
决策清晰
概览
一种可重复使用的代理技能,用于提升模糊、高价值、复杂性较大问题中的决策质量。 这项技能帮助LLM澄清真正的问题,揭示隐藏的假设,从基本事实推理,减少不必要的复杂性,并以更清晰的建议或下一步作结。 它结合了三种互补的推理模式,按实际顺序进行: 1. 澄清真正的问题 2.将问题简化为事实和约束 3.消除不必要的复杂性 - 识别真实决策或问题 - 揭示隐藏假设 - 将事实与惯性和惯性区分开来 - 将问题归结为约束、机制和因果结构 - 根据复杂性和假设负载比较选项 - 消除虚假复杂性而不删除现实 - 以建议、下一步或测试结束 当用户提出以下问题时,可以使用这个技能: - 我漏掉了什么?- 这真的非得这样吗?- 我该去掉什么?
概述(AI 翻译)
一种可重复使用的代理技能,用于提升模糊、高价值、复杂性较大问题中的决策质量。
这项技能帮助LLM澄清真正的问题,揭示隐藏的假设,从基本事实推理,减少不必要的复杂性,并以更清晰的建议或下一步作结。
它结合了三种互补的推理模式,按实际顺序进行:
- 苏格拉底式质疑以澄清真正的问题
- 第一原则思维将问题归结为事实和约束
- 奥卡姆式简化以消除不必要的复杂性
这项技能的作用
该技能旨在帮助特工:
- 识别真实的决策或问题
- 揭示隐藏假设
- 将事实与惯性和惯性区分开来
- 将问题归结为约束、机制和因果结构
- 按复杂度和假设负载比较选项
- 消除虚假复杂性而不删除现实
- 以建议、下一步或测试结束
最佳使用场景
当用户提出以下问题时,可以使用这个技能:
- 我漏掉了什么?
- 真的非得这样吗?
- 我应该去除什么?
- 最简单的解释是什么?
本页由 AI 翻译,技术细节请以原文为准。
README
以下为原文 README 节选,完整内容请查看原文。
decision-clarity
A reusable agent skill for improving decision quality in ambiguous, high-value, complexity-prone problems.
This skill helps an LLM clarify the real question, expose hidden assumptions, reason from fundamental facts, reduce unnecessary complexity, and end with a cleaner recommendation or next step.
It combines three complementary reasoning modes in a practical sequence:
- Socratic questioning to clarify the real issue
- First-principles thinking to reduce the issue to facts and constraints
- Occam-style simplification to remove unnecessary complexity
What this skill does
This skill is designed to help agents:
- identify the real decision or question
- expose hidden assumptions
- separate facts from inertia and convention
- reduce problems into constraints, mechanisms, and causal structure
- compare options by complexity and assumption load
- remove false complexity without deleting reality
- end with a recommendation, next move, or test
Best use cases
Use this skill when the user asks things like:
- What am I missing?
- Does this really have to be this way?
- What should I remove?
- What is the simplest explanation that still fits?
- Help me think this through.
- Which option actually makes the most sense?
It is especially useful for:
- startup and business decisions
- product and MVP decisions
- content and creator workflow design
- operations and process simplification
- strategy questions with multiple plausible options
- personal decisions distorted by confusion, self-justification, or excess complexity
- argument and reasoning audits
Structure
decision-clarity/
├── SKILL.md
├── Workflows/
│ ├── Clarify.md
│ ├── Deconstruct.md
│ ├── Simplify.md
│ └── Decide.md
└── references/
├── anti-patterns.md
├── business.md
├── content.md
├── examples.md
├── operations.md
├── output-patterns.md
├── product.md
└── trigger-questions.md
Workflow model
1. Clarify
Identify the real question, clean up vague framing, and expose initial assumptions.
2. Deconstruct
Reduce the problem into facts, constraints, mechanisms, and causal structure.
3. Simplify
Remove unnecessary complexity and compare options by assumption load while preserving adequacy.
4. Decide
Convert the clarified, decomposed, and simplified reasoning into a practical recommendation, next move, or test.
Included references
- anti-patterns.md — failure modes such as endless inquiry, false simplicity, and analysis without action
- business.md — startup, pricing, growth, distribution, and strategy questions
- product.md — MVP, feature decisions, onboarding, and scope simplification
- content.md — tutorials, creator workflows, and audience-value decisions
- operations.md — SOPs, approvals, handoffs, and process simplification
- trigger-questions.md — transforms vague prompts into sharper analysis frames
推荐工具
换一个关键词,或者移除筛选条件。
安装
npx skillfish add shanezzzz/decision-clarity-skill