OpenClaw Complete Guide 2026: Your Personal Open-Source AI Assistant

OpenClaw Features:
├── 🌐 Browser Automation
│   ├── Web scraping and data extraction
│   ├── Form filling and submission
│   └── Multi-page navigation
├── 📁 File Management
│   ├── Read, write, and organize files
│   ├── Code execution and editing
│   └── Directory management
├── ⏰ Task Scheduling
│   ├── Cron-like reminders
│   └── Automated workflows
├── 💬 Multi-Platform Integration
│   ├── Telegram
│   ├── Discord
│   ├── Feishu (飞书)
│   ├── DingTalk (钉钉)
│   └── WhatsApp
├── 🧠 Persistent Memory
│   ├── User preference recall
│   └── Context continuity
└── 🔌 Plugin System
    ├── Community plugins
    └── Custom skill development

OpenClaw Architecture:
┌─────────────────────────────────────────────────────┐
│                   Communication Layer               │
│  ┌─────────┐ ┌─────────┐ ┌─────────┐ ┌─────────┐   │
│  │Telegram │ │ Discord │ │ Feishu  │ │  WhatsApp│   │
│  └────┬────┘ └────┬────┘ └────┬────┘ └────┬────┘   │
└───────┼───────────┼───────────┼───────────┼────────┘
        │           │           │           │
        ▼           ▼           ▼           ▼
┌─────────────────────────────────────────────────────┐
│                    Gateway Layer                    │
│    (Message routing, authentication, rate limiting)  │
└─────────────────────────┬───────────────────────────┘
                          │
                          ▼
┌─────────────────────────────────────────────────────┐
│                    AI Brain Layer                   │
│    ┌─────────────┐    ┌─────────────────────────┐  │
│    │  LLM Engine │◄──►│  Task Planning & Logic  │  │
│    │  (OpenAI,   │    │  - Action decomposition │  │
│    │   Claude,   │    │  - Tool selection       │  │
│    │   Qwen,     │    │  - Execution planning   │  │
│    │   Local)    │    └─────────────────────────┘  │
│    └─────────────┘                                  │
└─────────────────────────┬───────────────────────────┘
                        │
        ┌───────────────┼───────────────┐
        ▼               ▼               ▼
┌───────────────┐ ┌───────────────┐ ┌───────────────┐
│   Browser    │ │  File System  │ │  System Cmd   │
│   Controller │ │   Operator    │ │   Executor    │
│              │ │               │ │               │
│ - Puppeteer  │ │ - Read/Write  │ │ - Shell       │
│ - Playwright │ │ - Code Exec   │ │ - Docker      │
│ - Web Scrap  │ │ - Git Ops     │ │ - API Calls   │
└───────────────┘ └───────────────┘ └───────────────┘

Technology Stack:
  Frontend:
    - Vue.js 3 (Admin Dashboard)
    - Canvas for screenshots
    - A2UI for agent interaction
  
  Backend:
    - Node.js (Primary runtime)
    - Express.js (Web framework)
    - WebSocket (Real-time communication)
  
  AI Integration:
    - OpenAI API
    - Anthropic Claude
    - Ollama (Local models)
    -火山引擎 (Doubao)
    - Qwen (Local/Cloud)
  
  Automation:
    - Puppeteer/Playwright (Browser control)
    - Node.js child_process (Shell commands)
    - fs/path (File operations)

Minimum Requirements:
  - Node.js 18+
  - 4GB RAM (8GB recommended)
  - 20GB Storage
  - Linux/Windows/macOS
  
Optional:
  - Docker
  - GPU (for local models)

# Create working directory
mkdir openclaw && cd openclaw

# Create docker-compose.yml
cat > docker-compose.yml << 'EOF'
version: '3.8'

services:
  openclaw:
    image: openclaw/openclaw:latest
    container_name: openclaw
    ports:
      - "3000:3000"
    environment:
      - NODE_ENV=production
      - OPENCLAW_PLATFORM=feishu  # or telegram, discord, etc.
      - OPENCLAW_LLM_PROVIDER=openai
      - OPENCLAW_OPENAI_API_KEY=your-api-key
    volumes:
      - ./data:/app/data
      - ./logs:/app/logs
    restart: unless-stopped
    network_mode: host
EOF

# Start the container
docker-compose up -d

# Check logs
docker logs -f openclaw

# Install Git
sudo apt update
sudo apt install git -y

# Install Node.js (using nvm)
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.0/install.sh | bash
source ~/.bashrc
nvm install 20
nvm use 20

# Clone OpenClaw repository
git clone https://github.com/openclaw/openclaw.git
cd openclaw

# Install dependencies
npm install

# Copy configuration
cp .env.example .env

# Edit configuration
nano .env

# .env configuration
# Platform: feishu, telegram, discord, whatsapp, dingtalk
OPENCLAW_PLATFORM=feishu

# LLM Provider: openai, anthropic, ollama, qwen
OPENCLAW_LLM_PROVIDER=openai
OPENCLAW_OPENAI_API_KEY=sk-xxx
OPENCLAW_OPENAI_MODEL=gpt-4

# Browser automation
OPENCLAW_BROWSER_HEADLESS=true
OPENCLAW_BROWSER_TIMEOUT=30000

# Memory and persistence
OPENCLAW_MEMORY_TYPE=redis
OPENCLAW_REDIS_URL=redis://localhost:6379

# Admin
OPENCLAW_ADMIN_USERS=user1,user2

# Feishu Configuration
# 1. Create app at https://open.feishu.cn/
# 2. Get App ID and Secret
# 3. Enable webhook and IM features

