Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› epsl-revision-and-rebuttal
epsl-revision-and-rebuttal
GitHub针对EPSL期刊修改意见的回复策略。指导如何平衡字数限制与审稿人要求，将新增数据移至补充材料，逐点回应审稿人，协调跨学科冲突，并确保数据一致性，以撰写高质量的修订说明和回复信。
Trigger Scenarios
收到EPSL主要或次要修改决定并准备回复时
不同子领域的审稿人提出相互矛盾的意见时
审稿人要求的补充内容可能导致正文超出字数限制时
起草给编辑的变化摘要说明时
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill epsl-revision-and-rebuttal -g -y
SKILL.md
Frontmatter
{
"name": "epsl-revision-and-rebuttal",
"description": "Use when responding to an Earth and Planetary Science Letters (EPSL) revision decision. EPSL decisions are made by a subject editor on the advice of referees who often span two sub-fields, so the response must satisfy each specialist with evidence while keeping the Letter concise. It structures the revision and response letter; it does not fabricate new results."
}
Revision & Rebuttal (epsl-revision-and-rebuttal)
An EPSL revision has a specific tension: referees ask for additions — more standards data, another sensitivity case, an alternative-hypothesis discussion — while the letters format refuses to grow. The craft is to satisfy every technical demand in the supplement and deposit while the main text stays near the cap, and to remember that the subject editor, not the referees, makes the call.
When to trigger
- A major/minor revision decision arrived and you are planning the response
- Referees from different sub-fields pull in different directions
- A requested addition would blow the word cap
- Drafting the summary-of-changes note to the editor
Strategy
- Mine the editor's letter first. The handling editor is a specialist and usually signals which referee points are decisive; resolve those with new evidence, not prose.
- Answer every point, point-by-point. Quote each comment, respond beneath it, and cite the exact location of the change (section, figure, table, supplement item, repository DOI). Unanswered comments read as evasion.
- Route additions to the supplement. New robustness tests, standards tables, and sensitivity cases go to the supplementary material and repositories; the main text absorbs only conclusions. State this routing openly — EPSL editors know the format's constraint and respect it.
- Rebut with data, concede with grace. A disagreement backed by a new calculation or a cited constraint is persuasive; a disagreement backed by adjectives is not. Where the referee is right, fix it and say so plainly.
- Reconcile cross-field conflicts explicitly. When the geochemist wants more caveats and the geodynamicist wants a bolder claim, choose a principled position and explain the trade-off to the editor — the adjudicator — rather than splitting the difference incoherently.
- Keep numbers synchronized. Any re-reduction or added analysis updates the manuscript, the supplement, and the deposited tables together; a mismatch discovered in round two is fatal to trust.
Triaging EPSL referee asks
| Referee ask | Default mode | What to show |
|---|---|---|
| "Report full standards/blank data" | concede → supplement | the per-session traceability table |
| "The feature may be a resolution artifact" | test, then rebut or scope | the recovery test at that node |
| "Consider alternative hypothesis H" | engage seriously | a discriminating calculation, or an honest scope limit |
| "Harmonize with published ages" | concede | recalculated comparison, constants stated |
| "Expand the regional context" | resist politely | letters format; one sentence + citations |
| "Tone down the global claim" | negotiate | keep the portable claim, bound its domain |
Response-letter format
> [Quoted referee comment]
Response: [what changed / why we respectfully disagree, with evidence].
Change: [Section X / Fig. Y / Table S3 / EarthChem DOI updated].
Open with a one-paragraph summary of the major changes for the editor; group by referee; keep the tone even toward a hostile review — the editor reads your temperament as a proxy for your rigor.
Worked micro-example (illustrative — answering a resolution objection)
Referee 2 (seismologist, illustrative): "The low-velocity conduit lies near the edge of ray coverage; I suspect vertical smearing."
> The low-velocity conduit lies near the edge of ray coverage; I suspect
> vertical smearing.
Response: We agree coverage thins there and have tested the concern directly.
A synthetic conduit restricted to <80 km depth, run through the identical
inversion, does not reproduce the observed anomaly at 120 km; recovery at the
key node is 78% (illustrative). The feature therefore cannot be explained as
smearing of shallow structure. New Supplementary Fig. S7 shows the full test;
the main text now states the recovery value where the conduit is first named.
Change: Section 4.2 (one sentence); Fig. S7; inversion configs updated at
Zenodo (DOI: illustrative).
The pattern: reproduce the referee's alternative inside your own machinery, show it fails, and spend one main-text sentence on it — the rest lives in the supplement.
Anti-patterns
- Growing the main text past the cap to appease referees instead of routing to the supplement
- Answering a technical objection with rhetoric or seniority
- Silently dropping a comment, or merging two comments to avoid one
- Capitulating on the portable claim, leaving a Letter with no letter-worthy result
- Updated figures whose underlying deposited tables were never refreshed
Output format
【Editor's decisive points】listed + resolved with evidence
【Coverage】every comment quoted and answered? [Y/N]
【Routing】additions → supplement/deposit; main text within cap? [Y/N]
【Rebuttals】each backed by a calculation/citation? [Y/N]
【Cross-field conflicts】reconciled for the editor? [Y/N or N/A]
【Sync】manuscript = supplement = deposit? [Y/N]
【Next】resubmit via Editorial Manager
Supplementary resources
../../resources/external_tools.md— tooling for referee-requested tests../../resources/official-source-map.md— decision flow and policy links
Version History
- 9f86f09 Current 2026-07-19 15:17
