alterlab-paper-reviewer

学术论文评审 — 多视角学术论文评审 Agent 团队

模拟完整的国际期刊同行评审流程：自动识别论文所属领域，动态配置 5 位评审人（主编 + 3 位同行评审人 + 唱反调者），从四个互不重叠的视角进行评审——方法论、领域专业知识、跨学科视角和核心论点质疑，最终生成结构化的编辑决定和修订路线图。

---

快速开始

最简命令：

Review this paper: [paste paper or provide file]

输出：

1. 自动识别论文所属领域和方法类型
2. 动态配置 5 位评审人的具体身份和专业领域
3. 5 份独立评审报告（每份来自不同视角）
4. 1 封编辑决定函 + 修订路线图

---

何时使用此技能

触发关键词

英文：review paper, peer review, manuscript review, referee report, review my paper, critique paper, simulate review, editorial review

不会触发的情况

场景	改用
撰写论文，或收到评审意见后进行修改	alterlab-paper-writer（revision / revision-coach modes）
深入研究某个主题	alterlab-deep-research
单份结构化审稿报告（由一位评审人根据清单进行评审）	alterlab-peer-review
对论文或学位论文进行基于评分标准的数字评分	alterlab-scholar-eval
检查稿件中的参考文献是否真实存在且未被撤稿	alterlab-citation-verifier

在 Claude Code 中，由真正独立的评审 Agent 组成的评审小组（4–6 位盲审评审人，每项主要疑虑都会对照稿件重新核查）以 /alterlab-workflows:review-panel 的形式提供（参见 alterlab-research-workflows）；当用户需要这种规模的独立评审时，请向其推荐。

快速模式选择指南

您的情况	推荐模式
需要全面评审（首次投稿）	full
检查修订是否回应了评审意见	re-review
快速质量评估（15 分钟）	quick
只关注方法/统计	methodology-focus
希望通过实践学习（引导式评审）	guided

不确定时：投稿前评审使用 full，修订后核查使用 re-review。

---

Agent 团队（7 个 Agent）

#	Agent	角色	阶段
1	field_analyst_agent	分析论文所属领域，动态配置 5 位评审人的身份	阶段 0
2	eic_agent	期刊主编——期刊匹配度、原创性、总体质量	阶段 1
3	methodology_reviewer_agent	同行评审人 1——研究设计、统计有效性、可复现性	阶段 1
4	domain_reviewer_agent	同行评审人 2——文献覆盖、理论框架、领域贡献	阶段 1
5	perspective_reviewer_agent	同行评审人 3——跨学科联系、实际影响、质疑基本假设	阶段 1
6	devils_advocate_reviewer_agent	唱反调者——质疑核心论点、检测逻辑谬误、提出最有力的反驳	阶段 1
7	editorial_synthesizer_agent	综合所有评审意见，识别共识与分歧，并作出编辑决定	阶段 2

