riiiku/clarify-skill
ResearchThis page was translated automatically. Technical details may differ from the original. View the original
Wittgenstein + Socrates + Polanyi: A Claude Code Skill for turning fuzzy thoughts into precise instructions
Overview
An AI Skill based on three philosophical systems helps you turn vague ideas into precise instructions or clear decisions. Break down vague ideas into precise instructions. Suitable for wanting AI to do something for you but not knowing exactly what you want. : Proposition classification (F/D/Q) → Breakdown of fuzzy words → Socrates' Three Questions → Polany's Backing → Assembling structured instructions Ask questions about unclear troubles, struggles, and impulses. Claude doesn't make decisions for you; she only helps you dig out your own answers. : Collect scenarios → Language clarification → Progressive questioning → Polanyi as a safety net → Feedback conclusion → Action anchor This is a Claude Code Skill file. Place the SKILL.md into the .claude/skills/clarify/ directory to use. Trigger methods: - /clarify — automatic judgment mode - /clarify prompt — forced prompt mode - /clarify decision — forced decision mode - Wittgenstein tells you: If you're not sure, don't speak up. - Socrates tells you: most of the time you've thought it through, but you're just not asked. - Polanyi tells you: there are things you can't get no matter how much you ask; you need to find a different approach
Overview (machine translated)
Wittgenstein propositional decomposition + Socratic introductory method + Polish Nimerian knowledge extraction
An AI Skill based on three philosophical systems helps you turn vague ideas into precise instructions or clear decisions.
Three-tier architecture
Two Models
Prompt Mode
Break down vague ideas into precise instructions. Suitable for wanting AI to do something for you but not knowing exactly what you want.
Process: Proposition classification (F/D/Q) → Breakdown of fuzzy words → Socratic Questions → Polanyi Backing → Assembling structured instructions
Decision Mode
Ask questions about unclear troubles, struggles, and impulses. Claude doesn’t make decisions for you; she only helps you dig out your own answers.
Process: Collect scenarios → Clarify language → Progressive questions → Polanyi provide a safety net → Feedback conclusions → Action anchors
Philosophical Foundations
| Philosopher | Core Ideas | Role in the System |
|---|---|---|
| Wittgenstein | Language is the boundary of thought; unclear = not clearly thought | Scan for blurred points |
| Socrates | The answer is in your heart and needs to be asked correctly | Ask back to bedrock |
| Polany | We can know more than we can tell | Extracting tacit knowledge |
How to use
This is a Claude Code Skill file. Place SKILL.md into the .claude/skills/clarify/ directory to use it.
Trigger method:
/clarify— Automatic check mode/clarify prompt— Forced prompt mode/clarify decision— Forced Decision mode
Design Concept
The relationship among these three philosophers is like a Tai Chi diagram:
- Wittgenstein tells you: If you’re unsure, don’t speak
- Socrates tells you: most of the time you think it through, just haven’t been asked
- Polanyi tells you: There are some things you can’t get no idea no matter how much you ask; you need to find a different approach
The three layers together cover all the dilemmas of human expression: the unclear-thought-out → the thought but unspoken→ the unspeakable.
License
MIT
This page was translated automatically. Technical details may differ from the original.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add riiiku/clarify-skill