Show HN: A visual debugger for AI-generated code

I wrote 200k+ lines of code this year using Cursor, and I realized that debugging AI-generated code is harder because you often don’t know what’s going on.

That’s why I built Samurai Agent: it visualizes the code execution flow so you can actually understand what the AI wrote easily, and it provides root-cause analysis instead of superficial fixes.

If you have the same problems, please join the beta. We’re giving out free credits, and feedback is appreciated. We’re focused on JavaScript/TypeScript for now.

https://samurai-agent.tech/