---

编排工作流（3 个阶段）

User: "Review this paper"
     |
=== Phase 0: FIELD ANALYSIS & PERSONA CONFIGURATION ===
     |
     +-> [field_analyst_agent] -> Reviewer Configuration Card (x5)
         - Reads the complete paper
         - Identifies: primary discipline, secondary discipline, research paradigm, methodology type, target journal tier, paper maturity
         - Dynamically generates specific identities for 5 reviewers:
           * EIC: Which journal's editor, area of expertise, review preferences
           * Reviewer 1 (Methodology): Methodological expertise, what they particularly focus on
           * Reviewer 2 (Domain): Domain expertise, research interests
           * Reviewer 3 (Perspective): Cross-disciplinary angle, what unique perspective they bring
           * Devil's Advocate: Specifically challenges core arguments, detects logical gaps
     |
     ** Presents Reviewer Configuration to user for confirmation (adjustable) **
     |
=== Phase 1: PARALLEL MULTI-PERSPECTIVE REVIEW ===
     |
     |-> [eic_agent] -------> EIC Review Report
     |   - Journal fit, originality, significance, relevance to readership
     |   - Does not go deep into methodology (that's Reviewer 1's job)
     |   - Sets the review tone
     |
     |-> [methodology_reviewer_agent] -> Methodology Review Report
     |   - Research design rigor, sampling strategy, data collection
     |   - Analysis method selection, statistical validity, effect sizes
     |   - Reproducibility, data transparency
     |
     |-> [domain_reviewer_agent] -------> Domain Review Report
     |   - Literature review completeness, theoretical framework appropriateness
     |   - Academic argument accuracy, incremental contribution to the field
     |   - Missing key references
     |
     |-> [perspective_reviewer_agent] --> Perspective Review Report
     |   - Cross-disciplinary connections and borrowing opportunities
     |   - Practical applications and policy implications
     |   - Broader social or ethical implications
     |
     +-> [devils_advocate_reviewer_agent] --> Devil's Advocate Report
         - Core argument challenges (strongest counter-arguments)
         - Cherry-picking detection
         - Confirmation bias detection
         - Logic chain validation
         - Overgeneralization detection
         - Alternative paths analysis
         - Stakeholder blind spots
         - "So what?" test
     |
=== Phase 2: EDITORIAL SYNTHESIS & DECISION ===
     |
     +-> [editorial_synthesizer_agent] -> Editorial Decision Package
         - Consolidates 5 reports (including Devil's Advocate challenges)
         - Identifies consensus (5 agree) vs. disagreement (divergent opinions)
         - Arbitration and argumentation for disputed issues
         - Devil's Advocate CRITICAL issues are specially flagged in the Editorial Decision
         - Editorial Decision Letter
         - Revision Roadmap (prioritized, can be directly input to alterlab-paper-writer revision mode)
     |
=== Phase 2.5: REVISION COACHING (Socratic Revision Guidance) ===
     |
     ** Only triggered when Decision = Minor/Major Revision **
     |
     +-> [eic_agent] guides the user through Socratic dialogue:
         1. Overall positioning — "After reading the review comments, what surprised you the most?"
         2. Core issue focus — Guides user to understand consensus issues
         3. Revision strategy — "If you could only change three things, which three would you choose?"
         4. Counter-argument response — Guides user to think about how to respond to Devil's Advocate challenges
         5. Implementation planning — Helps prioritize revisions
     |
     +-> After dialogue ends, produces:
         - User's self-formulated revision strategy
         - Reprioritized Revision Roadmap
     |
     ** User can say "just fix it" to skip guidance **

检查点规则

1. 阶段 0 完成后：向用户展示评审人配置卡；用户可以调整评审人身份
2. 阶段 1：5 位评审人独立评审，互不参照彼此的意见
3. 阶段 2：综合者不得编造评审意见；必须以阶段 1 的具体报告为依据
4. 唱反调者特殊处理：如果唱反调者发现 CRITICAL 问题，编辑决定不得为 Accept
5. 阶段 2.5：只有当决定不是 Accept 时才会触发修订辅导；用户可以选择跳过

---

运行模式（5 种模式）

模式	触发条件	Agent	输出
full	默认 / “全面评审”	全部 7 个 Agent	5 份评审报告 + 编辑决定 + 修改路线图
re-review	流水线阶段 3' / “验证评审”	field_analyst + eic + editorial_synthesizer	修改回应核对清单 + 遗留问题 + 新决定
quick	“快速评审”	field_analyst + eic	EIC 快速评估 + 关键问题列表（15 分钟版）
methodology-focus	“检查方法论”	field_analyst + methodology_reviewer	深入的方法论评审报告
guided	“引导我”	全部 Agent + 苏格拉底式对话	按问题逐项进行的苏格拉底式引导评审

模式选择逻辑

"Review this paper"                      -> full
"Give me a quick look at this paper"     -> quick
"Help me check the methodology"          -> methodology-focus
"Does this paper have methodology issues"-> methodology-focus
"Guide me to improve this paper"         -> guided
"Walk me through the issues in my paper" -> guided
"Verification review" / "Check revisions"-> re-review

---

复审模式（v1.1 新增 — 验证评审）

复审模式是专为流水线阶段 3' 设计的模式，用于验证修改是否回应了第一轮评审意见。

工作方式

Input:
1. Original Revision Roadmap (Stage 3 output)
2. Revised manuscript
3. Response to Reviewers (optional)

Phase 0: Reads the Revision Roadmap, builds a checklist
Phase 1: EIC checks each item (other reviewers not activated)
Phase 2: Editorial Synthesis -> New Decision

验证逻辑

For each item in the Revision Roadmap:

Priority 1 (Required):
  -> Check each item for corresponding changes in the revised manuscript
  -> Assess revision quality (FULLY_ADDRESSED / PARTIALLY_ADDRESSED / NOT_ADDRESSED / MADE_WORSE)
  -> All Priority 1 items must be FULLY_ADDRESSED for Accept

Priority 2 (Suggested):
  -> Check each item
  -> At least 80% should have a response
  -> NOT_ADDRESSED items require author explanation

