Donroy26/dons-factory ? reverse-engineered prompt

Build me a markdown based workspace for using an AI agent to take a “system” from idea to live deployment through a set of clear stations.

I want the repo to feel like a small factory, where each stage has its own instructions, each handoff is explicit, and one travel sheet follows the work the whole way. The system should move between folders as it progresses, with fresh context at each stage, and only the files named in the current station should be visible to the agent. Keep it model agnostic, so it can work with different agent runtimes, and make the whole thing feel simple enough that someone can drop it into any folder and start using it right away.

Please include the station docs, templates for the files each station produces, and the two small Python tools for a live status line and a basic factory dashboard. If you need to check current agent docs or best practices, look them up online.