David Ondrej · Video summary

OpenClaw 2.0 is here.

Leverage a VPS for OpenClaw instead of local hardware for cost savings, scalability, and remote access, enabling AI agent teams. Integrate with Cloud Code for complex software builds.

Watch on YouTube ↗

Key takeaways

• 💡Using a Virtual Private Server (VPS) for OpenClaw is significantly cheaper, easier, and more scalable than local hardware like Mac Minis, offering remote access from anywhere.00:41
• 🚀Hostinger provides a one-click setup for OpenClaw on a VPS, simplifying deployment to a matter of minutes without requiring DevOps or programming expertise.01:43
• 🚀Set up OpenClaw on a VPS using Hostinger's one-click installer for a cost-effective and scalable AI agent environment.01:43
• 🎯Leveraging OpenRouter as an AI model provider unlocks access to a wide array of cutting-edge LLMs (e.g., GPT-5.4, Gemini 3.1) through a single API key.05:14
• 🔗Configure OpenClaw to use OpenRouter for access to advanced AI models by obtaining an API key.05:14
• 🔑Integrating Cloud Code as a skill within OpenClaw allows the orchestrator agent to spawn and manage sub-agents for building complex software, enabling automated development workflows.09:16

Chapters

• 0:00🎯 Introduction to OpenClaw and Setup
• 0:48🎯 Why Use a VPS Instead of Hardware
• 2:39🎯 Sign-Up and Initial VPS Setup
• 3:30🎯 Connecting to the VPS Terminal
• 3:55🎯 Installing OpenClaw
• 5:40🎯 Setting Up API Keys and Connecting Models
• 6:50🎯 Configuring Telegram Integration
• 9:10🎯 Testing Telegram Connection and Basic Commands
• 9:20🎯 Installing Cloud Code and Managing Tasks
• 14:20🎯 Security Warnings and Best Practices
• 17:00🎯 Updating and Customizing OpenClaw Configuration
• 19:50🎯 Verifying and Using OpenClaw
• 24:40🎯 Demonstration of the ISS Tracker
• 26:30🎯 Final Thoughts and Next Steps

The digest

More from David Ondrej

All David Ondrej summaries →