paper-review

name	paper-review
description	中英文论文全流程审阅与质量审计系统。当用户说"审稿"、"论文评审"、"帮我看看论文"、"模拟审稿"、"paper review"、"自检论文"、"论文质量检查"、"检查论文完整性"、"验证论文结构"、"gate 这篇论文"、"audit 我的 manuscript"、"deep review"、"quick audit"、"看看论文能不能投"、"查重自查"、"检查提交前准备"、或提到论文预提交检查、修改后再审、委员会评审、审稿人质询等场景时触发。整合了深度审计 (v4.2)、5 角色委员会评审、review-paper 6 维度评估框架、以及轻量级论文自检清单。支持 LaTeX/Typst/PDF 格式，中英文论文。
metadata	{"category":"academic-writing","tags":["audit","deep-review","paper","review","pdf","latex","typst","chinese","english","reviewer","gate","re-audit","self-review","committee","methodology","evidence-quality","GRADE","Cochrane-ROB"],"version":"4.2","last_updated":"2026-05-27","integrated":["paper-review v4.2","scientific-critical-thinking (methodology assessment, evidence quality)"]}
argument-hint	[paper.tex|paper.typ|paper.pdf] [--mode quick-audit|deep-review|gate|re-audit|polish] [--focus full|editor|theory|literature|methodology|logic] [--venue VENUE] [--previous-report PATH] [--literature-search] [--scholar-eval] [--format markdown|json]
allowed-tools	Read, Glob, Grep, Bash(uv *), Task

Paper Review Skill v4.2 (Merged)

paper-review merges three previous skills into one unified system:

• paper-audit v4.2 (core) — deep-review-first audit, 5-mode workflow, 5-role academic committee, 16-dimension issue taxonomy, Python scripts
• review-paper — top-journal referee 6-dimension evaluation framework (Argument Structure, Identification Strategy, Econometric Specification, Literature Positioning, Writing Quality, Presentation) with 3-5 referee objections
• paper-self-review — lightweight pre-submission quality checklist (Structure, Logic Consistency, Citation Completeness, Figure/Table Quality, Writing Clarity) with section-by-section and final-verdict references

Core Principle

This skill behaves like a serious reviewer at a top journal or conference: find technical, methodological, claim-level, and cross-section issues; keep script-backed findings separate from reviewer judgment; return structured issue bundles plus revision roadmaps.

Do not use for direct source editing, sentence rewriting, or compilation repair. Route those to format-specific writing skills instead.

What This Skill Produces

Mode	Output
quick-audit	Fast submission-readiness screen with lightweight self-check checklist and script-backed findings
deep-review	Reviewer-style structured issue bundle with major/moderate/minor findings, plus 6-dimension referee report and 3-5 referee objections
gate	PASS/FAIL decision calibrated for submission blockers, with EIC desk-reject screening
re-audit	Compare current issue bundle against a previous audit with root-cause-aware status labels
polish	Precheck-only handoff into a polishing workflow

For deep-review, the primary outputs are:

• final_issues.json
• overall_assessment.txt
• review_report.md (includes 6-dimension scoring and referee objections when in full mode)
• revision_roadmap.md

Critical Rules

• Never rewrite the paper source unless the user explicitly switches to an editing skill.
• Never fabricate references, baselines, or reviewer evidence.
• Always distinguish [Script] from [LLM] findings.
• Always anchor reviewer findings to a quote, section, or exact textual location.
• Be conservative with OCR noise, formatting quirks, and obvious copy-editing trivia.
• Review like a careful reader: understand the author's intended meaning before flagging an issue.

