At some point, I think most of us had the same thought: "Why am I still doing this myself?" Open an app. Copy something. Paste it into ChatGPT. Wait. Open another tab. Download a file. Run a script. Come back. Do it again. AI can write the code, explain the document, browse the web, analyze the data… …but somehow, we're still the ones moving everything around. That started bothering me. I didn't want another chatbot. I started wondering: What if the computer itself was built around agents? Not an AI sitting inside an app. Not an agent that disappears when you close a tab. But something that can run, remember, use tools, interact with the system, and keep working in the background — almost like another process on your computer. That question eventually turned into **Formicx**. I'm building it as an agent-native OS/runtime — an environment where agents aren't just applications running on top of the system. They're part of the system. And honestly, Formicx is still very much a work in progress. But that's the fun part. The goal isn't to make AI better at answering: “What should I do?” It's to make it better at: “I'll take care of it.”

Formicx

Linux manages processes. Formicx manages agents.

Linux manages processes. Formicx manages agents.