科研 Skill
审稿人作者迭代改稿
reviewer-author-loop
Runs an iterative manuscript-improvement loop where a reviewer agent critiques, an author agent revises, and a verifier checks each revision. Use for simulated peer review or rebuttal prep.
SKILL.md
Reviewer-Author Loop
Purpose
Use this skill to run a closed-loop manuscript improvement process:
- A Reviewer Agent evaluates the manuscript as a skeptical but constructive peer reviewer.
- An Author Agent revises the manuscript and response plan to address the review.
- A Verifier Agent checks that the revision actually resolves the comments and does not introduce new problems.
- The loop repeats until the reviewer recommends acceptance or the author/verifier reaches a pause condition requiring human input.
This skill is distinct from one-pass peer review. Its value is the revision loop, traceable comment resolution, and explicit stop/pause decision.
Confidentiality
If private manuscripts, reviewer reports, review archives, unpublished data, grant proposals, or confidential comments are provided, use them only for the current task. Do not copy private text into reusable skill files, public outputs, examples, or repositories. When extracting lessons from private materials, convert them into abstract process rules with no titles, author names, manuscript identifiers, wording, or identifiable facts.
When To Use
Use this skill when the user asks to:
- simulate peer review and then revise the manuscript
- alternate between reviewer and author roles
- improve a paper until it is acceptable
- prepare a response to reviewers or rebuttal
- decide whether reviewer comments require new data, experiments, modeling, theory, or human judgment
- re-review a revised manuscript after changes
- audit whether author revisions truly addressed reviewer concerns
If the manuscript is in a specific technical field and a domain skill is available, use this skill as the process scaffold and the domain skill as the judgment layer.
Inputs To Identify
Before starting, identify what is available:
- manuscript source: DOCX, PDF, LaTeX, Markdown, Overleaf folder, or plain text
- target journal, conference, thesis committee, or funding program
- current stage: pre-submission, revision, resubmission, internal review, final polish
- existing reviewer comments or simulated review needed
- editable source file and build/render method
- user constraints: maximum loop count, acceptable aggressiveness, whether to edit files directly
Proceed with reasonable assumptions when the missing information is not blocking. Pause only when the missing information changes the revision path.
Loop Protocol
1. Reviewer Agent
Read the current manuscript and produce a decision-oriented review:
- recommendation: accept, minor revision, major revision, reject, or human input required
- core contribution and journal fit
- major comments that affect validity, novelty, methods, interpretation, or evidence
- minor comments that affect clarity, formatting, citations, or presentation
- required evidence or analysis
- acceptance criteria: what must change before the reviewer would recommend acceptance
Use references/reviewer-rubric.md for detailed review dimensions.
2. Author Agent
Convert reviewer comments into revisions:
- map each reviewer concern to a concrete action
- revise the manuscript, figures, tables, equations, references, and captions when possible
- add caveats or limitations when a claim cannot be fully supported
- avoid performative agreement: do not weaken or remove a claim if stronger evidence can support it
- preserve a response log with comment, action, manuscript location, and residual risk
Use references/author-revision-protocol.md for revision strategy and response-log structure.
3. Verifier Agent
Check the revision before returning to review:
- build/render the manuscript when possible
- verify citations, references, figures, tables, equation numbering, and cross-references
- check that every reviewer concern is addressed or explicitly marked unresolved
- check that new text does not overclaim, contradict earlier sections, or introduce unsupported statements
- check that figures/captions remain readable and aligned with the discussion
If editing files, do not claim completion until verification has been run or the inability to run it is reported.
4. Re-Reviewer Agent
Re-review the revised manuscript against the previous acceptance criteria:
- accepted or acceptable after editorial polish
- minor revision remains
- major revision remains
- human input required
If more revisions are possible with available information, loop back to the Author Agent. If not, pause.
Stop And Pause Rules
Stop when the reviewer recommendation is:
- accept
- accept after minor editorial revision
- ready for submission with only user-preference edits remaining
Pause and ask the user when progress requires:
- new experiments, simulations, datasets, images, measurements, or raw files
- an author decision on scope, claims, novelty, or target journal
- domain knowledge not present in the manuscript or available sources
- a new theory/model that cannot be responsibly invented from current evidence
- confidential permissions or coauthor judgment
- choosing among mutually exclusive revision strategies
Use references/pause-criteria.md to classify pause conditions and phrase the user request.
Output Format
During the loop, keep outputs concise and actionable:
- Reviewer Decision
- Required Revisions
- Author Actions Taken
- Verification
- Remaining Risks
- Next Loop Decision
When the loop stops, provide:
- final recommendation
- files changed
- verification performed
- unresolved risks or optional polish items
Coordination With Other Skills
In the manuscript workflow that motivated this skill, the reviewer-author loop acted as the process scaffold while other skills supplied specialist judgment or file operations.
When acting as the Reviewer Agent, use and cite companion skills as applicable:
academic-paper-reviewer: independent peer-review stance, editorial recommendation, major/minor comments, re-review after revision.mechanical-engineering-research: domain-specific technical review for thermal-fluid physics, boiling/heat-transfer models, experimental design, uncertainty, equations, figures, and interpretation.deep-research: targeted literature checks when a review comment depends on current or unfamiliar literature.
