← 返回 Skills 市场
868
总下载
0
收藏
3
当前安装
1
版本数
在 OpenClaw 中安装
/install ai-labs-builder
功能描述
AI Labs Builder - Create modern websites, AI applications, dashboards, and automated workflows. Unified system for building production-ready projects with Ne...
使用说明 (SKILL.md)
AI Labs Builder
Unified system for creating modern websites, AI applications, dashboards, and workflows.
Quick Start
# Create a website
ailabs create website my-portfolio --type portfolio
# Create an AI app
ailabs create ai-app my-chatbot --type chat
# Create a dashboard
ailabs create dashboard my-analytics --type analytics
# Create a workflow
ailabs create workflow my-automation --template automation
# Deploy
ailabs deploy my-project --platform vercel
Commands
Website Creation
ailabs create website \x3Cname> [options]
Options:
--type \x3Ctype> portfolio | saas | blog | ecommerce
--style \x3Cstyle> modern | glassmorphism | brutalism | minimal
--components Include shadcn/ui components
--animations Include Framer Motion animations
--seo Include SEO optimization
AI Application
ailabs create ai-app \x3Cname> [options]
Options:
--type \x3Ctype> chat | agent | rag | multimodal
--provider openai | claude | gemini | local
--streaming Enable streaming responses
--memory Enable conversation memory
--tools Enable tool calling
Dashboard
ailabs create dashboard \x3Cname> [options]
Options:
--type \x3Ctype> analytics | admin | personal | monitoring
--widgets Include widget system
--realtime Enable real-time updates
--charts Include chart components
Workflow
ailabs create workflow \x3Cname> [options]
Options:
--template automation | integration | pipeline
--mcp Include MCP server setup
--schedule Add cron scheduling
--webhook Add webhook triggers
Project Types
1. Websites
Portfolio
- Hero section with animations
- Projects showcase
- Skills/Experience timeline
- Contact form
- Blog integration
SaaS Landing
- Feature highlights
- Pricing tables
- Testimonials
- CTA sections
- FAQ accordion
Blog
- MDX support
- Tag/categories
- Search functionality
- RSS feed
- Newsletter signup
2. AI Applications
Chat Interface
- Message history
- Streaming responses
- Code syntax highlighting
- File attachments
- Voice input/output
AI Agent
- Autonomous task execution
- Tool calling
- Memory management
- Multi-step reasoning
- Human-in-the-loop
RAG System
- Document ingestion
- Vector database
- Semantic search
- Source citations
- Context management
3. Dashboards
Analytics
- KPI cards
- Charts (line, bar, pie)
- Date range picker
- Export functionality
- Real-time updates
Admin Panel
- User management
- Role-based access
- CRUD operations
- Audit logs
- Settings panel
Personal Dashboard
- Weather widget
- Calendar integration
- Task management
- Notes/quick capture
- Bookmarks
4. Workflows
Automation
- Scheduled tasks
- Email notifications
- Data processing
- Report generation
Integration
- API connections
- Webhook handling
- Data sync
- Event triggers
Pipeline
- CI/CD automation
- Testing workflows
- Deployment pipelines
- Quality checks
Design System
Colors
/* Modern */
--primary: #6366f1;
--secondary: #8b5cf6;
--accent: #ec4899;
/* Glassmorphism */
--glass-bg: rgba(255, 255, 255, 0.05);
--glass-border: rgba(255, 255, 255, 0.1);
--glass-blur: blur(20px);
/* Brutalism */
--brutal-black: #000;
--brutal-white: #fff;
--brutal-accent: #ff00ff;
Typography
- Headings: Inter, Geist, or JetBrains Mono
- Body: Inter or system-ui
- Code: JetBrains Mono or Fira Code
Components
All projects include:
- Button variants (primary, secondary, ghost, outline)
- Cards (default, hover, glass)
- Forms (input, textarea, select, checkbox, radio)
- Navigation (header, sidebar, breadcrumbs)
- Feedback (toast, alert, modal, tooltip)
- Data (table, pagination, tabs)
AI Integration
OpenAI
import { OpenAI } from 'openai';
const openai = new OpenAI({
apiKey: process.env.OPENAI_API_KEY,
});
// Streaming chat
const stream = await openai.chat.completions.create({
model: 'gpt-4',
messages: [{ role: 'user', content: prompt }],
stream: true,
});
Claude
