Explore, don’t decide: widen the field before the model converges

Divergence gateFreshModel-agnosticverified 2026-06-29sha 71a49ff7fa41

Born in: the post where it earned its place →

Most “give me options” prompts return one sensible idea wearing five jackets – five phrasings that share the same hidden assumption. Paste this as your message at the moment you are about to let AI move from exploring to recommending: fill the four fields, and don’t tack “so what should we do?” onto the same turn. It makes the model widen the option space and keep facts, inferences, and framings separate instead of handing you one seamless answer that has already chosen the question for you. Born from noticing that a plausible first answer has gravity – once it exists, later thinking becomes revision rather than discovery.

MODE: EXPLORE, NOT DECIDE

Question: {{question}}
My current framing: {{current_framing}}
Known constraints: {{known_constraints}}
What I may be assuming: {{assumptions}}

Return:

1. Facts you can support from what you can actually see this turn - the provided
   context, files read this run, or tool / browse output in this conversation. Link
   each. If you have no accessed sources this turn, say so plainly and do not invent
   citations.

2. Inferences, clearly labelled as inferences (not facts).

3. Three materially different framings of the problem - not three phrasings of one.
   For each, state its key assumption and its likely failure mode.

4. Divergence check: name the assumption all three framings share. If they share the
   decisive one, throw one out and replace it with a framing that challenges it.

5. Important unknowns, and the evidence that would resolve each.

6. Choices that depend on purpose, taste, or values. Mark these as mine to make,
   not yours to decide.

Do not recommend, rank, score, or name a "leading", "strongest", or "default"
framing - no "I would choose", "most teams", or "best option". End with the single
question whose answer would most change the available option space.

Fill in: question current_framing known_constraints assumptions

Shared as is, for reference. Read it and decide what it will do before you run it; using it is your responsibility, under the terms.