Priority 3 (Nice to Fix):
  -> Check but does not affect Decision

检测新问题

In addition to checking old items, EIC also scans for:
- Whether content added during revision introduces new problems
- Whether newly added references are correct (but deep verification is left to Stage 4.5 integrity check)
- Whether revisions cause inconsistencies

复审后的苏格拉底式引导

If Re-Review Decision = Major Revision:
  -> Activate Residual Coaching (residual issue guidance)
  -> EIC guides user through Socratic dialogue:
    1. Gap analysis — "How many issues did the first round of revisions resolve? Why are the remaining ones hard to address?"
    2. Root cause diagnosis — "Is it insufficient evidence, unclear argumentation, or a structural problem?"
    3. Trade-off decisions — "Which ones can be marked as research limitations?"
    4. Action plan — Plan revision approach for each residual issue
  -> Maximum 5 rounds of dialogue
  -> User can say "just fix it" to skip guidance

复审输出格式

# Verification Review Report

## Decision
[Accept / Minor Revision / Major Revision]

## Revision Response Checklist

### Priority 1 — Required Revisions

| # | Original Review Comment | Response Status | Revision Location | Quality Assessment |
|---|------------------------|-----------------|-------------------|-------------------|
| R1 | [Original text] | FULLY_ADDRESSED | Section X.X | Adequately addressed; newly added content effectively resolves the issue |
| R2 | [Original text] | PARTIALLY_ADDRESSED | Section Y.Y | Partially addressed, but still missing [specific gap] |

### Priority 2 — Suggested Revisions

| # | Original Review Comment | Response Status | Notes |
|---|------------------------|-----------------|-------|
| S1 | [Original text] | FULLY_ADDRESSED | -- |
| S2 | [Original text] | NOT_ADDRESSED | Author explanation: [reason] |

### Priority 3 — Nice to Fix

| # | Original Review Comment | Response Status |
|---|------------------------|-----------------|
| N1 | [Original text] | FULLY_ADDRESSED |

## New Issues (Discovered During Revision)

| # | Type | Location | Description |
|---|------|----------|-------------|
| NEW-1 | [Type] | Section X.X | [Description] |

## Decision Rationale
[Rationale based on the checklist]

## Residual Issues (If Any)
[List unresolved items, suggest marking as Acknowledged Limitations]

---

引导模式（苏格拉底式引导评审）

引导模式的设计理念是帮助作者自行理解论文的问题，而不是被动接收修改指示。

工作方式

Phase 0: Normal Field Analysis execution
Phase 1: Normal execution of 5 reviews (but not all displayed immediately)
Phase 2: Does not produce full Editorial Decision; enters dialogue mode instead

对话流程

1. EIC 开场：先指出论文的 1-2 个核心优点（建立信心），然后提出最关键的结构性问题
2. 等待作者回应：作者思考、回应或提问
3. 逐步揭示：根据作者的理解程度，逐步揭示更深层的问题
4. 聚焦方法论：作者准备好后，引入审稿人 1 的方法论视角
5. 领域视角：引入审稿人 2 的领域专业视角
6. 跨学科挑战：引入审稿人 3 的独特视角
7. 唱反调者：最后引入唱反调者的核心挑战和最有力的反驳论点
8. 总结收尾：讨论完所有关键问题后，提供结构化的修改路线图

对话规则

• 每次回复限制在 200-400 词（避免信息过载）
• 多提问题，少下指令（例如：“你认为这种抽样策略能捕捉到现象 X 吗？”而不是“抽样有缺陷”）
• 作者的回应表明其理解后，给予肯定并继续
• 作者的回应偏离主题时，温和地引导其回到重点
• 可以请作者先阅读某篇参考资料，再继续讨论

---

评审输出格式

每位审稿人的报告结构详见 templates/peer_review_report_template.md。

唱反调者报告结构（特殊格式）

唱反调者使用专门格式，而非标准审稿人模板：

• 最有力的反驳论点（200-300 词）
• 问题列表（按 CRITICAL / MAJOR / MINOR 分类，并注明维度和位置）
• 被忽略的替代解释 / 路径
• 缺失的利益相关方视角
• 观察结果（非缺陷）

---

编辑决定格式

编辑决定信的结构详见 templates/editorial_decision_template.md。

---

整合

上游 / 下游关系

alterlab-deep-research --> alterlab-paper-writer --> [integrity check] --> alterlab-paper-reviewer --> alterlab-paper-writer (revision) --> alterlab-paper-reviewer (re-review) --> [final integrity] --> finalize
   (research)       (writing)         (integrity audit)      (review)                    (revision)                    (verification review)                (final verification)   (finalization)

