Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› jacs-revision
jacs-revision
GitHub用于处理JACS稿件修改意见，协助分类评论、规划实验及起草点对点回复信。通过建立修订台账追踪关键证据缺口，确保回复专业严谨，不伪造数据，旨在高效完成论文修订与反驳。
Trigger Scenarios
收到JACS决定信（大修/小修/拒稿重投）
需要整理审稿人意见并分配至实验或编辑任务
正在撰写点对点回复信
需决定申诉或转投其他期刊
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill jacs-revision -g -y
SKILL.md
Frontmatter
{
"name": "jacs-revision",
"description": "Use when a decision letter\/referee reports arrive for a Journal of the American Chemical Society (JACS) manuscript, to plan experiments and draft the point-by-point response. Structures the revision and rebuttal — it does not fabricate data or guarantee acceptance."
}
Revision and Response to Reviewers (jacs-revision)
When to trigger
- A decision letter (major/minor revision, or reject-with-encouragement) arrived
- You need to triage referee comments into experiments vs edits vs rebuttals
- You are drafting the point-by-point response letter
- You must decide whether to appeal or to resubmit elsewhere
Triage the reports first
Sort every numbered comment into one bucket:
| Bucket | Action |
|---|---|
| Data-required | A control, extra characterization, X-ray, kinetics, or scope substrate is genuinely needed → do the experiment |
| Clarification | A real point the text/figure can fix without new data → edit + point to the change |
| Defensible disagreement | Reviewer is mistaken or asks beyond scope → rebut politely with evidence |
| Optional/over-reach | Nice-to-have; address briefly or explain why out of scope |
Address the rigor-critical comments (purity, characterization, mechanism support, reproducibility) first — these are the ones that block acceptance at JACS.
Response-letter structure
- Opening — thank the editor/reviewers; summarize the main changes and any new data/figures/SI added.
- Point by point — quote each comment, then respond. For each: state the change, where it is in the revised manuscript/SI (section, figure, page), and the new evidence.
- Quote the new text — paste added/changed passages so reviewers don't hunt.
- Rebuttals — for disagreements, be respectful and evidence-based; concede where the reviewer improves the paper.
- Marked manuscript — provide a tracked/colored version if requested.
Doing the new chemistry well
- New experiments must meet the same rigor bar: full characterization, controls,
CCDC deposition for any new structure, reproducible procedures (route to
jacs-methods). - New compounds → add full SI entries and spectra copies (route to
jacs-supplementary). - If a requested experiment fails or contradicts a claim, report it honestly and adjust the claim rather than hiding it.
Tone
- Professional, specific, and complete; never dismissive even when disagreeing.
- Make the reviewer's job easy: cross-reference every change to a location.
- If a comment is based on a misreading, fix the text so the next reader can't misread it too.
JACS revision ledger
Before drafting prose, make a revision ledger that separates editor-critical chemistry from ordinary cleanup. JACS revisions fail when a response looks complete but leaves one evidence gap unclosed. Track each item with four fields:
| Field | JACS-specific check |
|---|---|
| Claim under review | What exact novelty, scope, mechanism, selectivity, or structure claim did the referee challenge? |
| New evidence | Which experiment, characterization trace, spectrum, crystallographic deposit, kinetic run, or control now supports it? |
| Manuscript/SI location | Main-text figure/table, scheme, or SI page where the reader can verify the fix without searching. |
| Claim adjustment | Whether the language is strengthened, narrowed, or hedged because the new evidence changed the inference. |
Use the ledger to decide whether a comment is truly answered. A reviewer asking for "more evidence" is not answered by a paragraph; it is answered by a trace, table, structure, or explicitly narrowed claim. Conversely, a referee misunderstanding that came from ambiguous prose should produce both a response-letter explanation and a revised sentence in the manuscript so the same misreading cannot recur in the next round.
Checklist
- Every numbered comment has a response — none skipped
- Rigor-critical comments addressed with real data, not words
- New compounds/structures fully characterized and deposited (CCDC)
- Each response cites the exact location of the change
- Added/changed text quoted in the letter
- Disagreements rebutted with evidence and courtesy
- Claims softened or revised where new data require it
- Marked-up manuscript prepared if requested
Anti-patterns
- Responding "we have addressed this" without saying where or how
- Adding new compounds without adding their SI characterization
- Arguing with a referee instead of fixing the ambiguous text
- Quietly dropping an inconvenient result instead of reporting it
- Over-claiming again in the revision after softening in round one
- Ignoring a rigor request and hoping the editor overrides the referee
Output format
【Decision】major / minor / reject-with-encouragement
【Triage】data-required: [...]; clarify: [...]; rebut: [...]
【New experiments】done? characterized + deposited? Y/N
【Response letter】point-by-point complete? Y/N
【Claims adjusted】[...]
【Next】resubmit via Paragon Plus → await decision
Related resources
jacs-methods— rigor bar for new experimentsjacs-supplementary— adding SI for new compoundsjacs-referee-strategy— anticipated objections to map onto the actual reports
Version History
- 1839142 Current 2026-07-05 14:00
