菜单
Use when generating research hypotheses, designing experiments, or evaluating hypothesis feasibility
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when downloading academic papers from arXiv, searching for papers, or managing paper downloads
Use when coordinating the complete academic paper writing workflow, managing stage transitions, or initializing a new paper project
Use when implementing experimental code, scaffolding research projects, or following coding best practices for academic work
Use when running experiments, analyzing results, designing comparative experiments, or validating findings
Use when conducting literature review, searching for academic papers, identifying research gaps, or analyzing existing work
Use when drafting manuscript sections, formatting citations, polishing writing style, or preparing paper for submission
| name | paper-hypothesis |
| description | Use when generating research hypotheses, designing experiments, or evaluating hypothesis feasibility |
Generate and validate research hypotheses based on identified research gaps.
Based on literature review output, identify:
H1: [Independent variable] affects [dependent variable] because [mechanism]
Example:
H1: Using larger context windows improves reasoning performance
in LLM benchmarks because longer contexts provide more
contextual information for inference.
| Component | Description |
|---|---|
| Control | Baseline method to compare against |
| Treatment | Your proposed approach |
| Metrics | How to measure success |
| Statistics | Test selection, significance level |
{
"hypotheses": [
{
"id": "H1",
"statement": "...",
"variables": {"independent": "...", "dependent": "..."},
"feasibility": {"data": "available", "method": "feasible"},
"experiment_design": {...}
}
],
"recommended_experiments": ["exp1", "exp2"],
"potential_challenges": ["challenge1", "challenge2"]
}
scripts/hypothesis-generator.py: Generate hypotheses, assess feasibility, design experiments