import Anthropic from '@anthropic-ai/sdk';
const anthropic = new Anthropic({
apiKey: process.env.ANTHROPIC_API_KEY,
});
const message = await anthropic.messages.create({
model: 'claude-3-opus-20240229',
max_tokens: 1024,
messages: [{ role: 'user', content: prompt }],
});
Vector Database (Pinecone/Memory)
import { Pinecone } from '@pinecone-database/pinecone';
const pc = new Pinecone({
apiKey: process.env.PINECONE_API_KEY,
});
// Store embeddings
await index.upsert([{
id: '1',
values: embedding,
metadata: { text: content }
}]);
Deployment
Vercel (Recommended)
ailabs deploy my-project --platform vercel
Features:
- Automatic Git integration
- Preview deployments
- Edge functions
- Analytics
Netlify
ailabs deploy my-project --platform netlify
Features:
- Git-based deployment
- Branch previews
- Form handling
- Edge functions
GitHub Pages
ailabs deploy my-project --platform github
Features:
- Free hosting
- Custom domains
- Jekyll support
- Actions integration
Best Practices
Performance
- Use Next.js Image component
- Implement lazy loading
- Optimize fonts with next/font
- Use React Server Components
SEO
- Meta tags with next/head
- Sitemap generation
- robots.txt
- Structured data
Security
- Environment variables
- Input validation
- CSRF protection
- Content Security Policy
Accessibility
- ARIA labels
- Keyboard navigation
- Color contrast
- Screen reader support
Examples
Create a Portfolio
ailabs create website portfolio \
--type portfolio \
--style glassmorphism \
--components \
--animations \
--seo
Create an AI Chatbot
ailabs create ai-app chatbot \
--type chat \
--provider openai \
--streaming \
--memory
Create an Analytics Dashboard
ailabs create dashboard analytics \
--type analytics \
--widgets \
--realtime \
--charts
Create an Automation Workflow
ailabs create workflow daily-report \
--template automation \
--mcp \
--schedule "0 9 * * *"
Integration with Other Skills
- mcp-workflow: For advanced workflow automation
- gcc-context: For version controlling project context
- agent-reflect: For continuous improvement
Resources
安全使用建议
Summary of what to consider before installing/running this skill:
- Missing/undeclared secrets: The manifest lists no required environment variables, but the scripts and code expect provider API keys (OpenAI, Anthropic/Claude, Pinecone). Do not set or expose high-privilege API keys to this skill until you audit the code and understand where keys are used and stored.
- Inconsistent env names: The code uses different variable names for the same provider (e.g., ANTHROPIC_API_KEY vs CLAUDE_API_KEY). Verify and standardize names to avoid confusion and unintended key leakage.
- Network and package installs: The scripts run npx/create-next-app and npm install many packages. Running the skill will cause network activity and install third-party packages — review package lists and consider running in an isolated environment (container, VM) or offline audit first.
- Missing referenced scripts: The main CLI references deploy.sh and integrate.sh, but those files are not present in the provided manifest. That could break functionality or be a sign the package is incomplete; request the missing files or avoid running commands that call them.
- File-system and command execution scope: The skill's purpose requires writing files and running Node/npm commands. If you allow the agent to run Bash/Node, understand it can execute arbitrary commands on the host. Prefer to run the CLI inside a disposable container or sandbox.
- MCP/stdio server: The workflow generator builds an MCP server that listens on stdio. Understand what exposes and how workflows/tools are registered before enabling network or production use.
Immediate recommendations:
- Inspect the full package (missing files) locally and review package.json and the actual list of npm dependencies before running npm install.
- Run the tool in a sandbox (container/VM) rather than on a sensitive machine.
- Provide least-privilege API keys (scoped tokens) if you want to test AI features; rotate keys after testing.