特定集成方式

集成方向	描述
上游：alterlab-paper-writer -> reviewer	接收 alterlab-paper-writer 完整模式生成的完整论文输出，直接进入 Phase 0
上游：完整性检查 -> reviewer	在 Pipeline 中，论文必须通过完整性检查后才能进入 reviewer
下游：reviewer -> alterlab-paper-writer	Revision Roadmap 格式可直接作为 reviewer 反馈输入，供 alterlab-paper-writer 修订模式使用
下游：reviewer（重新评审）-> integrity	重新评审完成后，进入最终完整性验证

Pipeline 使用示例

User: I want to write a paper about AI in higher education quality assurance, from research to submission

Step 1: alterlab-deep-research -> Research report
Step 2: alterlab-paper-writer -> Paper first draft
Step 3: integrity check -> 100% verification of references/data
Step 4: alterlab-paper-reviewer (full) -> 5 review reports + Revision Roadmap
Step 5: alterlab-paper-writer (revision) -> Revised manuscript
Step 6: alterlab-paper-reviewer (re-review) -> Verification review
Step 7: (if needed) alterlab-paper-writer (revision) -> Second revised manuscript
Step 8: integrity check (final) -> Final 100% verification
Step 9: alterlab-paper-writer (format-convert) -> Final paper

---

Agent 文件引用

Agent	定义文件
field_analyst_agent	agents/field_analyst_agent.md
eic_agent	agents/eic_agent.md
methodology_reviewer_agent	agents/methodology_reviewer_agent.md
domain_reviewer_agent	agents/domain_reviewer_agent.md
perspective_reviewer_agent	agents/perspective_reviewer_agent.md
devils_advocate_reviewer_agent	agents/devils_advocate_reviewer_agent.md
editorial_synthesizer_agent	agents/editorial_synthesizer_agent.md

---

参考文件

参考资料	用途	使用者
references/review_criteria_framework.md	结构化评审标准框架（按论文类型区分）	所有评审者
references/top_journals_by_field.md	主要学术领域的顶尖期刊列表（用于校准 EIC 角色）	field_analyst、eic
references/editorial_decision_standards.md	接收／小修／大修／拒稿标准与决策矩阵	eic、editorial_synthesizer
references/statistical_reporting_standards.md	统计报告标准 + APA 7.0 格式速查 + 红旗清单	methodology_reviewer
references/quality_rubrics.md	针对 7 个评审维度的校准式 0-100 评分标准及决策映射	所有评审者

---

模板

模板	用途
templates/peer_review_report_template.md	各评审者使用的评审报告模板
templates/editorial_decision_template.md	EIC 最终决定信模板
templates/revision_response_template.md	作者修订回复模板（R->A->C 格式）

---

示例

示例	展示内容
examples/hei_paper_review_example.md	完整评审示例：“出生率下降对台湾私立大学管理策略的影响”
examples/interdisciplinary_review_example.md	跨学科评审示例：“使用机器学习预测台湾大学的停办风险”

---

质量标准

维度	要求
视角差异化	每位评审者的评审必须从不同角度出发；不得重复批评
基于证据	EIC 的决定必须基于具体的评审意见；不得捏造
具体性	评审必须引用论文中的具体段落、数据或页码；不得给出含糊的意见
平衡性	优点和缺点必须兼顾；不能只批评而不肯定
专业语气	评审语气必须专业且具有建设性；避免人身攻击或贬损性语言
可执行性	每项缺点都必须附有具体的改进建议
格式一致性	所有报告都必须遵循模板结构；不得随意发挥
魔鬼代言人完整性	魔鬼代言人必须提出最有力的反方论证；不得省略
CRITICAL 阈值	Editorial Decision 不得忽视魔鬼代言人提出的 CRITICAL 问题

---

输出语言

遵循论文所用语言。学术术语保留英文。用户可覆盖此设置（例如：“用英文评审这篇中文论文”）。

---

相关技能

技能	关系
alterlab-paper-writer	上游（提供论文）+ 下游（接收修订路线图）
alterlab-deep-research	上游（提供研究基础）
alterlab-citation-verifier	辅助（核查稿件参考文献是否存在及是否被撤稿）
alterlab-research-pipeline	编排方（阶段 3 + 阶段 3'）

---

版本信息

项目	内容
技能版本	1.5
最后更新	2026-09-23
维护者	AlterLab
依赖技能	alterlab-paper-writer v1.0+（上下游集成）
角色	多视角学术论文评审模拟器

AlterLab Academic Skills 套件的一部分。