Thinking simulator¶
Challenge this note with critical-thinking prompts adapted to its gaps.
Called thinking-simulator in the step's YAML.
A Knowledge action. It asks nothing in the wizard. It writes questions that push you to stress-test the idea, adapted to the note's actual gaps in the knowledge model — no text or AI inference. Deterministic, offline, and fully usable with AI turned off.
Settings¶
- Result property — where the prompts are written. Default:
thinkingPrompts. - Write to — Frontmatter (a property on the note) or Context (only for the next steps). Default: Frontmatter. Either way the next steps can use them as
{{<result property>}}, e.g.{{thinkingPrompts}}. - Target note (optional) — the note to challenge. Empty uses the note being built.
The prompts¶
Always asked¶
These four apply to any idea, so the list is never empty — even for a brand-new note:
- What if this is false?
- Who would disagree, and why?
- What happens in the extreme case?
- What hidden assumption does this rest on?
Asked when the note has the gap¶
| Prompt | When |
|---|---|
| What evidence is missing to support this? | the note makes a claim but cites no source |
| What contradicts this? | no contradicts relation in or out |
| What is a concrete example? | no outgoing example relation |
| How does this connect to what you already know? | nothing links to it |
| What open question does this raise? | no question relation |
Suggest next move reads the same gaps and gives a to-do ("add a source"); this one gives a challenge ("what evidence is missing — and what if it's false?").
What it writes¶
---
thinkingPrompts:
- What if this is false?
- Who would disagree, and why?
- What happens in the extreme case?
- What hidden assumption does this rest on?
- What evidence is missing to support this?
---
A notice reports the count. If the index isn't ready, the action does nothing.
Capabilities¶
File-system read + the disclosed result-property write. No network, no AI.