Post-generation diversity audit that scores whether a set of ideas are actually different from each other or just surface variations of the same thought. Use AFTER any brainstorming, ideation, or creative generation — from any source, any skill, any person — when you need to know if your "ten different ideas" are actually ten ideas or three ideas wearing costumes. Triggers on "are these actually different," "these all feel the same," "did I just write the same idea five times," "check these for diversity," "which of these overlap," or any moment after generating a list where the user wants to know what they actually have before moving forward. Also use automatically as a quality gate after any other creative skill produces output, or when a user pastes in a list of ideas from anywhere and wants them audited. Works on any problem in any domain — strategy, engineering, science, policy, design, education, medicine, law, research. This is the bullshit detector for brainstorms.
You just generated ten ideas. How many ideas do you actually have?
The research says: probably three.
Doshi & Hauser (2024) proved that AI-assisted creative output is rated more creative individually but is significantly more similar collectively. Wenger & Kenett (2025) confirmed that LLMs show high creative homogeneity even across different model families. Ideas from a single AI perspective cluster at 0.92 cosine similarity.
That means most brainstorms — whether from AI, from a team meeting, or from a solo session at midnight — produce the illusion of variety while actually circling the same center.
This happens in every domain. Marketing brainstorms produce ten ideas that are all "get attention through someone else's audience." Engineering brainstorms produce ten ideas that are all "add another layer of abstraction." Research brainstorms produce ten ideas that are all "run the same experiment with a larger sample." Policy brainstorms produce ten ideas that are all "expand the existing program." Education brainstorms produce ten ideas that are all "make the same lesson more engaging."
The surface looks different. The skeleton is the same.
This skill is the x-ray. It shows you the skeleton underneath the ten ideas that look different on the surface, so you can see how many bones you're actually working with.
Feed this skill any list of ideas. From any source. Any format. Any domain.
It does three things:
It does not generate new ideas. It does not fix the list. It tells you what you have so you can decide what to do about it.
For every idea in the list,