framework-matching
$
npx mdskill add yogsoth-ai/de-anthropocentric-research-engine/framework-matchingMatches the most suitable RQ framework based on research type
- Solves the problem of selecting the right RQ framework for a given research type
- Depends on subagent-spawning for deeper analysis if needed
- Uses a framework knowledge base to evaluate fit based on research type and domain
- Returns a recommended framework with rationale and candidate comparison
SKILL.md
.github/skills/framework-matchingView on GitHub ↗
---
name: framework-matching
description: "SOP: 根据研究类型匹配最适合的 RQ 框架"
version: 1.0.0
category: hypothesis-formation
type: sop
campaign: research-question
input: "研究类型 + 领域 + 假设特征"
output: "推荐框架 + 理由 + 候选框架对比"
dependencies:
skills:
- subagent-spawning
---
# Framework Matching
根据研究类型匹配最适合的 RQ 框架。
## HARD-GATE
<HARD-GATE>
输入必须包含: 研究方向/假设 + 研究类型(或足够信息推断研究类型)。
</HARD-GATE>
## Pipeline
1. **前置检查**: 研究方向和类型是否明确
2. **类型判定**: 确定研究类型(量化干预/定性探索/评估/混合)
3. **框架匹配**: 根据类型匹配候选框架
4. **适用性评估**: 对每个候选框架评估适用性
5. **推荐**: 选择最适合的框架 + 理由
6. **输出**: 推荐框架 + 候选对比表
## Framework Knowledge Base
| 框架 | 适用研究类型 | 核心组件 | 优势 | 局限 |
|------|------------|---------|------|------|
| PICO/PICOTS | 量化/干预 | P-I-C-O(-T-S) | 结构清晰,广泛认可 | 不适合定性研究 |
| SPIDER | 定性 | S-PI-D-E-R | 专为定性设计 | 不适合干预研究 |
| SPICE | 评估 | S-P-I-C-E | 评估导向 | 范围较窄 |
| ECLIPSE | 混合方法 | E-C-L-I-P-S-E | 全面,适合复杂研究 | 组件多,可能过度 |
## Output Format
```
{
recommended: "PICO",
rationale: "...",
candidates: [
{ framework: "PICO", fit_score: 9, reason: "..." },
{ framework: "SPIDER", fit_score: 4, reason: "..." }
]
}
```