OPENCLAW_PLATFORM=feishu
OPENCLAW_FEISHU_APP_ID=cli_xxx
OPENCLAW_FEISHU_APP_SECRET=xxx
OPENCLAW_FEISHU_VERIFICATION_TOKEN=xxx
OPENCLAW_FEISHU_ENCRYPT_KEY=xxx

# Telegram Configuration
# 1. @BotFather to create bot
# 2. Get bot token
# 3. Configure webhook

OPENCLAW_PLATFORM=telegram
OPENCLAW_TELEGRAM_BOT_TOKEN=xxx
OPENCLAW_TELEGRAM_ADMIN_IDS=123456789

# Discord Configuration
# 1. Create application at https://discord.com/developers/
# 2. Add bot to server
# 3. Get token

OPENCLAW_PLATFORM=discord
OPENCLAW_DISCORD_BOT_TOKEN=xxx
OPENCLAW_DISCORD_GUILD_ID=xxx
OPENCLAW_DISCORD_CHANNEL_IDS=xxx

# After deployment, you can chat with OpenClaw:

User: "Search for the latest AI news"
OpenClaw: [Searches the web and returns latest AI news]

User: "Create a file called meeting.md with today's agenda"
OpenClaw: [Creates the file in your specified directory]

User: "Remind me to call John at 3pm today"
OpenClaw: [Sets a reminder notification]

User: "Browse Product Hunt and summarize top 5 products"
OpenClaw: [Opens browser, navigates to Product Hunt, extracts data]

// Programmatic interaction via API
const OpenClawAPI = require('openclaw-api');

const client = new OpenClawAPI({
    baseUrl: 'http://localhost:3000',
    apiKey: 'your-api-key'
});

// Send message
await client.sendMessage({
    platform: 'telegram',
    chatId: '123456789',
    message: 'Summarize the latest tech news'
});

// Execute automation workflow
await client.executeWorkflow({
    name: 'research',
    params: {
        topic: 'OpenClaw AI assistant',
        depth: 'detailed',
        outputFormat: 'markdown'
    }
});

// OpenClaw can control browser for complex tasks

// Example 1: Data Extraction
"Extract all job listings from LinkedIn for 'AI Engineer'"

// Example 2: Form Filling
"Fill out the contact form on example.com with my info from profile"

// Example 3: Multi-page Navigation
"Browse through pages 1-5 of news site and extract headlines"

// my-custom-skill/index.js
module.exports = {
    name: 'my-custom-skill',
    version: '1.0.0',
    description: 'Custom skill for specific tasks',
    
    // Trigger phrases
    triggers: [
        'my skill',
        'custom action'
    ],
    
    // Skill logic
    async execute(context) {
        const { message, args, user } = context;
        
        // Your custom logic here
        const result = await performCustomAction(args);
        
        return {
            success: true,
            response: `Action completed: ${result}`
        };
    },
    
    // Required permissions
    permissions: [
        'browser',
        'file:read',
        'shell'
    ]
};

// config/skills.js
module.exports = {
    skills: [
        require('./skills/web-search'),
        require('./skills/file-manager'),
        require('./skills/reminder'),
        require('./skills/code-runner'),
        require('./skills/my-custom-skill')  // Your custom skill
    ]
};

# Install Ollama
curl -fsSL https://ollama.com/install.sh | sh

# Pull a model
ollama pull qwen:7b
ollama pull llama3:8b

# Configure OpenClaw to use Ollama
OPENCLAW_LLM_PROVIDER=ollama
OPENCLAW_OLLAMA_BASE_URL=http://localhost:11434
OPENCLAW_OLLAMA_MODEL=qwen:7b

Ollama Comparison:
  Model        Parameters  Use Case
  ─────────────────────────────────────────
  Llama 3      8B         General purpose
  Qwen 2.5     7B-72B     Coding, reasoning
  Mistral      7B         Fast, efficient
  CodeLlama    7B-34B     Code generation

LLM Provider Comparison:
  Provider     Cost/M1K Tokens  Speed    Best For
  ─────────────────────────────────────────────────────
  OpenAI       $0.01-0.03      Fast     Production
  Anthropic    $0.015-0.075    Fast     Reasoning
  Ollama       $0 (local)       Medium   Privacy/Cost
  Qwen Cloud   $0.001-0.005    Fast     Chinese language

Security Settings:
  # Rate limiting
  RATE_LIMIT_ENABLED=true
  RATE_LIMIT_WINDOW=60000  # 1 minute
  RATE_LIMIT_MAX=10
  
  # Access control
  ALLOWED_USERS=user1,user2
  ADMIN_ONLY_COMMANDS=shell,exec
  
  # Sandbox
  SHELL_TIMEOUT=30000
  BROWSER_SANDBOX=true
  
  # Audit
  AUDIT_LOG_ENABLED=true
  AUDIT_LOG_PATH=/app/logs/audit.log

Security Checklist:
✓ Use strong API keys
✓ Enable rate limiting
✓ Restrict allowed users
✓ Enable audit logging
✓ Use sandbox for browser
✓ Regular updates
✓ Backup configuration
✓ Monitor resource usage

# Enable debug logging
DEBUG=openclaw:* npm start

# Check specific module
DEBUG=openclaw:browser npm start
DEBUG=openclaw:llm npm start