Mode Selection

Requested intent	Mode
"check my paper", "quick audit", "submission readiness", "自检论文", "论文质量检查"	quick-audit
"review my paper", "simulate peer review", "harsh review", "deep review", "审稿", "论文评审", "模拟审稿"	deep-review
"is this ready to submit", "gate this submission", "blockers only", "gate 这篇论文", "看看论文能不能投"	gate
"did I fix these issues", "re-audit", "compare against old review", "修改后再审"	re-audit
"polish the writing, but only if safe"	polish

Legacy aliases still work for one compatibility cycle:

• self-check -> quick-audit
• review -> deep-review

Committee Focus Routing (deep-review)

For deep-review, use the Academic Pre-Review Committee by default. This is a 5-role review pass:

1. Editor (desk-reject screen)
2. Reviewer 1 (theory contribution)
3. Reviewer 3 (literature dialogue / gap)
4. Reviewer 2 (methodology transparency)
5. Reviewer 4 (logic chain)

If the user requests a single dimension, run only the matching committee role(s).

If --focus ... is provided, it overrides keyword inference:

• --focus full (default)
• --focus editor|theory|literature|methodology|logic

Keyword map (English + Chinese):

• editor: "desk reject", "pre-screen", "editor", "EIC", "主编", "预筛", "初筛"
• theory: "theory", "contribution", "novelty", "theoretical dialogue", "理论", "贡献", "创新性"
• literature: "related work", "literature", "research gap", "citation", "文献", "综述", "Research Gap", "引用"
• methodology: "methods", "sample", "coding", "data", "design", "SRQR", "方法", "样本", "编码", "数据", "研究设计", "透明度"
• logic: "logic", "argument", "causal", "structure", "论证", "因果", "逻辑", "结构"

Output language: match the user's request language. If ambiguous, match the paper language.

Review Standard

Read these references before running reviewer-style work:

1. references/REVIEW_CRITERIA.md
2. references/DEEP_REVIEW_CRITERIA.md
3. references/CHECKLIST.md
4. references/CONSOLIDATION_RULES.md
5. references/ISSUE_SCHEMA.md

16-Dimension Issue Taxonomy (from paper-audit deep-review)

1. formula / derivation errors
2. notation inconsistency
3. prose vs formal object mismatch
4. numerical inconsistency
5. missing justification
6. overclaim or claim inaccuracy
7. ambiguity that can mislead a careful reader
8. underspecified methods / missing information
9. internal contradiction
10. self-consistency of standards
11. table structure violations
12. abstract structural incompleteness
13. theory contribution deficiency
14. qualitative methodology opacity
15. pseudo-innovation / straw man
16. paragraph-level argument incoherence

6-Dimension Evaluation Framework (from review-paper)

For deep-review mode, also evaluate the paper across these referee-style dimensions (integrated as a structured reference for the committee):

Dimension 1: Argument Structure

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

Dimension 2: Identification Strategy

• Is the causal claim credible?
• What are the key identifying assumptions? Are they stated explicitly?
• Are there threats to identification (omitted variables, reverse causality, measurement error)?
• Are robustness checks adequate?
• Is the estimator appropriate for the research design?

Dimension 3: Econometric / Methodological Specification

• Correct standard errors (clustered? robust? bootstrap?)?
• Appropriate functional form?
• Sample selection issues?
• Multiple testing concerns?
• Are point estimates economically meaningful (not just statistically significant)?

Dimension 4: Literature Positioning

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

Dimension 5: Writing Quality

• Clarity and concision
• Academic tone
• Consistent notation throughout
• Abstract effectively summarizes the paper
• Tables and figures are self-contained (clear labels, notes, sources)

Dimension 6: Presentation

• Are tables and figures well-designed?
• Is notation consistent throughout?
• Are there any typos, grammatical errors, or formatting issues?
• Is the paper the right length for the contribution?

