We use rules for AI purposes. These can be placed in the Project root directory and the chosen LLM follows them. Being fans of Cline we follow their guidelines for creating these files, documentation is here (skip down to here to read about Clines implementation) if you are interested.
My suggestion is to allow for the creation of a similar rules definition within the Wappler AI implementation, accompanied with custom 'instructions' (as explained here for Cline) definition (in Cline/Roo and other tools this is available as a guideline for the LLM to follow). Both greatly improve the results returned by the LLM and can save repeating the same rules/instructions with every prompt.
These rules can significantly improve the results!
So please vote if you think this would be useful... Thank you.
We could even exchange/offer our own rules/instructions here within the Community... ![]()
![]()