- Ask the skill author for the missing deploy/integrate scripts and clarification about env var names and what state is saved under .ailabs.
If you want, I can: (1) list the exact locations where environment variables are referenced, (2) extract the npm dependency lists from the scripts, or (3) produce a short checklist and a safe command sequence to run the scaffolder inside a container.
功能分析
Type: OpenClaw Skill
Name: ai-labs-builder
Version: 1.0.0
The skill is suspicious due to critical shell injection vulnerabilities present in multiple core scripts, specifically `scripts/create-ai-app.sh`, `scripts/create-dashboard.sh`, `scripts/create-website.sh`, and `scripts/create-workflow.sh`. In these scripts, the `$PROJECT_NAME` variable, which is derived from user input, is used directly in `mkdir -p "$PROJECT_NAME"` and `cd "$PROJECT_NAME"` commands without proper sanitization. This allows for potential directory traversal or arbitrary command execution on the host system. Additionally, the `assets/js/dashboard-widgets.js` file, intended for generated dashboards, contains client-side XSS vulnerabilities if widget titles or todo items are sourced from untrusted input.
能力评估
Purpose & Capability
The skill's stated purpose is to scaffold websites, AI apps, dashboards and workflows — which explains the included Bash/Node scripts and npm installs. However, metadata declares no required environment variables or credentials while the code and SKILL.md clearly expect provider API keys (OpenAI/Anthropic/Claude, Pinecone, etc.). This mismatch (no declared env requirements but many references in code) is incoherent and increases risk.
Instruction Scope
SKILL.md and the provided scripts instruct the agent to run shell scripts that create projects, call npx/create-next-app and npm install packages, and write many files into the current working directory. Instructions allow Read/Write/Edit/Bash/Node tools (broad file and command access). The instructions do not attempt obvious exfiltration, but they do instruct network operations (npm installs, possible API calls) and create an MCP stdio server in workflow code. Also, some referenced commands/files (deploy.sh, integrate.sh) are invoked by the CLI but are absent from the manifest — a functional inconsistency.
Install Mechanism
No formal install spec is declared (instruction-only), but the included scripts perform live package installs via npm and run npx. That means the skill will pull many packages from public registries at runtime (moderate risk). There are no downloads from unknown URLs in the manifest, but running npm install will fetch third-party packages (traceable but should be reviewed).
Credentials
Although the registry metadata declares no required env vars, the code and SKILL.md reference multiple secrets: OPENAI_API_KEY, CLAUDE_API_KEY / ANTHROPIC_API_KEY (inconsistent naming), PINECONE_API_KEY, and NEXT_PUBLIC_* vars. Requiring these without declaring them is disproportionate and confusing. The skill will encourage use of provider credentials to enable AI features; the user should not provide high-privilege keys without audit or scoping.
Persistence & Privilege
always:false (normal). The skill will write many files and create project directories in the current working directory (expected for a scaffolder). The SKILL.md metadata references a state dir ".ailabs" (stateDirs) — this indicates it may store per-project state. There is no evidence it modifies other skills or system-wide agent settings, but because allowed-tools include Bash and Node, the agent could execute arbitrary shell commands if invoked — that capability is expected here but increases blast radius.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install ai-labs-builder - 安装完成后,直接呼叫该 Skill 的名称或使用
/ai-labs-builder触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Unified system for creating websites, AI applications, dashboards, and workflows with Next.js, TypeScript, Tailwind, and MCP integration
元数据
常见问题
AI Labs Builder 是什么?
AI Labs Builder - Create modern websites, AI applications, dashboards, and automated workflows. Unified system for building production-ready projects with Ne... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 868 次。
如何安装 AI Labs Builder?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install ai-labs-builder」即可一键安装,无需额外配置。
AI Labs Builder 是免费的吗?
是的,AI Labs Builder 完全免费(开源免费),可自由下载、安装和使用。
AI Labs Builder 支持哪些平台?
AI Labs Builder 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 AI Labs Builder?
由 slemo54(@slemo54)开发并维护,当前版本 v1.0.0。
推荐 Skills