fengmo11 Paper Five Reviewer Panel
Use this skill after the manuscript has a complete draft, figures, references, and data/code availability notes. The goal is to simulate a serious journal review round before the real submission.
Inputs
- Full manuscript draft.
- Target journal, conference, or venue tier.
- Figures, tables, supplementary material, and captions.
- References or BibTeX.
- Data/code availability statement.
- Known constraints: no new experiments, limited time, confidential data, or author decisions.
Reviewer Agents
Run five independent reviews. Do not let one reviewer see another review until all five reports are complete.
| Reviewer | Main Focus |
|---|---|
| Editor-in-Chief | Venue fit, novelty, significance, reader value, decision risk. |
| Methodology/Statistics Reviewer | Design validity, baseline choice, sample size, statistics, reproducibility. |
| Domain Reviewer | Literature coverage, theoretical framing, field contribution, missing citations. |
| Figure/Data Reviewer | Figure logic, panel evidence, captions, source data, data/code availability. |
| Devil's Advocate | Core claim attacks, alternative explanations, overclaim, hidden assumptions. |
Workflow
- Build a reviewer configuration card for the target venue and field.
- Run the five reviewer agents independently.
- Require each reviewer to provide a decision recommendation, confidence, major concerns, minor concerns, required evidence, and scorecard.
- Synthesize the reviews into an editorial decision.
- Convert every actionable comment into stable IDs.
- Route the comment map to
paper-review-response. - After revision, run re-review verification before
paper-submission-gate.
Decision Labels
- Accept-ready: only minor edits remain.
- Minor revision: no new core evidence needed.
- Major revision: important evidence, analysis, structure, citation, or figure changes are needed.
- Reject-risk: contribution, method, evidence, or venue fit is not yet credible.
- Desk-reject-risk: the paper is misaligned with the venue or lacks a clear publishable contribution.
Output Contract
Return:
- Reviewer configuration card.
- Five independent reviewer reports.
- Editorial synthesis letter.
- Decision label and confidence.
- Major revision map with stable IDs.
- Minor revision map with stable IDs.
- Re-review checklist.
- Blocking author-input questions.
Review Integrity Rules
- Do not invent manuscript sections, results, citations, datasets, or line numbers.
- Mark any comment that requires author judgment or new experiments.
- Separate fixable writing problems from scientific validity problems.
- Make disagreements visible when reviewers conflict.
- A manuscript cannot move to submission until all high-risk comments are addressed, explicitly deferred, or acknowledged as a limitation.
When To Read References
- For reviewer roles and expected outputs, read
references/reviewer-roles.md. - For scoring and re-review criteria, read
references/review-scorecard.md.