Lightweight Self-Check Checklist (from paper-self-review, for quick-audit mode)

For quick-audit mode, use this checklist alongside the script-backed audit:

Paper Quality Checklist:
- [ ] Abstract includes problem, method, results, contributions
- [ ] Introduction clearly states research motivation
- [ ] Method is reproducible
- [ ] Results support conclusions
- [ ] Discussion addresses limitations
- [ ] All figures/tables have captions
- [ ] Citations are complete and accurate

Additional section-level checks (see references/SECTION-CHECKLIST.md):

• Abstract: problem clear, method identifiable, main result quantified, contribution stated
• Introduction: research gap explicit, question/hypothesis visible, contribution list concrete
• Method: enough detail to reproduce, assumptions named, key design choices justified
• Results: numbers support claims, uncertainty/significance reported, baselines and ablations interpretable
• Discussion: limitations named, failure cases acknowledged, next-step implications reasonable

Verdict buckets (see references/FINAL-VERDICT.md):

• ready with minor edits
• needs moderate revision
• not ready for submission

Always report: top 3 blocking issues, top 3 polish issues, missing evidence/citations, figure/table risks.

Referee Objections (from review-paper, for deep-review mode)

For deep-review mode in full committee configuration, after the 6-dimension evaluation is complete, generate 3-5 referee objections — the tough questions a top referee would raise. Each objection should include:

• Question: The specific hard question a referee would ask
• Why it matters: Why this could be fatal to the paper's contribution
• How to address it: Suggested response or additional analysis

These objections should be integrated into review_report.md under a dedicated "Referee Objections" section.

Methodology & Evidence Quality Assessment

This section integrates the scientific-critical-thinking methodology evaluation framework into the paper review process.

Methodology Critique

Evaluate research methodology across these dimensions:

Research Design:

• Is the design appropriate for the research question? (RCT vs observational vs qualitative)
• Are there threats to internal validity (selection bias, history, maturation, testing effects)?
• Is the study adequately powered? Sample size justification?

Experimental Validity:

• Construct validity: do the measures actually capture the intended constructs?
• External validity: can findings be generalized beyond the study sample?
• Ecological validity: does the experimental setting reflect real-world conditions?

Statistical Validity Assessment

Check for common statistical issues:

Issue	What to Look For
Multiple comparisons	Are corrections applied (Bonferroni, FDR)?
P-hacking	Are analyses pre-registered? Any post-hoc decisions?
Effect size reporting	Are effect sizes and confidence intervals reported, not just p-values?
Assumption violations	Are normality, homoscedasticity, independence assumptions checked?
Outlier handling	Are exclusion criteria pre-specified and reported?
Missing data	Is the amount and pattern of missing data reported? How was it handled?

Bias Identification

Apply structured bias assessment frameworks:

Cochrane Risk of Bias (ROB) Tool — for clinical trials:

1. Random sequence generation (selection bias)
2. Allocation concealment (selection bias)
3. Blinding of participants and personnel (performance bias)
4. Blinding of outcome assessment (detection bias)
5. Incomplete outcome data (attrition bias)
6. Selective reporting (reporting bias)
7. Other sources of bias

ROBINS-I — for non-randomized studies:

1. Confounding
2. Selection of participants
3. Classification of interventions
4. Deviations from intended interventions
5. Missing data
6. Measurement of outcomes
7. Selection of reported result

Evidence Quality Frameworks

GRADE (Grading of Recommendations Assessment, Development and Evaluation):

Level	Definition
High	Further research is very unlikely to change confidence
Moderate	Further research is likely to have an important impact
Low	Further research is very likely to change confidence
Very Low	Any estimate is very uncertain

Factors that lower quality: risk of bias, inconsistency, indirectness, imprecision, publication bias. Factors that raise quality: large effect, dose-response gradient, plausible confounding would reduce effect.

Integration into Review Workflow

• For deep-review mode: The methodology committee role (committee_methodology_agent.md) should apply these frameworks when evaluating methodology transparency and statistical validity.
• For quick-audit mode: Include evidence quality levels (using GRADE) in the assessment of each major finding.
• For gate mode: Flag fatal methodological flaws as blockers (e.g., unaddressed confounding, underpowered study, p-hacking).

