Openclaw Chinese AI Assistant

OpenClaw Chinese AI Assistant Skill

Skill by ara.so — Hermes Skills collection.

Overview

OpenClaw Chinese Translation is a fully localized (Simplified Chinese) distribution of OpenClaw, an open-source personal AI assistant platform with 195,000+ GitHub stars. It runs locally on your machine and provides AI assistance through popular chat platforms (WhatsApp, Telegram, Discord) with both CLI and web dashboard interfaces fully translated to Chinese.

Key Features:

• Automatic hourly sync with upstream OpenClaw (< 1 hour delay)
• Full Chinese localization for CLI and Dashboard
• Multi-platform support: WhatsApp, Telegram, Discord
• Multiple LLM providers: Claude, OpenAI, compatible APIs
• Tool calling and skill extensions
• Daemon/service mode for background operation
• Web-based management dashboard

Installation

Prerequisites

Node.js >= 22 is required. Check version:

node -v

If needed, download from https://nodejs.org/

Install OpenClaw Chinese

npm install -g @qingchencloud/openclaw-zh@latest

Verify Installation

openclaw --version

Initialization and Configuration

Initial Setup with Daemon

openclaw onboard --install-daemon

This interactive wizard will guide you through:

1. Selecting an LLM provider (Claude, OpenAI, or compatible)
2. Configuring API credentials
3. Setting up chat channels (WhatsApp/Telegram/Discord)
4. Installing gateway as daemon service

Manual Configuration

View current config:

openclaw config

Edit config file directly (located at ~/.openclaw/config.yml):

llm:
  provider: openai
  api_key: ${OPENAI_API_KEY}
  model: gpt-4
  base_url: https://api.openai.com/v1  # For compatible providers

gateway:
  port: 3000
  host: 0.0.0.0

channels:
  - type: telegram
    token: ${TELEGRAM_BOT_TOKEN}
  - type: discord
    token: ${DISCORD_BOT_TOKEN}

Configure LLM Provider

For OpenAI-compatible providers (like the free gpt.qt.cool):

openclaw config set llm.provider openai-compatible
openclaw config set llm.base_url https://gpt.qt.cool/v1
openclaw config set llm.api_key ${YOUR_API_KEY}
openclaw config set llm.model gpt-4

For Claude:

openclaw config set llm.provider anthropic
openclaw config set llm.api_key ${ANTHROPIC_API_KEY}
openclaw config set llm.model claude-3-5-sonnet-20241022

Core Commands

Gateway Management

# Start gateway (foreground, for debugging)
openclaw gateway run

# Start as daemon (background, recommended)
openclaw gateway start

# Stop daemon
openclaw gateway stop

# Restart gateway
openclaw gateway restart

# Check status
openclaw gateway status

# Install as system service (auto-start on boot)
openclaw gateway install

# Uninstall system service
openclaw gateway uninstall

Dashboard

# Open web dashboard (auto-opens browser)
openclaw dashboard

# Open on specific port
openclaw dashboard --port 8080

# Access remotely with token
openclaw dashboard --host 0.0.0.0 --token ${DASHBOARD_TOKEN}

Dashboard URL: http://localhost:3000 (default)

Language Setting: After opening dashboard, go to Overview page → scroll to bottom → set Language to 简体中文 (Simplified Chinese) → refresh.

Skills Management

# List available skills
openclaw skills list

# Install a skill
openclaw skills install 1password

# Remove a skill
openclaw skills remove 1password

# Update all skills
openclaw skills update

Maintenance

# Update OpenClaw CLI
openclaw update

# Diagnose issues (auto-fix)
openclaw doctor

# View logs
openclaw logs

# Clear cache
openclaw cache clear

Real-World Usage Examples

Example 1: Deploy with Custom LLM Provider

#!/bin/bash

# Set environment variables
export OPENAI_API_KEY="sk-your-key-here"
export TELEGRAM_BOT_TOKEN="123456:ABC-DEF-your-token"

# Install
npm install -g @qingchencloud/openclaw-zh@latest

# Configure programmatically
openclaw config set llm.provider openai-compatible
openclaw config set llm.base_url https://gpt.qt.cool/v1
openclaw config set llm.api_key "${OPENAI_API_KEY}"
openclaw config set llm.model gpt-4

# Add Telegram channel
openclaw config set channels.0.type telegram
openclaw config set channels.0.token "${TELEGRAM_BOT_TOKEN}"

# Start daemon
openclaw gateway start

# Verify running
openclaw gateway status

Example 2: Docker Deployment

Create docker-compose.yml:

version: '3.8'

services:
  openclaw:
    image: ghcr.io/1186258278/openclaw-chinese:latest
    container_name: openclaw-zh
    restart: unless-stopped
    ports:
      - "3000:3000"
    environment:
      - OPENAI_API_KEY=${OPENAI_API_KEY}
      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
      - TELEGRAM_BOT_TOKEN=${TELEGRAM_BOT_TOKEN}
      - DASHBOARD_TOKEN=${DASHBOARD_TOKEN}
    volumes:
      - ./data:/root/.openclaw
      - ./logs:/var/log/openclaw

Run:

docker-compose up -d

Example 3: Remote Access with Nginx

Nginx config for HTTPS dashboard access:

server {
    listen 443 ssl;
    server_name openclaw.example.com;

    ssl_certificate /path/to/cert.pem;
    ssl_certificate_key /path/to/key.pem;

    location / {
        proxy_pass http://127.0.0.1:3000;
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection 'upgrade';
        proxy_set_header Host $host;
        proxy_cache_bypass $http_upgrade;
        
