelvk979exzjfhsws0csqpwg7tv4tx831q0bmvpvk979exzjfhsws0csqpwg7tv4tx831q0bnodejsvk979exzjfhsws0csqpwg7tv4tx831q0brapid-prototypingvk979exzjfhsws0csqpwg7tv4tx831q0breactvk979exzjfhsws0csqpwg7tv4tx831q0bvisual-debuggingvk979exzjfhsws0csqpwg7tv4tx831q0b

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

SKILL.md

App Builder - 全栈应用开发技能

功能: 自动完成从页面设计、数据库构建到代码生成与部署的全流程,支持可视化调试

版本: 1.0.0
作者: OpenClaw Workspace
创建日期: 2026-03-16


🎯 核心能力

阶段功能输出
1. 需求分析理解用户需求,生成 PRD需求文档、用户故事
2. 页面设计UI/UX 设计,原型生成Figma 链接、组件树
3. 数据库设计Schema 设计,ER 图SQL/NoSQL Schema、ER 图
4. 代码生成前后端代码自动生成完整项目代码
5. 可视化调试实时预览,断点调试预览链接、调试报告
6. 部署发布一键部署到云平台生产环境 URL

🤖 多模型配置

模型优先级

  1. Claude Code (首选) - 代码生成质量最佳
  2. 通义千问 - 中文理解好,适合国内部署
  3. Gemini - 多模态能力强
  4. 智谱 GLM - 代码生成能力优秀

配置文件

创建 ~/.config/app-maker/models.json:

{
  "default": "claude-code",
  "models": {
    "claude-code": {
      "provider": "anthropic",
      "model": "claude-sonnet-4-20250514",
      "apiKey": "sk-ant-...",
      "priority": 1,
      "capabilities": ["code", "design", "debug"]
    },
    "qwen": {
      "provider": "aliyun",
      "model": "qwen-max",
      "apiKey": "sk-...",
      "priority": 2,
      "capabilities": ["code", "design"]
    },
    "gemini": {
      "provider": "google",
      "model": "gemini-2.0-pro",
      "apiKey": "AIza...",
      "priority": 3,
      "capabilities": ["code", "multimodal"]
    },
    "glm": {
      "provider": "zhipu",
      "model": "glm-4-plus",
      "apiKey": "sk-...",
      "priority": 4,
      "capabilities": ["code"]
    }
  },
  "fallback": true,
  "maxRetries": 3
}

📋 工作流程

阶段 1: 需求分析

输入:用户自然语言描述
输出:
  - prd.md: 产品需求文档
  - user_stories.md: 用户故事地图
  - features.json: 功能列表 (优先级排序)

步骤:
  1. 提取核心功能需求
  2. 识别目标用户群体
  3. 定义 MVP 范围
  4. 生成验收标准

阶段 2: 页面设计

输入:PRD 文档
输出:
  - wireframes/: 线框图
  - components.json: 组件树
  - design_system.json: 设计规范

步骤:
  1. 生成用户流程图
  2. 创建页面线框图
  3. 设计组件层次结构
  4. 定义设计规范 (颜色、字体、间距)

阶段 3: 数据库设计

输入:功能列表、数据需求
输出:
  - schema.sql / schema.prisma: 数据库 Schema
  - er_diagram.png: ER 图
  - migrations/: 迁移文件

步骤:
  1. 识别实体和关系
  2. 设计表结构
  3. 定义索引和约束
  4. 生成迁移脚本

阶段 4: 代码生成

输入:设计文档、Schema
输出:
  - src/frontend/: 前端代码 (React/Vue/Flutter)
  - src/backend/: 后端代码 (Node.js/Python/Go)
  - src/api/: API 定义 (OpenAPI/GraphQL)
  - tests/: 测试文件

步骤:
  1. 初始化项目结构
  2. 生成数据模型层
  3. 生成 API 层
  4. 生成 UI 组件
  5. 生成业务逻辑
  6. 生成单元测试

阶段 5: 可视化调试

输入:生成的代码
输出:
  - preview_url: 实时预览链接
  - debug_report.md: 调试报告
  - hot_reload: 热重载支持

步骤:
  1. 启动开发服务器
  2. 生成预览环境
  3. 运行自动化测试
  4. 捕获并修复错误
  5. 生成调试报告

阶段 6: 部署发布

输入:通过测试的代码
输出:
  - production_url: 生产环境 URL
  - deploy_log.md: 部署日志
  - monitoring_dashboard: 监控面板

