agentleFS
Felpix-StudiosFelpix-Studios/social-science-research/skills/revise-paper/SKILL.md

revise-paper

Skill8 starsChanged 4 months agoMITView on GitHub

Apply revisions from a review report to a manuscript. Reads the review, presents prioritized plan, applies fixes section-by-section with user approval. Use when systematically addressing review issues, not drafting a new paper or running a fresh review.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. CISLab-HKUST/revise-paper★ 55same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository12 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If $ARGUMENTS is provided, use that path.

    line 18
  2. Distinctive:

    If multiple reports found, use AskUserQuestion:

    line 23
  3. Very specific:

    Read the review report and extract the manuscript path from its File: field.

    line 31
  4. Distinctive:

    If the file doesn't exist at that path, glob for:

    line 33
  5. Distinctive:

    If multiple manuscripts found, use AskUserQuestion to let the user pick.

    line 37
  6. Distinctive:

    Read the full manuscript.

    line 39
  7. Distinctive:

    Prose Track rules apply: If the user deferred any items, skip them and note them in the summary.

    line 95
  8. Very specific:

    After all approved revisions are applied, dispatch the proofreader agent via Task to catch any issues introduced by the edits:

    line 100
  9. Distinctive:

    Prose Track only: Never apply edits without user approval.

    line 148
  10. Distinctive:

    Work from the review report: Don't invent new issues — address what the review found.

    line 149
  11. Distinctive:

    Preserve author voice: When rewriting prose, maintain the manuscript's existing style and tone.

    line 150
  12. Distinctive:

    Track everything: Every revision applied or deferred must appear in the summary.

    line 151

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Revise Paper from Review Report

Systematically apply revisions from a review report to a manuscript. Follows Prose Track rules — never auto-edits without approval.

Input: $ARGUMENTS — path to a review report, or leave blank to auto-detect.

Step 1: Locate the Review Report

If $ARGUMENTS is provided, use that path. Otherwise glob for:

  • quality_reports/paper_review_*.md
  • quality_reports/*_substance_review.md
  • quality_reports/*_proofread.md

If multiple reports found, use AskUserQuestion:

  • header: "Review Report"
  • question: "Multiple review reports found. Which should I use for revisions?"
  • options: list found reports (label: filename, description: date and type). User can select one.

Step 2: Locate the Manuscript

Read the review report and extract the manuscript path from its File: field.

If the file doesn't exist at that path, glob for:

  • manuscripts/**/*.tex
  • manuscripts/**/*.qmd

If multiple manuscripts found, use AskUserQuestion to let the user pick.

Read the full manuscript.

Step 3: Parse the Review Report

Extract all actionable items and classify them:

PrioritySource SectionAction
CriticalMajor Concerns with "CRITICAL" or "BLOCKING"Must fix before any submission
MajorMajor Concerns, Referee ObjectionsShould fix — a reviewer would flag these
MinorMinor Concerns, Specific CommentsPolish — fix if time allows

For each item, extract:

  • The issue description
  • The suggested fix (if provided)
  • The location in the manuscript (section, line, table)

Step 4: Present Revision Plan (WAIT FOR APPROVAL)

Present the revision plan grouped by priority:

## Revision Plan

### Critical (N items) — must fix
1. [MC1: Title] — Section X — [brief description of fix]
2. ...

### Major (N items) — should fix
1. [MC2: Title] — Section Y — [brief description of fix]
2. ...

### Minor (N items) — polish
1. [mc1: Title] — Section Z — [brief description of fix]
2. ...

Total: N revisions across M sections.

Do NOT start editing until the user approves the plan or selects which items to address.

Step 5: Apply Approved Revisions

Work through approved items in this order:

  1. Critical items first
  2. Major items second
  3. Minor items last

For each revision:

  1. Show the current text and the proposed replacement
  2. Apply the edit using the Edit tool
  3. Move to the next item

Prose Track rules apply: If the user deferred any items, skip them and note them in the summary.

Step 6: Post-Revision Review

After all approved revisions are applied, dispatch the proofreader agent via Task to catch any issues introduced by the edits:

Task prompt: "Review the manuscript at [path] for grammar, typos, layout issues,
and consistency. Focus on sections that were recently revised: [list changed sections].
Return your report — the calling skill handles saving."

Present any new issues found by the proofreader.

Step 7: Save Revision Summary

Save to quality_reports/revision_summary_[paper-name].md:

# Revision Summary: [Paper Name]

**Date:** [YYYY-MM-DD]
**Review report used:** [path]
**Manuscript:** [path]

## Revisions Applied

| # | Priority | Issue | Section | Status |
|---|----------|-------|---------|--------|
| 1 | Critical | [title] | Section X | APPLIED |
| 2 | Major | [title] | Section Y | APPLIED |
| 3 | Minor | [title] | Section Z | DEFERRED — user chose to skip |

## Issues Deferred

- [Item]: [reason deferred]

## Post-Revision Proofreader Findings

- [Any new issues introduced by edits]

## Recommended Next Steps

- **`/quality-gate`** — verify claim consistency after edits
- **`/validate-bib`** — check citations if references were added or changed
- **`/review-paper`** — run a fresh review to confirm issues are resolved

Key Rules

  • Prose Track only. Never apply edits without user approval.
  • Work from the review report. Don't invent new issues — address what the review found.
  • Preserve author voice. When rewriting prose, maintain the manuscript's existing style and tone.
  • Track everything. Every revision applied or deferred must appear in the summary.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.