brainstorming
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).
s4h-identity-character-testingAsks what a person or organisation of genuine integrity would do — grounding decisions in character rather than calculation. Triggers: 'character test', 'what would a person of integrity do', 'what does this say about us', 'gut check', 'would I be proud of this', 'are we being who we want to be'.
Independent GitHub snapshot · Automated classification · Original documentation governs
Ethical calculation — weighing outcomes, mapping trade-offs — can be gamed. Character cannot. The question "what would a person of genuine integrity do here?" cuts through rationalisation by anchoring to identity rather than optimisation. It works best when something feels wrong but is hard to name.
Step 1: Describe the Situation and Decision State what is happening and what is being considered. Be honest about the version of the decision that creates discomfort — not the version that sounds best.
Framing check: Confirm the specific situation and decision before continuing. State what you've identified — the actual situation, the decision being considered, and who it concerns — in one sentence, then use AskUserQuestion:
Step 2: Describe the Character Not aspiration, but genuine commitment — what kind of person or organisation is this, when at its best? What does it actually stand for? Describe this in concrete behavioural terms, not in values words.
Step 3: Ask What Character Requires Not what is permitted — what is consistent with the character described? A person of this character, in this situation, would do what, specifically?
Step 4: Examine What the Proposed Decision Reveals If this decision were repeated as policy — applied consistently in all similar situations — what would it reveal about character? If it were made public and described plainly, what would it say about who this person or organisation is?
Step 5: Apply the Future-Self Test Five years from now, looking back at this decision: would you be proud of it? Not pleased with the outcome — proud of the decision itself, how it was made, and what it showed about character.
Step 6: Analyse the Divergence If the character test points in a different direction from the proposed decision: what is driving that divergence? Is it a genuine competing consideration — or is it pressure, convenience, or fear? Is the divergence legitimate?
Before proceeding, use the AskUserQuestion tool. State your interpretation of the situation in 1–2 sentences — what is being analyzed and what the core question is — then ask:
Proceed based on their selection. If the user reframes, incorporate the correction before running any analysis.
[Honest description — including the uncomfortable version]
[Concrete, behavioural — not abstract values words]
In this situation, a person or organisation of this character would: [specific]
If repeated as policy or made public: [what it would say about character]
Looking back in five years: proud / uncomfortable / regretful — and why?
| The character test points toward | The proposed decision points toward | What's driving the divergence | Is it legitimate? |
|---|---|---|---|
| ... | ... | ... | Yes / No |
[What the character test implies — and whether it should override the original proposal]
This test is most valuable when the calculation says one thing and something else says no. The discomfort is information — it deserves analysis rather than suppression.
After delivering this output, use AskUserQuestion to offer the next move:
/s4h-identity-values-clarification — Clarify values that the character test revealed tensions in/s4h-ethics-check — Check whether character under pressure met ethical standards/s4h-decision-premortem-analysis — Stress-test decisions for consistency with character