nature-response

Draft and audit mutually isolated point-by-point reviewer responses and revision packages.

Updated Aug 26, 2026
One-click install
npx skills add https://github.com/Yjj0333/skills-manager-backup --skill nature-response-yjj0333
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nature-response
Source: https://github.com/Yjj0333/skills-manager-backup/tree/main/nature-response
Command: npx skills add https://github.com/Yjj0333/skills-manager-backup --skill nature-response-yjj0333

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Responding to journal peer review requires coordinating editor instructions, multiple mutually blind reviewers, manuscript edits, and cover letters without leaking cross-reviewer information or fabricating evidence. This Skill structures that entire revision-correspondence workflow so nothing is missed, misattributed, or overstated. ## Core Features & Use Cases - Reviewer-Isolated Response Drafting: Splits editor and reviewer comments into stable IDs (E.1, R1.1, R2.3) and produces a standalone point-by-point response letter per reviewer that never reveals another reviewer's comments, numbering, or recommendations. - Evidence-Calibrated Tracking: Classifies each comment by action type, work status, and blocking state, refusing to mark anything verified without an inspectable artifact and flagging missing author input instead of inventing experiments, statistics, or line numbers. - LaTeX Revision Package Auditing: Ships templates for cover letters, response letters, and red-marked manuscripts, plus a consistency script that checks verbatim quote drift, comment-response count mismatches, and clean-versus-marked manuscript divergence. - Use Case: Paste an editor decision email with two reviewer reports and your Chinese revision notes; receive a decision-type-gated strategy summary, an internal master tracker, separate English response letters per reviewer, a cover letter draft, and a list of facts you still must supply. ## Quick Start Paste your editor decision letter and reviewer comments and ask the assistant to generate a point-by-point response framework with per-reviewer isolated response letters.

Frequently Asked Questions about nature-response

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a point-by-point response to reviewers?▼

Provide the editor decision letter, reviewer comments, and your completed or planned changes. The workflow assigns stable IDs to each comment, classifies the response action, and drafts a formal standalone reply per reviewer with manuscript locations or visible placeholders.

How do I handle conflicting reviewer comments in a rebuttal?▼

Reconcile the conflict only in an internal editor-facing strategy, prioritizing explicit editor instructions and the study design. Each reviewer then receives an independent, scientifically consistent explanation that never mentions the other reviewer's request.

Should reviewers see each other's comments in my response letter?▼

No. Reviewers are treated as mutually blind by default, so each reviewer-facing file contains only that reviewer's comments with neutral local numbering. Repeated concerns are answered fully in each file rather than cross-referenced.

What if a reviewer asks for an experiment I cannot do?▼

Acknowledge the scientific value, explain the study-design or scope boundary, offer alternative evidence if available, and soften the claim or add a limitation. Avoid citing time, budget, or convenience as the primary reason.

How do I check a LaTeX revision package for consistency?▼

Run scripts/check_package_consistency.py with the manuscript, response letter, and clean and marked files. It verifies that quoted excerpts appear verbatim in the manuscript, comment and response counts match, and clean and marked texts agree after markup removal.

What are the limitations of automated rebuttal drafting?▼

The workflow never invents experiments, statistics, citations, or line numbers; missing facts are flagged as author input needed and the package cannot be labeled ready to submit. Compiled PDF page numbers, colors, and references still require manual final verification.