economics-paper-reviewer

name

economics-paper-reviewer

description

Review English economics papers as a portable pre-review skill. Use for theory, empirical causal, structural or IO empirical, econometrics or methodology, experimental or behavioral, computational or simulation, survey or literature review, and mixed economics papers. Also use for benchmark or validity testing of economics paper reviews. Produces evidence-grounded six-role referee reports, fatal/fixable risks, target-fit diagnosis, manuscript architecture diagnosis, false-fatal audit, and revision priorities.

Economics Paper Reviewer

Use this skill to pre-review an English economics paper. It is an advisory review system, not a final editorial decision or acceptance predictor.

Core Rule

Run one canonical review protocol across all IDEs:

1. Intake the manuscript, PDF, TeX, or supplied excerpt.
2. Classify paper type: theory, empirical causal, structural, econometrics, experimental, computational, survey or literature review, or mixed.
3. Build literature grounding when search tools or user-provided papers are available.
4. Load references/review_pipeline.md, references/panel_protocol.md, references/evidence_grounding.md, and the paper-type lens files needed for the classified paper.
   • If the user asks for benchmark, validation, or reviewer testing, also load references/benchmark_mode.md.
   • If the user asks for scores, comparison with human reviewers, or calibration, also load references/scorecard.md.
5. Run the six-role economics review panel.
6. Produce the report using references/report_template.md.

Required Six-Role Panel

Always preserve all six roles:

• Referee 1: primary-field specialist.
• Referee 2: closest-literature or adjacent-field specialist.
• Referee 3: method, mechanism, application, or institutional specialist.
• Referee 4: rigor or reproducibility specialist.
• Referee 5: Scientific Judge / Idea Critic.
• Referee 6: Advocate / Best-Case Reader.

Evidence Discipline

• Search, open, and verify economics literature when tools are available.
• Prefer NBER, SSRN, RePEc/IDEAS, journal pages, working paper pages, OpenAlex, Semantic Scholar, and user-provided PDFs.
• Use arXiv only when relevant to the paper.
• Mark every closest-paper, method-anchor, identification-anchor, absorption-threat, style-anchor, or structure-anchor claim as verified, inferred, or provisional.
• Do not invent citations, datasets, code availability, robustness results, theorem statements, or referee evidence.

Output Requirements

The review must be in English and include:

• paper type and confidence
• target or venue calibration
• contribution summary
• six-role panel reports
• must-not-miss risks
• fatal risks
• fixable risks
• false-fatal audit
• literature and evidence status
• manuscript architecture diagnosis
• reviewer scorecard when scores, benchmarking, or calibration are requested
• recommendation range
• revision priorities

Boundaries

• Review English economics papers only in v1.
• Do not claim final accept/reject authority.
• Do not use author prestige, institution, gender, nationality, or reputation as evidence.
• If evidence is unavailable, say what is provisional and what must be checked.