Research Ideas

Research Idea Generator

When to Use

After major literature additions, or when looking for new directions.

Steps

1. Read all @*.md notes in literature/zotero-notes/ — extract YAML frontmatter (citekey, paper_type, tags, authors, year, journal) and note content
2. Also read: literature/missing-lit.md, existing strategy memos, literature/synthesis/frontier-map.md, .claude/references/domain-profile.md
3. Generate ideas, requiring for each:

• Research question (one sentence)
• Empirical strategy (DiD/IV/RDD/SC + what provides variation)
• Data feasibility (what exists, what's accessible, what's missing)
• Contribution (what's new relative to the frontier)
• Feasibility score (1–5: 1 = publishable with existing data, 5 = data may not exist)
• Target journal (based on journal-profiles.md)
• Relation to existing papers (complement, extension, or independent)

1. Append to literature/research-ideas.md with date stamp

Output Format

Prepend to literature/research-ideas.md:

## YYYY-MM-DD — Research Ideas

### Idea 1: [One-sentence question]
- **Strategy:** [Estimator + variation source]
- **Data:** [Dataset(s) + accessibility]
- **Contribution:** [What's new]
- **Feasibility:** [N]/5
- **Target journal:** [Journal name]
- **Relation:** [How it connects to Papers 1–3]

### Idea 2: ...

Important

• Follow append-only protocol
• Flag ideas that complement existing three objectives
• Be specific about identification — "use DiD" is insufficient; specify what provides the treatment variation

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: saptarsibhowmick
• Source: saptarsibhowmick/claude-workflow-for-econ-phd
• License: MIT

Install and usage instructions live in the source repository linked above.