返回提示词目录第 1363 页

起草期刊审稿决定通知信

依据既定编辑决定与审稿意见,拟写说明处理结果、必要修改及提交要求的作者通知信。

完整提示词

<context>
Authors read the decision letter more closely than anything else the journal sends, and they read it for what it lets them do next. A good one states the decision in the first two sentences using the journal's decision category, tells authors which concerns they must address for the paper to be acceptable and which are optional, resolves conflicting reviewer requests instead of passing them on, says exactly what to submit and by when, and is courteous without false encouragement. Poor letters forward the reviews with a one-line verdict, leave authors to satisfy contradictory demands, promise acceptance after a major revision, blur "reject" and "reject and resubmit" so authors cannot tell whether a new submission is welcome, or turn a desk reject into a page of criticism nobody asked for. This prompt drafts the letter once the editor has decided; weighing the reports to reach a decision is a separate job. Review material is confidential, and some journals restrict the use of AI tools with it.
</context>

<task>
Draft the decision letter{{#journal}} for {{journal}}{{/journal}}. Decision: {{decision}}.
{{#reviews}}
<reviews>
{{reviews}}
</reviews>
{{/reviews}}

<editor_notes>
{{editor_notes}}
</editor_notes>

1. Consistency check. Compare the decision with the reviews and the editor's notes. If the decision seems at odds with them (for example "minor" when a reviewer reports a flaw in the main analysis that the editor does not dismiss, or a reason in the notes that is about the authors rather than the paper), say so and say what would reconcile it. Do not change the decision yourself.
2. Write the letter. Every letter opens with the manuscript title and number (or placeholders), the decision in plain words in the first two sentences, and one or two sentences on why, in terms of the paper's contribution and the main issues. Then, by decision:
   - accept: the final checks still needed (data availability statement, reporting checklist, figure files, licence or copyright form) and what happens next (proofs, publication).
   - minor or major: "Essential revisions", numbered, each stating the problem, why it matters and what would resolve it, with its source (R1, R2 or Editor). Merge overlapping reviewer points. Where reviewers conflict, state which approach the editor prefers, following the editor's notes. Then "Optional suggestions", numbered and short. Then what to submit (a point-by-point response, a marked-up version) and the deadline from the notes or a placeholder. For major, say plainly that the revised paper will be re-reviewed and that acceptance is not guaranteed; for minor, say whether it will go back to reviewers or be checked by the editor.
   - reject-resubmit: say that this version is declined and a new submission is welcome, list the changes a new submission would need, and say it will be treated as a new manuscript, possibly with new reviewers.
   - reject: the main reasons, respectfully and specifically enough to help the authors elsewhere; a transfer offer only if the notes include one; no wording that implies a resubmission here is welcome.
   - desk-reject: short. The editor's reason in two or three sentences, that the paper was not sent for review so this is not a judgement of its full scientific merit, and a pointer to more suitable venues only if the notes give one. Do not add criticisms of your own.
   - Close courteously, refer to the reviewer reports below (except for a desk reject), and leave a signature placeholder.
3. Before you answer, check that every essential revision traces to a reviewer or the editor's notes, nothing in the letter reveals or hints at reviewer identity, the decision wording in the opening matches {{decision}}, and the tone fits it.
</task>

<constraints>
- Start with one line reminding the user to check that the journal allows AI assistance with confidential review material.
- Do not introduce new scientific criticisms of your own; if you notice a gap, put it in the note to the editor.
- Do not repeat hostile or personal remarks from a review in the letter; flag them in the note to the editor.
- Do not promise acceptance, and do not invent journal policies, deadlines or transfer options; use placeholders.
- If the decision is anything but desk-reject and the reviews are missing, or the editor's notes are too thin to justify the decision, ask for them and stop.
</constraints>

<output_format>
One reminder line, then:
## Consistency check
Two or three sentences: consistent, or the mismatch and what would resolve it.
## Decision letter
The letter, ready to edit, with placeholders in square brackets.
## Note to the editor
Points for the editor only: reviewer comments not to forward as written, possible conflicts, gaps you noticed, or "None".
</output_format>

可替换内容

  • {{#journal}}
  • {{journal}}
  • {{/journal}}
  • {{decision}}
  • {{#reviews}}
  • {{reviews}}
  • {{/reviews}}
  • {{editor_notes}}

分类与标签

办公效率、期刊编辑、审稿通知、修改要求、工作函件

前往 AI 图片创作