light-review-rebuttal

Builds evidence-bound peer-review revision and author-response packages from reviewer comments.

572|72|Updated Jun 7, 2026
One-click install
npx skills add https://github.com/Light0305/Light-skills --skill light-review-rebuttal
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: light-review-rebuttal
Source: https://github.com/Light0305/Light-skills/tree/main/skills/light-review-rebuttal
Command: npx skills add https://github.com/Light0305/Light-skills --skill light-review-rebuttal

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Responding to peer review often means juggling verbatim reviewer comments, revision tracking, venue rules, and response letters by hand, which leads to missed comments, overstated claims of completed work, and responses that violate venue limits. This Skill turns reviewer feedback into an auditable, machine-checked revision and rebuttal package.

Core Features & Use Cases

  • Lossless review atomization: Preserves reviewer, editor, and meta-review text verbatim with SHA-256 hashed source spans, then splits feedback into typed issues (request, claim, question, misunderstanding, editorial) with root causes.
  • Truthful commitment tracking: Enforces PLANNED vs DONE states so planned experiments are never written as completed, and DONE items require real change locators plus run provenance hashes.
  • Venue-rule-aware budgeting and gating: Checks response length only against the selected venue's current sourced rules (never borrowed presets), gates new-experiment requests on verified policy and user authorization, and routes rejection-driving issues back to novelty, experiment, or writing stages only after user approval.
  • Use Case: After receiving a major-revision decision with three reviewer reports, use this Skill to capture the reviews (including public OpenReview forums), atomize every comment, bind each to manuscript claims and evidence, draft a point-by-point response letter, and verify the package before resubmission.

Quick Start

Use the light-review-rebuttal skill to turn these reviewer comments and my decision letter into a tracked issue matrix and draft response letter for my selected venue.

Frequently Asked Questions about light-review-rebuttal

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

FAQPage Schema
How do I write a rebuttal or response letter to peer reviewers?

Capture the reviewer comments verbatim, atomize them into individual issues with root causes, bind each to a strategy and owner, then draft point-by-point responses. This Skill automates that structure and validates that no comment is dropped and no planned work is claimed as done.

How do I fetch reviews from OpenReview programmatically?

Use the fetch_openreview.py script with a forum ID to capture public OpenReview API v2 notes, or pass a commit-pinned PeerRead snapshot URL when the live API is unavailable. Network or auth failures are recorded as UNAVAILABLE rather than treated as absent reviews.

Can I add new experiments during a revision if the venue rules are unclear?

No. The experiment request gate blocks RUN commitments unless the venue's new-results policy is VERIFIED from an official source with a check date, feasibility is adequate, and the user authorizes it. Unknown or unverified rules keep the request non-passing.

Does this skill check response letter word or page limits?

Yes, but only against the selected venue's current sourced rules. The rebuttal_budget.py script counts words and characters and returns PASS or FAIL for available rules; unknown, stale, or page-only limits stay explicitly UNKNOWN instead of borrowing another venue's preset.

What are the limitations of automated review-response checking?

The scripts verify structure: source-span hashes, issue coverage, DONE locators, run provenance, and policy authorization. They do not judge whether a scientific response is persuasive or adequate, so a human re-review of the revised artifact is still required.