Setting up a QQ robot shouldn't take hours of configuration and debugging. With OpenClaw's one-click installation, you can have a fully functional AI-powered QQ bot running in under 10 minutes.
This guide covers the fastest path from zero to a working OpenClaw QQ robot on Tencent Cloud Lighthouse.
Traditional QQ bot deployment involves:
One-click installation automates all of this into a single command.
SSH into your Lighthouse instance and run:
curl -fsSL https://raw.githubusercontent.com/AstraGPT/openclaw/main/scripts/qq-install.sh | bash
This script handles:
[1/6] Detecting system environment... Ubuntu 22.04 (OK)
[2/6] Installing system dependencies... Done (32s)
[3/6] Installing OpenClaw core... Done (45s)
[4/6] Configuring QQ adapter... Done (8s)
[5/6] Setting up process manager... Done (5s)
[6/6] Running health check... All systems operational
✅ OpenClaw QQ Robot installed successfully!
Next steps:
1. Edit /opt/openclaw/.env with your QQ credentials
2. Run: openclaw qq login
3. Scan the QR code with your QQ mobile app
Edit the environment file:
nano /opt/openclaw/.env
Fill in your QQ application details:
# QQ Robot Configuration
QQ_APP_ID=your_app_id
QQ_APP_SECRET=your_app_secret
QQ_BOT_TOKEN=your_bot_token
# AI Model Configuration
OPENAI_API_KEY=your_api_key
# Or use other LLM providers
# ANTHROPIC_API_KEY=your_key
# LOCAL_LLM_URL=http://localhost:11434
# Server Configuration
PORT=3000
NODE_ENV=production
# Start the login process
openclaw qq login
Follow the on-screen instructions to authenticate your QQ robot account. For enterprise applications, use the QQ Open Platform bot token flow:
# Token-based authentication (recommended for production)
openclaw qq auth --token
# Check service status
openclaw status
# Expected output:
# OpenClaw Status:
# Core: Running (PID: 12345)
# QQ Adapter: Connected
# AI Engine: Ready
# Uptime: 5m 23s
# Memory: 128MB / 2048MB
Send a test message to your QQ robot:
You: "Hello, are you there?"
Bot: "Hi! I'm your OpenClaw AI assistant. How can I help you today?"
Or via CLI:
# Send test message
openclaw qq test --message "What can you do?"
The one-click installer already configures PM2 for auto-start, but verify:
# Check PM2 startup configuration
pm2 list
pm2 save
# Verify auto-start is enabled
systemctl status pm2-root
Extend your QQ bot's capabilities:
# Browse available skills
openclaw plugin list --compatible qq
# Install popular skills
openclaw plugin install auto-reply group-manager trivia-game
# Restart to activate
pm2 restart openclaw
| Issue | Solution |
|---|---|
| Script fails on download | Check network: curl -I https://raw.githubusercontent.com |
| Port 3000 already in use | Change PORT in .env or kill existing process |
| QQ login timeout | Ensure server can reach QQ servers; check firewall |
| Memory errors | Upgrade Lighthouse instance to 2-core 4GB |
| Permission denied | Run with sudo or fix directory ownership |
If you need to remove the installation:
# Stop services
pm2 stop openclaw && pm2 delete openclaw
# Remove files
rm -rf /opt/openclaw
# Remove system service
pm2 unstartup
For QQ robot hosting, Lighthouse is the ideal choice:
Get started with Tencent Cloud Lighthouse special offers.
One-click installation removes the friction from QQ robot deployment. In under 10 minutes, you go from a bare Lighthouse instance to a fully operational AI-powered QQ bot. No manual dependency management, no configuration headaches — just run one command and configure your credentials.
For advanced configuration options, check the OpenClaw documentation.