write-paper

Write Paper Draft

Draft an academic manuscript from existing analysis outputs. Follows propose-first — the outline is always shown for approval before writing begins.

Input: $ARGUMENTS — a paper title, research question, or from spec to read from quality_reports/specs/.

Step 1: Orient

Read project context:

• CLAUDE.md — project name, author, institution
• references/domain-profile.md — field (used to calibrate outline conventions in Step 3)
• Most recent file in quality_reports/specs/ — research question and identification strategy
• Most recent file in quality_reports/lit_review_*.md — related literature (if it exists)
• Most recent file in quality_reports/research_ideation_*.md — hypotheses (if it exists)

If no spec exists, ask the user for the core research question before proceeding.

Step 2: Inventory Analysis Outputs

Glob for available outputs:

• output/tables/**/*.tex — regression tables
• output/tables/**/*.html — HTML table versions
• output/figures/**/*.pdf, output/figures/**/*.png — figures
• output/**/*.rds, output/**/*.pkl, output/**/*.parquet — saved objects

List what exists. Note any gaps (e.g., "no summary statistics table found").

Step 3: Propose Outline (WAIT FOR APPROVAL)

Draft a paper outline using this standard structure:

1. Abstract
2. Introduction
3. Background
4. Data & Methods
5. Results
6. Discussion
7. Appendix
References

The structure is flexible. Default to the hardcoded outline above unless (a) the user requests a different structure, or (b) the field in references/domain-profile.md has strong conventions that warrant deviation (e.g., merging Background into Introduction, splitting Data & Methods into separate Data and Empirical Strategy sections, or reordering sections). If you deviate from the default, state explicitly why and confirm with the user before writing.

Adapt the writing style — not the structure, by default — to the field from references/domain-profile.md. A political science paper reads differently than an econometrics paper even with the same section headings: econ prose leads with identification and is terse about theory; poli sci prose leads with theoretical framing and treats measurement validity as a first-class concern; sociology leans on theoretical framework prose. Mirror the conventions of the field's top journals when drafting each section.