---
name: creative_methodologies
description: "Applies three structured creativity methods to real problems: TRIZ contradiction solving for hard trade-offs, SCAMPER prompts for generating product variants, and SIT inside-the-box tools for ideas that reuse existing resources. Includes worked problems and step-by-step facilitation guidance for solo or group ideation sessions. Use when someone wants to brainstorm better, resolve a design trade-off, run an innovation workshop, or get past the obvious ideas."
---

# Creative Methodologies

## Purpose

Brainstorming usually produces the ideas people already had. These three methods produce ideas people would not have had. Each is documented from primary sources (references list them), each has a different shape of problem it fits, and each is applied through a concrete workflow, not a lecture. The skill also covers how to actually run a session: solo in 30 minutes or a group workshop in 90, with roles, timing, and convergence rules.

The three methods:

- **TRIZ** (Genrich Altshuller, built from studies of patent literature starting 1956): for problems with a built-in trade-off. You map the trade-off to a standard contradiction, then apply historically successful invention patterns (40 inventive principles) instead of compromising.
- **SCAMPER** (Alex Osborn's question lists, 1953, reorganized into a mnemonic by Bob Eberle, 1971): for generating variants of an existing product or service. Seven transformation prompts force angles the team would skip on its own.
- **SIT / Systematic Inventive Thinking**: for new-product ideas that use only what is already in the situation. Five thinking tools plus the "closed world" constraint produce surprising but buildable concepts.

## Workflow

1. **Frame the problem (5 minutes).** Write it down before touching any method:
   - What is the thing? Who uses it, and what do they need from it?
   - What must improve? What gets worse or blocks improvement if you try?
   - If the blocker is a trade-off (better X means worse Y), you are heading for TRIZ. If you have an existing thing and want more directions, SCAMPER. If you want new ideas without new resources, SIT.

2. **Pick the method.** Use this, do not guess:
   - Clear tug-of-war between two requirements (stronger but heavier, faster but hungrier) -> TRIZ. See references/triz.md.
   - "We have this product, what else could it be?" -> SCAMPER. See references/scamper.md.
   - "We need new ideas and cannot add cost, parts, or headcount" -> SIT. See references/sit.md.
   - If none fits cleanly, run SCAMPER first (it is the fastest warm-up), then the one that fits.

3. **Run the method.** Follow the reference file for the chosen method exactly:
   - TRIZ: state the contradiction in the template sentence, map to standard parameters, pick 2-4 principles, write at least one concrete concept per principle. See references/triz.md.
   - SCAMPER: walk all seven prompts, at least one idea per prompt, 2-3 minutes each. See references/scamper.md.
   - SIT: list the components of the existing situation first, then apply tools one at a time to build "virtual products," then work backwards to benefits (function follows form). See references/sit.md.

4. **Converge (10 minutes).** Every method session ends the same way:
   - Tag each idea with provenance: which method and which principle or tool produced it. Ideas without provenance are guesses, label them as such.
   - Score each on two axes: does it resolve the original framing, and what is the cheapest test that could kill it in a week? Pick the top 3 by a quick vote or dot count.
   - Write the cheap test down next to each idea. A session that ends with ideas and no tests produced conversation; this skill aims for candidates with next steps.

5. **Facilitation mode.** Running this for a group? See references/facilitation.md for the 90-minute workshop agenda, roles, and anti-patterns. Solo? Use the 30-minute compressed version there. Use assets/session-worksheet.md to record each run.

## Output Contract

Every completed run of this skill delivers:

1. The problem framing (the thing, the user, the improvement, the blocker).
2. The method chosen and why (one sentence).
3. Ideas, each with: short description, producing principle or tool, and the cheapest kill-test for it.
4. The top 3, with the chosen cheap tests written out.
5. The worksheet filled in (assets/session-worksheet.md), suitable for reuse in the next session.

## Operating Rules

1. Never invent facts about the methods. The histories, dates, and principle lists in references/ are verified against the cited sources; do not extend them from memory.
2. Worked problems in references/ are labeled illustrative. They teach the mechanics, they are not proof the method works in general.
3. Tag every idea with its provenance. If an idea came from a chat rather than a method step, say so.
4. Do not promise novelty. Checking whether an idea is patentable or genuinely new is out of scope; say that plainly if asked.
5. Style for everything this skill produces: plain, direct language. No em dashes. No inspirational filler, no stacked adjectives, no rhetorical questions answered by the author. If a sentence reads like a press release, rewrite it.
6. End every session with cheap tests attached to each surviving idea. The deliverable is candidates plus next steps.