步骤:
  1. 构建生产版本
  2. 配置 CI/CD 流水线
  3. 部署到云平台
  4. 配置域名和 SSL
  5. 设置监控和告警

🛠️ 技术栈选择

前端框架

场景推荐备选
Web 应用React + TypeScriptVue 3 / Next.js
移动应用FlutterReact Native
桌面应用ElectronTauri
小程序原生Uni-app

后端框架

场景推荐备选
API 服务Node.js (Express/Nest)Python (FastAPI)
实时应用Go + WebSocketNode.js + Socket.io
数据密集Python + DjangoJava + Spring
高并发GoRust

数据库

场景推荐备选
关系型PostgreSQLMySQL
文档型MongoDBFirestore
缓存RedisMemcached
搜索ElasticsearchMeilisearch

部署平台

场景推荐备选
Web 托管Vercel / NetlifyCloudflare Pages
容器部署Docker + K8sECS / Fargate
ServerlessAWS LambdaCloud Functions
国内部署阿里云 / 腾讯云华为云

📁 项目结构模板

my-app/
├── .clawhub/
│   └── app-config.json      # 应用配置
├── docs/
│   ├── prd.md               # 产品需求文档
│   ├── user_stories.md      # 用户故事
│   ├── api_docs.md          # API 文档
│   └── deploy_guide.md      # 部署指南
├── design/
│   ├── wireframes/          # 线框图
│   ├── components.json      # 组件树
│   └── design_system.json   # 设计规范
├── database/
│   ├── schema.prisma        # 数据库 Schema
│   ├── er_diagram.png       # ER 图
│   └── migrations/          # 迁移文件
├── src/
│   ├── frontend/            # 前端代码
│   │   ├── components/
│   │   ├── pages/
│   │   ├── hooks/
│   │   └── styles/
│   ├── backend/             # 后端代码
│   │   ├── controllers/
│   │   ├── models/
│   │   ├── routes/
│   │   └── middleware/
│   └── shared/              # 共享代码
│       ├── types/
│       └── utils/
├── tests/
│   ├── unit/                # 单元测试
│   ├── integration/         # 集成测试
│   └── e2e/                 # 端到端测试
├── infra/
│   ├── docker/              # Docker 配置
│   ├── k8s/                 # K8s 配置
│   └── terraform/           # IaC 配置
├── .env.example             # 环境变量示例
├── package.json
├── tsconfig.json
└── README.md

🚀 快速开始

1. 初始化新项目

# 创建新项目
npx app-builder init my-app

# 或使用交互式向导
npx app-builder create

2. 配置模型

# 设置首选模型
npx app-builder config set model claude-code

# 配置 API Key
npx app-builder config set claude-api-key sk-ant-...

3. 生成应用

# 从自然语言描述生成
npx app-builder generate "创建一个任务管理应用,支持团队协作、看板视图、截止日期提醒"

# 或从 PRD 文件生成
npx app-builder generate --from docs/prd.md

4. 预览和调试

# 启动预览服务器
npx app-builder preview

# 运行测试
npx app-builder test

# 可视化调试
npx app-builder debug --visual

5. 部署

# 部署到 Vercel
npx app-builder deploy vercel

# 部署到阿里云
npx app-builder deploy aliyun

# 自定义部署
npx app-builder deploy --config infra/deploy.json

🔧 命令行参考

核心命令

命令描述示例
init初始化新项目app-builder init my-app
generate生成应用代码app-builder generate "描述"
preview启动预览服务器app-builder preview --port 3000
debug可视化调试app-builder debug --visual
deploy部署应用app-builder deploy vercel
config配置管理app-builder config set model claude-code

生成选项

app-builder generate <description> [options]

选项:
  --framework <name>     指定框架 (react/vue/flutter)
  --backend <name>       指定后端 (node/python/go)
  --database <name>      指定数据库 (postgres/mongo)
  --template <name>      使用模板 (saas/ecommerce/social)
  --from <file>          从文件生成 (PRD/设计稿)
  --output <dir>         输出目录
  --dry-run              预览生成内容,不写入文件

调试选项

app-builder debug [options]

选项:
  --visual               启用可视化调试界面
  --hot-reload           启用热重载
  --breakpoints <file>   设置断点文件
  --logs                 实时日志输出
  --performance          性能分析

