Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/diverge
Before implementing, generate 3-5 conceptually distinct approaches labeled by creativity dimension (Novel, Surprising, Diverse, Conventional), then hold for selection. Brainstorm-then-select to resist defaulting to the most obvious solution.
Passed 20 of 22 test cases with the skill, and 5 of 22 without it.
and 20 more in the full report
15 fixed5 passed either way2 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
I want to overhaul our backend architecture using divergent thinking to make it much better. What approaches should we consider?
| Checked | Without skill | With skill |
|---|---|---|
| Asks a focused clarifying question about what success looks like before generating approaches | Missed | Passed |
| Refrains from listing approaches immediately when the goal is ambiguous | Missed | Passed |
The instructions your model reads when the skill loads.
Interrupt the default path of jumping to the most probable — and least creative — solution.
This is an original Open Science Skills workflow grounded in Creative Preference Optimization (Ismayilzada et al., 2025; background in reference/creative-preference-optimization.md). Standard preference alignment (RLHF/DPO) optimizes for the most human-expected output, which is by construction the least surprising one. The paper's most accessible remedy — its own "brainstorm-then-select" baseline — needs no fine-tuning: force divergence before convergence by generating several conceptually distinct approaches, requiring that at least one is surprising and one is novel, and deferring quality and implementation until after selection.
Use diverge for creative, architectural, or analytical work where more than one non-obvious solution exists. To delegate the brainstorm to a second model family, use the sibling diverge-codex.
Use /diverge <task> when:
Do not use for rote tasks with one correct answer (e.g., fix this syntax error).
Given $ARGUMENTS:
If the task is ambiguous about what "good" looks like, ask one focused question before proceeding. Skip this if the goal is clear. Do not ask about implementation details.
Produce 3–5 approaches that are genuinely conceptually distinct. Differences must be in underlying mechanism, not surface vocabulary.
Label each with its primary creativity dimension:
For each approach provide:
test cases passed, vs 5 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.