When acting as the Author Agent, use and cite companion skills as applicable:
academic-paper: manuscript restructuring, abstract/introduction/discussion revision, response-to-reviewer style writing, and paper-level coherence.mechanical-engineering-research: technical rewriting, mechanistic framing, model explanation, analysis design, and engineering interpretation.documents:documents: DOCX editing, rendering, visual QA, comments, and tracked document artifact work.spreadsheets:Spreadsheets: spreadsheet inspection, data extraction, recalculation, and analysis tables.zotero:Zotero: citation lookup, BibTeX export, reference cleanup, and citation-key insertion when Zotero or BibTeX is used.presentations:Presentations: figure or slide-deck integration when manuscript figures are being adapted from presentation material.
For one-pass critique only, a peer-review skill is sufficient. Use this skill when the user wants iterative review, revision, verification, and re-review. In outputs, name the companion skills used for each loop stage so the workflow remains auditable.
Minimal Flow
flowchart TD
A["Manuscript + target journal"] --> B["Reviewer critiques"]
B --> C["Decision + required revisions"]
C --> D{"Can revise with available information?"}
D -- "Yes" --> E["Author revises manuscript"]
E --> F["Verifier checks build, citations, figures, claims"]
F --> G["Reviewer re-reviews"]
G --> H{"Acceptable?"}
H -- "Yes" --> I["Stop: ready or near-ready"]
H -- "No, revision possible" --> C
D -- "No" --> J["Pause for human input"]
J --> K["Human provides data, decisions, or theory"]
K --> E
示例提示词
帮我模拟审稿人和作者的多轮修改,打磨这篇稿件直到可接收。
所属插件
安装该插件会一并获得同领域的 17 个 Skill / MCP。
同插件的其他科研 Skill
- 控制系统分析与PID设计Models, analyzes, and designs feedback control systems with python-control and scipy.signal: transfer functions, Bode plots, root locus, and PID tuning. Use for SISO control design.
- FEniCSx有限元分析Runs FEA with FEniCSx or scikit-fem: mesh generation, boundary conditions, linear elasticity, heat conduction, and result visualization. Use for custom open-source finite-element workflows.
- 测量不确定度分析Builds a GUM-style measurement uncertainty budget from Type A/B evaluation, RSS combination, and error propagation to a derived result. Use when a reported value needs a defensible plus-or-minus.
- 热流体系统研究方法指导Applies source-aware mechanical-engineering judgment to research tasks spanning thermal-fluids, turbomachinery, experiments, and writing. Use for broad mechanical-engineering research support.
- 可靠性工程分析Covers reliability engineering methods: Weibull analysis, fault tree analysis, FMEA, reliability block diagrams, and accelerated life testing. Use for failure-rate or reliability-prediction tasks.
- 工程实验数据分析设计Designs and analyzes engineering experiments or simulations with baseline cases, hypothesis-driven DOE, and uncertainty quantification. Use for CFD, lab data, or parameter-study analysis.
- 科研导师式反馈Gives constructive, evidence-based mentoring feedback on student manuscripts, proposals, code, and figures without impersonating an advisor. Use for advisor-style review of student work.
- 研究计划书撰写Develops research proposals with solicitation alignment, reviewer-focused logic, evidence-bounded preliminary results, and milestones. Use for federal, foundation, or internal research proposals.
- 科研示意图制作Creates publication-quality, editable scientific and engineering schematics with explicit physical meaning and clear labels. Use for experimental-facility diagrams, mechanisms, or graphical abstracts.
- 科研汇报幻灯片设计Designs clean, graphics-first technical presentations and posters with logical progression and visual hierarchy. Use for research talks, conference slides, or defenses.
- 科研写作与文献综述Writes and revises research narratives, literature reviews, manuscript sections, and figure discussions with proper citations. Use for drafting or revising technical manuscript text.
- ROS2机器人开发Covers ROS2 robotics development with rclpy: nodes, topics, services, actions, tf2 transforms, and Nav2 navigation. Use when building or debugging a ROS2 node or navigation stack.
- 数字信号处理Covers digital signal processing with scipy.signal: filter design, spectral analysis, peak detection, and PSD estimation. Use for vibration, biomedical, or time-series signal tasks.
- SimPy离散事件仿真Builds discrete-event simulations with SimPy: M/M/c queues, manufacturing lines, resource contention, and confidence intervals via replications. Use for queueing or throughput analysis.
- 热流体系统分析指导Analyzes thermal-fluid systems, experiments, CFD, and correlations with explicit assumptions, validity limits, and uncertainty. Use for heat transfer or thermal-management analysis.
- 机械振动与故障诊断Analyzes mechanical vibration: SDOF/MDOF natural frequencies, resonance checks, and FFT-based diagnosis of rotating-machinery faults such as imbalance or bearing defects. Use for vibration diagnosis.
常见问题
审稿人作者迭代改稿能做什么?
Runs an iterative manuscript-improvement loop where a reviewer agent critiques, an author agent revises, and a verifier checks each revision. Use for simulated peer review or rebuttal prep。
如何安装审稿人作者迭代改稿?
在科搭 Codar 左侧点击「插件」,在「组件」页签搜索「审稿人作者迭代改稿」并安装;在 Claude Code 或 Codex 中可以发送本页的安装提示词,或下载 Skill 包后解压到 ~/.claude/skills 或 ~/.codex/skills。
审稿人作者迭代改稿需要哪些权限?
该科研 Skill没有声明额外权限;执行具体操作时仍由 Agent 按自身规则请求确认。
审稿人作者迭代改稿经过官方审核吗?
是。SpeedSkills 收录的每个版本都会记录内容摘要与权限声明,当前版本已通过安全审核。