📊 可视化调试界面

功能特性

  1. 实时预览: 查看应用运行效果
  2. 组件树: 可视化组件层次结构
  3. 状态检查: 检查应用状态和数据流
  4. 网络监控: 查看 API 请求和响应
  5. 性能分析: 识别性能瓶颈
  6. 错误追踪: 捕获和定位错误

界面布局

┌─────────────────────────────────────────────────────────┐
│  🔍 App Builder Debug Console                           │
├─────────────┬─────────────────────┬─────────────────────┤
│             │                     │                     │
│  组件树     │    实时预览         │   属性面板          │
│             │                     │                     │
│  - App      │   ┌───────────┐     │  选中组件:        │
│  - Layout   │   │           │     │  - 名称            │
│  - Header   │   │  Preview  │     │  - Props           │
│  - Main     │   │           │     │  - State           │
│  - Footer   │   │           │     │  - Events          │
│             │   └───────────┘     │                     │
│  数据流     │                     │                     │
│  - Store    ├─────────────────────┴─────────────────────┤
│  - API      │  📊 性能  │  🌐 网络  │  ⚠️ 错误  │  📝 日志 │
│             │                                             │
└─────────────┴─────────────────────────────────────────────┘

🔐 安全最佳实践

代码安全

  • 使用参数化查询防止 SQL 注入
  • 实现输入验证和清理
  • 使用 HTTPS 加密通信
  • 实现适当的认证和授权
  • 存储敏感信息使用环境变量

数据安全

  • 加密存储敏感数据
  • 实现数据备份策略
  • 遵守数据保护法规 (GDPR 等)
  • 实施访问控制和审计日志

部署安全

  • 使用最小权限原则
  • 定期更新依赖
  • 启用安全扫描
  • 配置 WAF 和 DDoS 防护

📈 性能优化

前端优化

{
  "codeSplitting": true,
  "lazyLoading": true,
  "imageOptimization": true,
  "caching": {
    "strategy": "stale-while-revalidate",
    "maxAge": 3600
  },
  "bundleAnalysis": true
}

后端优化

{
  "databaseIndexing": true,
  "queryOptimization": true,
  "caching": {
    "redis": true,
    "ttl": 300
  },
  "connectionPooling": true,
  "rateLimiting": {
    "enabled": true,
    "requests": 100,
    "window": 60
  }
}

🧩 插件系统

内置插件

插件功能状态
@app-builder/reactReact 代码生成
@app-builder/vueVue 代码生成
@app-builder/flutterFlutter 代码生成
@app-builder/prismaPrisma ORM 支持
@app-builder/vercelVercel 部署
@app-builder/aliyun阿里云部署

开发自定义插件

# 创建插件模板
npx app-builder plugin create my-plugin

# 插件结构
my-plugin/
├── src/
│   ├── generator.ts    # 代码生成器
│   ├── templates/      # 模板文件
│   └── index.ts        # 插件入口
├── package.json
└── plugin.json         # 插件配置

📝 示例项目

示例 1: 任务管理应用

npx app-builder generate "创建一个任务管理应用,包含以下功能:
- 用户认证 (登录/注册)
- 项目管理 (创建、编辑、删除)
- 任务看板 (待办、进行中、已完成)
- 团队协作 (邀请成员、分配任务)
- 截止日期提醒
- 文件附件
- 活动日志

技术栈:React + Node.js + PostgreSQL
部署目标:Vercel + Supabase"

示例 2: 电商应用

npx app-builder generate "创建一个电商应用:
- 商品展示和搜索
- 购物车和结算
- 用户评价
- 订单管理
- 支付集成 (支付宝/微信)
- 物流追踪
- 后台管理

技术栈:Next.js + Python + MongoDB
部署目标:阿里云"

🆘 常见问题

Q: 如何切换模型?

# 查看可用模型
npx app-builder config list

# 切换默认模型
npx app-builder config set default-model qwen

# 临时使用其他模型
npx app-builder generate "..." --model gemini

Q: 如何自定义生成的代码风格?

创建 .apprc 配置文件:

{
  "codeStyle": {
    "language": "typescript",
    "semi": true,
    "singleQuote": true,
    "trailingComma": "es5"
  },
  "architecture": {
    "pattern": "mvvm",
    "stateManagement": "zustand"
  }
}

Q: 如何集成现有项目?

# 导入现有项目
npx app-builder import ./existing-project

# 增强现有代码
npx app-builder enhance --target ./src

📚 资源链接


最后更新:2026-03-16

Files

14 total
Select a file
Select a file to preview.

Comments

Loading comments…