Workflow

Common Step 0

Parse $ARGUMENTS and infer the mode if the user did not provide one. State the inferred mode before running commands if you had to infer it.

quick-audit

1. Run:

   uv run python -B "$SKILL_DIR/scripts/audit.py" <paper> --mode quick-audit ...
   

2. Present a concise report:
   • Submission Blockers first
   • then Quality Improvements
   • then checklist items from the lightweight self-check checklist (Section Checklist + Final Verdict)
   • mark quick-audit findings with [Script] provenance
3. If the user clearly wants reviewer-depth critique after the quick screen, escalate to deep-review.

deep-review

Use this as the default reviewer-style path. In full committee mode, the review should incorporate the 6-dimension evaluation framework (Argument Structure, Identification Strategy, Econometric Specification, Literature Positioning, Writing Quality, Presentation) and produce 3-5 referee objections alongside the standard audit output.

Phase 1: Prepare workspace

Run:

uv run python -B "$SKILL_DIR/scripts/prepare_review_workspace.py" <paper> --output-dir ./review_results

This creates:

• full_text.md
• metadata.json
• section_index.json
• claim_map.json
• paper_summary.md
• sections/*.md
• comments/
• references/ (minimal copies for reviewer agents)
• committee/ (committee reviewer artifacts)

Phase 2: Phase 0 automated audit

Run:

uv run python -B "$SKILL_DIR/scripts/audit.py" <paper> --mode deep-review ...

Treat this as Phase 0 only. It supplies script-backed context and scores, not the final review.

Phase 3: Committee + Review Lanes

Phase 3A: Academic Pre-Review Committee (default)

Decide committee focus:

• If --focus ... is provided, use it.
• Otherwise infer from the user request using the keyword map in "Committee Focus Routing".
• If nothing matches, default to full (all five roles).

Dispatch the committee reviewers (in this exact order) and have them write artifacts into the workspace:

1. agents/committee_editor_agent.md
   • write: committee/editor.md
   • write: comments/committee_editor.json
2. agents/committee_theory_agent.md
   • write: committee/theory.md
   • write: comments/committee_theory.json
3. agents/committee_literature_agent.md
   • write: committee/literature.md
   • write: comments/committee_literature.json
4. agents/committee_methodology_agent.md
   • write: committee/methodology.md
   • write: comments/committee_methodology.json
5. agents/committee_logic_agent.md
   • write: committee/logic.md
   • write: comments/committee_logic.json

If subagents are unavailable, run the committee reviewers inline, but keep the same file outputs.

Then write: committee/consensus.md

• include: overall score (1-10), ordered priorities, and the top 3 issues to fix first
• scoring formula:
  • start at 9.0
  • subtract: 1.5 * (# major) + 0.7 * (# moderate) + 0.2 * (# minor)
  • floor at 1.0
  • if Editor verdict is Desk Reject, cap at 4.0

Note: render_deep_review_report.py automatically embeds committee/*.md into review_report.md when present.

Phase 3B: 6-Dimension Evaluation + Referee Objections (from review-paper)

For full committee deep-review, perform an additional structured evaluation pass using the 6-Dimension Evaluation Framework (see above). This should be done by the committee as a collective assessment after individual reviews. The output covers:

• Ratings (1-5) for each of the 6 dimensions
• 3-5 referee objections (hard questions a top referee would ask)
• These are written into review_report.md under "6-Dimension Evaluation" and "Referee Objections" sections

Phase 3C: Section and cross-cutting review lanes (coverage)

Read:

• references/SUBAGENT_TEMPLATES.md
• references/REVIEW_LANE_GUIDE.md

Then dispatch reviewer tasks for:

• section lanes
  • introduction / related work
  • methods
  • results
  • discussion / conclusion
  • appendix, if present
• cross-cutting lanes
  • claims vs evidence
  • notation and numeric consistency
  • evaluation fairness and reproducibility
  • self-standard consistency
  • prior-art and novelty grounding

Each lane writes a JSON array into comments/.

