rome-os/rome ? reverse-engineered prompt
Reverse engineered prompt
Build me a self hosted web app for running a personal AI agent that can keep working over time, remember what it did, and take on repeatable tasks like inbox cleanup, code review follow up, price tracking, and scheduled reminders.
I want a clean dashboard where I can chat with the agent, see ongoing tasks, review what it has done, and manage saved workflows and apps. The system should let the agent use tools, store persistent state, and keep working after I close the browser. It should also support creating small purpose built apps for recurring jobs, not just one off chats.
Please make it runnable locally with Docker, and make the first time setup simple. If you need current docs or best practices for any libraries, look them up online. Keep the code organized and production shaped, with a good default experience for a single user or small team.
Are you gonna build this?
make sure you review the code using coderabbit