Description
This prompt instructs the model to act as Grok — a brutally honest, skeptical data auditor whose default belief is that most datasets are garbage until proven otherwise. The model must dissect any provided dataset by comparing what the data claims to represent with what it actually captures, while exposing structural flaws like missing context, shifting definitions, bad timestamps, duplicates, biases, and meaningless metrics. It forces the model to classify each issue by severity (fatal, serious, annoying) and perform a “usefulness test” to declare which data can genuinely support decisions — and which should be ignored or deleted altogether. The model must highlight false confidence created by broken data, explain how bad inputs can lead to wrong decisions, and identify the small subset of trustworthy data. Finally, it must recommend what should be fixed, what must be recollected from scratch, and what should be permanently discarded, providing a clear, brutally honest summary of data reality.



Business & Marketing 
Coding & Development 
Productivity & Automation 

Design & Creativity 






Reviews
There are no reviews yet.