        # WebSocket support
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
    }
}

Example 4: Programmatic Agent Creation (JavaScript)

// Create custom agent configuration
const fs = require('fs');
const path = require('path');
const yaml = require('js-yaml');

const configPath = path.join(process.env.HOME, '.openclaw', 'config.yml');
const config = yaml.load(fs.readFileSync(configPath, 'utf8'));

// Add new agent
config.agents = config.agents || [];
config.agents.push({
  name: 'research-assistant',
  model: 'gpt-4',
  system_prompt: '你是一个专业的研究助手，帮助用户查找和总结信息。',
  skills: ['web-search', 'summarize', 'markdown'],
  max_tokens: 4000,
  temperature: 0.7
});

// Save config
fs.writeFileSync(configPath, yaml.dump(config), 'utf8');
console.log('Agent created successfully');

Example 5: Monitoring Gateway Status

#!/bin/bash

# Check if gateway is running
if openclaw gateway status | grep -q "running"; then
    echo "Gateway is healthy"
    exit 0
else
    echo "Gateway is down, restarting..."
    openclaw gateway restart
    sleep 5
    if openclaw gateway status | grep -q "running"; then
        echo "Gateway restarted successfully"
        exit 0
    else
        echo "Failed to restart gateway"
        exit 1
    fi
fi

Common Patterns

Pattern 1: Multi-Model Setup

Configure different models for different tasks:

agents:
  - name: fast-assistant
    model: gpt-3.5-turbo
    use_case: quick_questions
  
  - name: deep-thinker
    model: claude-3-opus-20240229
    use_case: complex_reasoning
  
  - name: code-helper
    model: gpt-4
    skills: ['code-interpreter', 'github']

Pattern 2: Environment-Based Config

Use environment variables for secrets:

# .env file
OPENAI_API_KEY=sk-proj-...
ANTHROPIC_API_KEY=sk-ant-...
TELEGRAM_BOT_TOKEN=123456:ABC...
WHATSAPP_PHONE_NUMBER=+1234567890
DASHBOARD_TOKEN=secure-random-token-here

# Load in config.yml
llm:
  api_key: ${OPENAI_API_KEY}

channels:
  - type: telegram
    token: ${TELEGRAM_BOT_TOKEN}
  - type: whatsapp
    phone: ${WHATSAPP_PHONE_NUMBER}

Pattern 3: Skill Chains

Combine multiple skills for complex workflows:

// Example: Research and summarize workflow
const workflow = {
  steps: [
    { skill: 'web-search', query: '${user_query}' },
    { skill: 'summarize', input: '${search_results}' },
    { skill: 'markdown', format: 'report', input: '${summary}' }
  ]
};

Troubleshooting

Gateway Won't Start

# Check if port 3000 is already in use
lsof -i :3000  # macOS/Linux
netstat -ano | findstr :3000  # Windows

# Run diagnostics
openclaw doctor

# Check logs
openclaw logs --tail 100

# Try alternative port
openclaw gateway run --port 3001

Dashboard Connection Issues

1. Cannot access dashboard remotely:

# Start with external access
openclaw dashboard --host 0.0.0.0 --token your-secure-token

# Or edit config
openclaw config set gateway.host 0.0.0.0
openclaw config set gateway.dashboard_token your-secure-token
openclaw gateway restart

2. CORS errors:

# config.yml
gateway:
  cors:
    enabled: true
    origins:
      - https://yourdomain.com
      - http://localhost:*

LLM Provider Errors

# Test API connection
curl https://gpt.qt.cool/v1/models \
  -H "Authorization: Bearer ${OPENAI_API_KEY}"

# Verify config
openclaw config get llm

# Switch provider
openclaw config set llm.provider openai-compatible
openclaw config set llm.base_url https://api.openai.com/v1

Daemon Service Issues (Windows)

If gateway install fails on Windows:

# Use daemon start instead (doesn't require schtasks)
openclaw gateway start

# Or use Docker
docker run -d -p 3000:3000 \
  -v ~/.openclaw:/root/.openclaw \
  ghcr.io/1186258278/openclaw-chinese:latest

Memory/Performance Issues

# Check resource usage
openclaw gateway status --verbose

# Limit memory (Node.js)
export NODE_OPTIONS="--max-old-space-size=4096"
openclaw gateway restart

# Clear cache
openclaw cache clear

Update Issues

# Force reinstall
npm uninstall -g @qingchencloud/openclaw-zh
npm install -g @qingchencloud/openclaw-zh@latest

# Clear npm cache if needed
npm cache clean --force

Uninstallation

# Stop services
openclaw gateway stop
openclaw gateway uninstall  # If installed as service

# Uninstall package
npm uninstall -g @qingchencloud/openclaw-zh

# Remove config (optional)
rm -rf ~/.openclaw

Additional Resources

• Official Chinese Documentation: https://openclaw.qt.cool
• GitHub Repository: https://github.com/1186258278/OpenClawChineseTranslation
• Free AI API Platform: https://gpt.qt.cool
• ClawPanel (GUI Manager): https://github.com/qingchencloud/clawpanel
• ClawApp (Mobile Client): https://github.com/qingchencloud/clawapp
• Detailed Install Guide: docs/INSTALL_GUIDE.md
• Docker Guide: docs/DOCKER_GUIDE.md
• FAQ: docs/FAQ.md