paper-revision
lingzhi227/agent-research-skills
Systematically revise papers by mapping reviewer feedback to sections, executing targeted edits, and verifying improvements.
What is paper-revision?
This skill helps researchers revise academic papers in response to peer review. It parses reviewer comments, prioritizes concerns, plans and executes targeted edits (including new experiments if needed), and verifies that revisions address all feedback. Use it after receiving reviewer comments on a submitted paper.
- Parse and classify reviewer concerns (major/minor revisions, questions, suggestions)
- Map feedback to specific paper sections and prioritize by impact
- Create a structured revision plan linking concerns to required actions
- Execute targeted edits while preserving paper structure and marking changes
- Run new experiments or generate figures/tables as needed via related skills
- Verify improvements by re-running self-review and checking all concerns are addressed
How to install paper-revision
npx skills add https://github.com/lingzhi227/agent-research-skills --skill paper-revision- Current paper draft (main.tex or paper directory)
- Reviewer comments or feedback document
- Access to related skills: experiment-code, figure-generation, table-generation, self-review, paper-compilation
How to use paper-revision
- 1.Provide reviewer comments and current paper draft as input
- 2.Review the parsed concerns and prioritization to ensure accuracy
- 3.Examine the revision plan mapping concerns to affected sections and required actions
- 4.Execute each planned revision, using related skills for experiments or figures as needed
- 5.Mark revised text clearly (e.g., with \textcolor{blue}{...}) for reviewer visibility
- 6.Run self-review to verify improved scores and check all concerns are addressed
- 7.Generate and review the revision summary before submission
Use cases
- Revise a paper after receiving peer review feedback from a conference or journal
- Address major revision requests by planning and executing structural changes and new experiments
- Handle mixed feedback by prioritizing major concerns first and systematically working through minor revisions
- Generate a revision summary that cross-references each change to the reviewer concern it addresses
- Verify that revisions improve paper quality scores without introducing new issues
- Researchers and academics preparing revised manuscripts
- Authors responding to peer review at conferences or journals
- Teams managing multi-reviewer feedback across multiple papers
paper-revision FAQ
Extract the specific concern as best as possible, classify it, and map it to the most relevant section. If ambiguous, prioritize clarification edits and note the ambiguity in the revision summary.
No. Only run new experiments when the revision plan explicitly requires them (e.g., 'Additional experiment' action). Many revisions are clarifications, rewrites, or citations that don't need new data.
Parse each reviewer's concerns separately, then consolidate by section. If concerns conflict, address both perspectives or note the conflict in the revision summary for the rebuttal.
No. The skill enforces addressing EVERY reviewer concern. If a concern seems invalid, document it and address it in the rebuttal separately.
Keep a copy before revising. Use it to verify that revisions don't introduce new issues and to generate a clear diff summary for reviewers.
Full instructions (SKILL.md)
Source of truth, from lingzhi227/agent-research-skills.
name: paper-revision description: Revise papers based on reviewer feedback. Map reviewer concerns to specific sections, apply targeted edits, run additional experiments if needed, and verify improvements. Use after receiving peer review with revision requests. argument-hint: [reviews-or-draft]
Paper Revision
Systematically revise papers based on reviewer feedback.
Input
$0— Reviewer comments/feedback$1— Current paper draft (main.tex or paper directory)
References
- Revision workflow and prompts:
~/.claude/skills/paper-revision/references/revision-prompts.md
Workflow
Step 1: Parse and Prioritize Concerns
For each reviewer comment:
- Extract the specific concern
- Classify: major revision, minor revision, question, suggestion
- Map to affected paper section(s)
- Prioritize: address major concerns first
Step 2: Plan Revisions
Create a revision plan:
Concern → Affected Section → Required Action → New Content/Experiment
Categories of actions:
- Clarification: Rewrite text for clarity
- Additional experiment: Run new experiment, add results
- New analysis: Add ablation, statistical test, or comparison
- Structural change: Move, merge, or split sections
- Citation: Add missing references
Step 3: Execute Revisions
For each planned revision:
- Read the current section
- Apply targeted edits (preserve surrounding structure)
- If new experiments needed: use experiment-code skill
- If new figures/tables needed: use figure-generation / table-generation skills
- Mark changes (use
\textcolor{blue}{...}for revised text)
Step 4: Verify Improvements
- Re-run self-review skill to check if scores improved
- Verify all reviewer concerns are addressed
- Check that revisions don't introduce new issues
- Ensure page count still fits venue requirements
Step 5: Write Revision Summary
Generate a diff summary:
- List all changes made with section references
- Note any new experiments, figures, or tables added
- Cross-reference each change to the reviewer concern it addresses
Rules
- Address EVERY reviewer concern — do not skip any
- Preserve paper structure unless structural change is explicitly needed
- New results must come from actual experiments, not hallucinated
- Mark all revised text clearly for the reviewers
- Keep a copy of the previous version before revising
- Compare new scores vs previous scores after revision
Related Skills
- Upstream: self-review
- Downstream: paper-compilation
- See also: rebuttal-writing, paper-writing-section
Related skills
More from lingzhi227/agent-research-skills and the wider catalog.
paper-to-code
Convert ML research papers into complete, runnable code repositories via structured planning, analysis, and generation.
paper-writing-section
Write publication-quality sections for academic papers with section-specific guidance and two-pass refinement.
rebuttal-writing
Write evidence-based rebuttals to peer review comments with structured point-by-point responses.
related-work-writing
Write publication-quality Related Work sections that organize literature by theme and position your research.
research-planning
Design structured research plans and paper architectures from a topic or idea.
self-review
Automatically review academic papers using NeurIPS form with three reviewer personas and ensemble scoring.