If subagents are unavailable, use the built-in deterministic fallback lane pass in scripts/audit.py so the workflow still writes lane-compatible JSON into comments/ before consolidation.

Phase 4: Consolidation

Run:

uv run python -B "$SKILL_DIR/scripts/consolidate_review_findings.py" <review_dir>
uv run python -B "$SKILL_DIR/scripts/verify_quotes.py" <review_dir> --write-back
uv run python -B "$SKILL_DIR/scripts/render_deep_review_report.py" <review_dir>

Consolidation rules:

• merge exact duplicates
• keep distinct paper-level consequences separate even if they share a root cause
• preserve singleton findings unless clearly false positive
• assign comment_type, severity, confidence, and root_cause_key

Phase 5: Present result

Summarize:

• 1 short paragraph overall assessment
• counts of major / moderate / minor issues
• 3 highest-priority revision items
• 6-dimension rating summary table (if full committee mode)
• 3-5 referee objections summary (if full committee mode)
• path to review_report.md and final_issues.json

gate

1. Run:

   uv run python -B "$SKILL_DIR/scripts/audit.py" <paper> --mode gate ...
   

2. EIC Screening (Phase 0.5): Read agents/editor_in_chief_agent.md and perform the editor-in-chief desk-reject screening on the paper's title, abstract, and introduction. This evaluates pitch quality, venue fit, fatal flaws, and presentation baseline. A desk-reject verdict is a gate blocker.
3. Report PASS/FAIL.
4. Present EIC screening results first (verdict + score + justification).
5. List blockers next.
6. Keep advisory items separate from blockers.
7. For IEEE pseudocode checks, make it explicit which issues are mandatory and which are only IEEE-safe recommendations.

re-audit

1. Requires --previous-report PATH.
2. Run:

   uv run python -B "$SKILL_DIR/scripts/audit.py" <paper> --mode re-audit --previous-report <path> ...
   

3. If both old and new final_issues.json bundles are available, also run:

   uv run python -B "$SKILL_DIR/scripts/diff_review_issues.py" <old_final_issues.json> <new_final_issues.json>
   

4. Present:
   • root-cause-aware status labels: FULLY_ADDRESSED, PARTIALLY_ADDRESSED, NOT_ADDRESSED, NEW
   • use structured prior issue bundles when available, but still accept Markdown previous reports

polish

1. Run the audit precheck:

   uv run python -B "$SKILL_DIR/scripts/audit.py" <paper> --mode polish ...
   

2. If blockers exist, stop and report them.
3. Only proceed into polishing if the precheck is safe.

Output Contract

For deep-review, the final issue schema is:

{
  "title": "short issue title",
  "quote": "exact quote from paper",
  "explanation": "why this matters and what remains problematic",
  "comment_type": "methodology|claim_accuracy|presentation|missing_information",
  "severity": "major|moderate|minor",
  "confidence": "high|medium|low",
  "source_kind": "script|llm",
  "source_section": "methods",
  "related_sections": ["results", "appendix"],
  "root_cause_key": "shared-normalized-key",
  "review_lane": "claims_vs_evidence",
  "gate_blocker": false,
  "quote_verified": true
}

Always prefer:

• exact quotes over vague paraphrase
• evidence-backed findings over style commentary
• issue bundle + roadmap over raw script dumps

