When to invoke
Invoke this skill during backlog refinement to generate combined ideas (morphological matrix mode) or during sprint review to extract effects from a completed prototype (effects-extraction mode).
Inputs needed
Either a list of backlog items (for morphological mode) OR a prototype description or completed feature (for effects-extraction mode).
Reference data used
reference/principles.md(for both modes)reference/matrix.json(only when effects-extraction surfaces a contradiction)
Procedure
- Receive inputs and determine the mode.
- For morphological mode (backlog refinement):
- Reframe backlog items as contradictions.
- Enumerate candidate principles per row.
- Generate a bounded set of combined ideas by making one selection per row.
- Name the principles each idea draws from.
- Present the ideas to the user. Do not automatically apply them.
- For effects-extraction mode (sprint review):
- Take the prototype description.
- List the positive effects achieved.
- List the harmful side-effects or unresolved issues.
- Reframe each harmful effect as a fresh contradiction (improving X / worsening Y) for the next sprint backlog.
- Stop. Ask the user to confirm. Do not proceed until they reply.
- Finalize the output based on user feedback. No code edits are performed by this skill.
Output format
- For morphological mode: A small ranked list of combined ideas with principle attributions.
- For effects-extraction mode: An updated contradiction backlog listing positive effects and reframed harmful effects.