Menu
Challenge the paper's arguments, methodology, and framing with 5-7 specific critical questions. Identifies weaknesses before reviewers do.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Search for published papers that justify a specific methodological decision. Returns precedents diversified across empirical domains, with top papers read in full via PDF.
Build the "Related Literature" section by mapping each literature stream and drafting its contribution paragraph, using Semantic Scholar. Proposes the streams with you first, then searches per stream and reads 1-2 anchor papers in full.
Stage and commit changes, grouped into logical commits (e.g. writing vs. code). Optionally push with the "push" flag.
Compile the LaTeX research paper with latexmk. Use when compiling the draft after making changes.
Check that all dependencies required to run this template are installed. Verifies R, LaTeX, required packages, and runs a full smoke test of the pipeline.
Validate bibliography entries against citations in all paper sections. Find missing entries and unused references.
| name | devils-advocate |
| description | Challenge the paper's arguments, methodology, and framing with 5-7 specific critical questions. Identifies weaknesses before reviewers do. |
| disable-model-invocation | true |
| argument-hint | [section name, or 'full paper'] |
Critically examine the paper and challenge its arguments with 5-7 specific questions a tough [Target Journal] referee would ask.
.claude/rules/knowledge-base.mdGenerate 5-7 challenges from these categories:
"What if [key independent variable] is endogenous to [outcome]?"
"Could this result be driven by [alternative mechanism] rather than [proposed mechanism]?"
"Does this generalize beyond [your specific context / sample] to [broader population or setting]?"
"Is [your measure of X] the right way to capture [underlying construct] for [your population]?"
"What happens with a different [lag structure / functional form / sample restriction / control set]?"
"How is this different from [specific paper] that studies similar dynamics?"
"The coefficient is statistically significant, but is the effect economically meaningful?"
# Devil's Advocate: [Section or Full Paper]
## Challenges
### Challenge 1: [Category] -- [Short title]
**Question:** [The specific critical question]
**Why it matters:** [What a referee would worry about]
**Suggested response:** [How to address this in the paper]
**Section affected:** [Which section(s)]
**Severity:** [High / Medium / Low]
[Repeat for 5-7 challenges]
## Summary Verdict
**Strengths:** [2-3 things done well]
**Critical changes:** [0-2 changes needed before submission]
**Suggested improvements:** [2-3 nice-to-have improvements]