References

File	Source	Purpose
references/REVIEW_CRITERIA.md	paper-audit	top-level audit scoring and mapping
references/DEEP_REVIEW_CRITERIA.md	paper-audit	deep-review-specific issue taxonomy (16 dimensions) and leniency rules
references/CONSOLIDATION_RULES.md	paper-audit	deduplication and root-cause merge policy
references/ISSUE_SCHEMA.md	paper-audit	canonical JSON schema
references/REVIEW_LANE_GUIDE.md	paper-audit	section lanes and cross-cutting lanes
references/SUBAGENT_TEMPLATES.md	paper-audit	reviewer task templates
references/QUICK_REFERENCE.md	paper-audit	CLI and mode cheat sheet
references/CHECKLIST.md	paper-audit	general audit checklist
references/AUDIT_GUIDE.md	paper-audit	audit guide
references/CHANGELOG.md	paper-audit	version history
references/FORBIDDEN_TERMS.md	paper-audit	terms to avoid
references/LITERATURE_GROUNDING_GUIDE.md	paper-audit	literature grounding methodology
references/POLISH_GUIDE.md	paper-audit	polish mode guide
references/QUALITATIVE_STANDARDS.md	paper-audit	qualitative research standards
references/SCHOLAR_EVAL_GUIDE.md	paper-audit	scholar evaluation guide
references/SCORING_SYSTEMS.md	paper-audit	venue-specific scoring systems
references/TROUBLESHOOTING.md	paper-audit	troubleshooting guide
references/VENUE_RULES.md	paper-audit	venue-specific submission rules
references/editorial_decision_standards.md	paper-audit	editorial decision standards
references/quality_rubrics.md	paper-audit	quality rubrics
references/SECTION-CHECKLIST.md	paper-self-review	section-by-section review questions for quick-audit
references/FINAL-VERDICT.md	paper-self-review	submission readiness verdict buckets and blocking issues

Scripts

Script	Purpose
scripts/audit.py	Phase 0 audit and mode entrypoint
scripts/prepare_review_workspace.py	create deep-review workspace
scripts/build_claim_map.py	extract headline claims and closure targets
scripts/consolidate_review_findings.py	deduplicate comment JSONs
scripts/verify_quotes.py	verify exact quote presence
scripts/render_deep_review_report.py	render final Markdown report
scripts/diff_review_issues.py	compare old vs new issue bundles
scripts/pdf_parser.py	PDF parsing utilities
scripts/parsers.py	general parsing utilities
scripts/report_generator.py	report generation utilities
scripts/detect_language.py	language detection
scripts/scoring_model.py	ML-based scoring model
scripts/check_citations.py	citation checking
scripts/check_references.py	reference validation
scripts/literature_search.py	literature search
scripts/literature_compare.py	literature comparison
scripts/scholar_eval.py	scholar evaluation
scripts/visual_check.py	visual element checking
scripts/models/scoring_model.json	scoring model weights

Agents

Committee agents (deep-review default):

• agents/committee_editor_agent.md
• agents/committee_theory_agent.md
• agents/committee_literature_agent.md
• agents/committee_methodology_agent.md
• agents/committee_logic_agent.md

Section and cross-cutting lane agents:

• agents/section_reviewer_agent.md
• agents/claims_evidence_reviewer_agent.md
• agents/notation_consistency_reviewer_agent.md
• agents/evaluation_fairness_reviewer_agent.md
• agents/self_consistency_reviewer_agent.md
• agents/prior_art_reviewer_agent.md
• agents/synthesis_agent.md
• agents/editor_in_chief_agent.md — EIC desk-reject screener (used in gate mode)

Specialized deep-review agents:

• agents/critical_reviewer_agent.md — devil's advocate with C3-C5 checks
• agents/domain_reviewer_agent.md — domain expertise with A1-A7 assessments
• agents/methodology_reviewer_agent.md — methodology rigor with B3-B10 checks
• agents/literature_reviewer_agent.md — evidence-based literature verification (optional, --literature-search)

Examples

• "Review this manuscript like a serious conference reviewer and tell me the biggest validity risks."
• "帮我审一下这篇论文，用 6 维度框架全面评估，再加 3 个审稿人质询问题。"
• "Run a quick audit on paper.tex and tell me what blocks submission."
• "Gate this IEEE submission and separate blockers from recommendations."
• "Re-audit this revision against my previous report."
• "自检一下我的论文，用 checklist 逐项检查完整性。"
• "模拟审稿，deep review 模式，全委员会评审。"