question-refinement-loop
$
npx mdskill add yogsoth-ai/de-anthropocentric-research-engine/question-refinement-loop迭代精化研究问题 — 反复打磨直到通过 FINER 5 项标准。
SKILL.md
.github/skills/question-refinement-loopView on GitHub ↗
---
name: question-refinement-loop
description: "Tactic: 迭代精化研究问题直到通过 FINER 5 项标准"
version: 1.0.0
category: hypothesis-formation
type: tactic
campaign: research-question
sops:
- finer-criteria-check
- scope-assessment
- success-criteria-definition
dependencies:
skills:
- subagent-spawning
---
# Question Refinement Loop
迭代精化研究问题 — 反复打磨直到通过 FINER 5 项标准。
## 编排意图
将"精化问题"结构化为:检验 → 识别问题 → 修正 → 重新检验的循环。每轮聚焦于未通过的标准项。
## 可用 SOPs
| SOP | 职责 | 何时调用 |
|-----|------|---------|
| finer-criteria-check | FINER 5 项标准逐项检验 | 每轮循环的检验步骤 |
| scope-assessment | 评估问题范围是否合适 | 当 F(Feasible) 不通过时 |
| success-criteria-definition | 定义可测量的成功标准 | 循环结束后,确认 RQ 可衡量 |
## 编排模式
**标准循环**:
1. finer-criteria-check → 获得 5 项判定
2. 如果全通过 → 调用 success-criteria-definition → 结束
3. 如果有未通过项 → 针对性修正:
- F 不通过 → scope-assessment → 缩小范围
- I 不通过 → 增强理论动机
- N 不通过 → 检查文献确认新颖性
- E 不通过 → 调整研究设计
- R 不通过 → 强化实践意义
4. 修正后回到步骤 1
**最大迭代**: 3 轮。超过 3 轮仍不通过 → 报告哪项持续不通过 + 建议根本性调整方向。
## Minimum Yield
- RQ 通过 FINER 5/5
- Success criteria 定义完成(可测量)
- 如果 3 轮未通过:明确报告问题项 + 建议
## Yield Report
执行完成后报告:
- 迭代轮次
- 每轮的 FINER 结果
- 最终 RQ 表述
- Success criteria
More from yogsoth-ai/de-anthropocentric-research-engine
- abductive-hypothesis-generationStrategy: 面对异常的最佳解释推理
- ablation-brainstormRemove components one by one, observe system changes to reveal hidden dependencies and generate ideas from structural gaps.
- ablation-component-mappingMap system architecture to ablatable units for ablation studies
- ablation-designDesign ablation studies to isolate component contributions in ML systems
- ablation-executionRemove components one by one from a system, record the response/impact of each removal.
- abp-vulnerability-classificationClassify assumptions on 2 axes — load-bearing (how much conclusion depends on it) × vulnerable (how likely to be false). Focuses attention on High-Load × High-Vulnerable quadrant.
- abstraction-extractionExtract abstract principles from concrete domain cases. Strips domain-specific details to reveal transferable mechanisms.
- abstraction-ladderPerform bisociation at multiple abstraction levels
- abstraction-ladderingMove between concrete and abstract framings — 3 levels up (Why?) and 3 levels down (How?) to find the most productive research level.
- abstraction-to-designAbstract biological principle to design principle. Bridge from biology to engineering.