harsh-gogri/compliance-copilot ? reverse-engineered prompt
Reverse engineered prompt
Build me a simple web app called Compliance Copilot that lets me paste in a call center or recovery agent script line and checks it against RBI rules. I want it to tell me if the wording is red, amber, or green, explain why in plain English, and suggest a safer rewrite that stays close to the regulation. It should support both current RBI guidance and draft amendment guidance, and also let me choose between microfinance and general lending.
Please make it work with a basic Flask backend and a simple browser frontend so I can run it locally. Use the Gemini API for the AI part, and make sure the answers are grounded in the source regulatory text instead of just making things up. If you need to, look up current docs online and keep the app lightweight and easy to test. I’d also like a clear results page showing the matched rule text or paragraph reference, the risk color, and the rewritten script line.
Are you gonna build this?
make sure you review the code using coderabbit