rebsonm/meuartigoskill ? reverse-engineered prompt
Reverse engineered prompt
Build me an AI writing assistant for researchers that helps turn a rough idea into a submission ready scientific manuscript.
I want it to keep everything traceable, so it should record sources, notes, decisions, progress, pending tasks, and journal requirements in one continuous workspace instead of treating each chat like a fresh start. It should support the C.A.D.A. flow for tasks, capture, assign, set a deadline, and follow up. It should also let the user work in their own language, while still handling the manuscript in the language they choose for the paper.
Make it careful about research integrity, so it never invents references, data, or prior work, and always distinguishes what comes from literature, what is an inference, and what is an original proposal. If possible, include a simple dashboard view that shows current status, decisions, evidence, and history, with a plain text fallback when visual panels are not available. Use the existing docs and README as the source of truth, and look up current docs online if you need to.
Are you gonna build this?
make sure you review the code using coderabbit