review-paper

name	review-paper
description	Rigorous manuscript review workflow for academic papers in PDF, Word, TeX, or Quarto format. Use when Codex is asked to review a manuscript, produce a referee report, evaluate identification or econometric specifications, or generate Markdown/PDF paper-review outputs from files in papers/ or direct manuscript paths.

Manuscript Review

Produce a thorough, constructive academic referee report. The report should cover the paper's argument structure, literature positioning, identification strategy, econometric specification, writing quality, presentation, and likely referee objections.

Inputs

Accept a direct path to a .pdf, .docx, .tex, or .qmd manuscript, or a filename located in papers/.

Resolve the manuscript path in this order:

1. The path provided by the user.
2. papers/<provided filename>.
3. A case-insensitive partial filename match under papers/.

If the manuscript cannot be read fully, state the limitation explicitly in the report and avoid inventing page, table, equation, or section references.

Workflow

1. Read the manuscript end to end. For long PDFs or Word manuscripts, extract and review the text in manageable chunks. For .docx files, use python scripts/extract_docx_text.py <manuscript.docx> when a plain-text copy is useful. Keep extraction artifacts out of papers/; use review_reports/extracted_[sanitized_name].txt only as a temporary working file.
2. Evaluate the paper across the six review dimensions below.
3. Identify 3 to 5 strong referee objections.
4. Write the Markdown report to review_reports/paper_review_[sanitized_name].md.
5. Export the PDF with python scripts/export_review_pdf.py review_reports/paper_review_[sanitized_name].md when Pandoc and XeLaTeX are available.
6. Verify the Markdown file exists.
7. Verify the PDF exists, or record why PDF export could not be run.
8. Delete any extracted manuscript .txt working files after verification.
9. Verify that no extracted .txt artifact remains in papers/ after the Markdown and PDF outputs are generated.

Review Dimensions

Argument Structure

• Is the research question clearly stated?
• Does the introduction motivate the question effectively?
• Is the logical flow sound from question to method to results to conclusion?
• Are conclusions supported by the evidence?
• Are limitations acknowledged?

Literature Positioning

• Are the key papers cited?
• Is prior work characterized accurately?
• Is the contribution clearly differentiated from existing work?
• Are there missing citations that a referee would flag?

Identification Strategy

• Is the causal claim credible?
• What are the identifying assumptions, and are they stated explicitly?
• Are there threats from omitted variables, reverse causality, measurement error, or selection?
• Are robustness checks adequate?
• Is the estimator appropriate for the research design?

Econometric Specification

• Are standard errors appropriate, including clustering or robustness choices?
• Is the functional form appropriate?
• Are sample selection issues addressed?
• Are multiple testing concerns relevant?
• Are point estimates economically meaningful, not only statistically significant?

Writing Quality

• Is the prose clear, concise, and academically toned?
• Is notation consistent?
• Does the abstract summarize the paper effectively?
• Are tables and figures self-contained?

Presentation

• Are tables and figures well designed?
• Are labels, notes, sources, and definitions clear?
• Are there typographical, grammatical, or formatting issues?
• Is the paper length appropriate for the contribution?

Output Format

Use this Markdown structure:

# Manuscript Review: [Paper Title]

**Date:** [YYYY-MM-DD]
**Reviewer:** Codex review-paper workflow
**File:** [path to manuscript]

## Summary Assessment

**Overall recommendation:** [Strong Accept / Accept / Revise & Resubmit / Reject]

[2-3 paragraph summary: main contribution, strengths, and key concerns]

## Strengths

1. [Strength 1]
2. [Strength 2]
3. [Strength 3]

## Major Concerns

### MC1: [Title]
- **Dimension:** [Identification / Econometrics / Argument / Literature / Writing / Presentation]
- **Issue:** [Specific description]
- **Suggestion:** [How to address it]
- **Location:** [Section/page/table if applicable]

## Minor Concerns

### mc1: [Title]
- **Issue:** [Description]
- **Suggestion:** [Fix]

## Referee Objections

These are the tough questions a top referee would likely raise:

### RO1: [Question]
**Why it matters:** [Why this could be fatal]
**How to address it:** [Suggested response or additional analysis]

## Specific Comments

[Line-by-line or section-by-section comments, if any]

## Summary Statistics

| Dimension | Rating (1-5) |
|-----------|-------------|
| Argument Structure | [N] |
| Identification | [N] |
| Econometrics | [N] |
| Literature | [N] |
| Writing | [N] |
| Presentation | [N] |
| **Overall** | **[N]** |

Do not append workflow verification notes to the Markdown review report. Report verification status only in the final response to the user.

Principles

• Be constructive: every criticism should come with a practical suggestion.
• Be specific: reference exact sections, pages, equations, tables, or figures only when verified.
• Think like a rigorous journal referee: distinguish fatal flaws from fixable issues.
• Acknowledge what the paper does well.
• Do not fabricate details from unreadable or inaccessible manuscript sections.