Posts
All the articles I've posted.
-
Hard-Won Lessons from Multi-Agent Collaboration: How One config.patch Nearly Took Down the Whole System
Real-world lessons from a week of running a multi-agent system: a config management incident, TypeScript import errors, a wrong publish date, cost optimization in practice, and best practices for team collaboration.
-
AI Agent Frontend Workflow (Part 2): Intelligent Code Review and Automated Testing
Use AI Agents for intelligent code review and automated test case generation. From Git Hook integration to E2E testing, this post shares a complete hands-on setup with real project metrics.
-
Safely Exposing a Local AI Assistant to the Internet: SSH Reverse Tunnels in Practice
Expose a locally running OpenClaw Gateway to the public internet with an SSH reverse tunnel plus an Nginx reverse proxy, accessible via your own domain. All data stays local, with multiple layers of security — at zero cost.
-
AI Agent Frontend Workflow (Part 1): Understanding Agents and Automated Component Generation in Practice
An AI Agent is not just a chatbot — it's an intelligent assistant that can invoke tools and manage context on its own. This post breaks down how Agents actually work, then walks through a hands-on React component generator to show how AI can reshape the frontend development workflow.
-
OpenClaw Multi-Agent Setup in Practice: Pitfalls and Best Practices
Setting up multiple OpenClaw agents and multiple Telegram accounts comes with plenty of pitfalls. Based on hands-on experience, this post covers every common problem and its fix so you can skip the pain.
-
AI Agent Dev Tools Compared 2026: Claude Code vs OpenClaw vs Cursor — Which One Should You Pick
A hands-on comparison of Claude Code, OpenClaw, and Cursor — the three big AI coding tools. From runtime model, memory systems, and model support to skill mechanisms, this guide covers how to choose an AI agent dev tool in 2026, plus a deep dive into their config systems and a cross-tool migration guide.