Skip to content
OpenAgentSkill
Browsing is open · Catalog captured 2026-11-21 · Automatic updates and account services are paused.
Other skills

sciwrite-publish

Navigate publication: choosing a journal, the submission process, responding to reviewers after a revise-and-resubmit, performing a peer review yourself, authorship decisions and order, plagiarism and self-plagiarism, and spotting predatory journals. Based on Stanford's Writing in the Sciences. Use this skill whenever someone mentions submitting a paper, a decision letter, reviewer comments, a rebuttal or response letter, "revise and resubmit", "my paper was rejected", being invited to review a manuscript, who should be an author or in what order, whether a passage counts as plagiarism, reusing their own text from a previous paper, an unsolicited invitation from a journal or conference, or whether a journal is legitimate. Also use it when someone asks how to choose where to submit. Do NOT use it for writing the manuscript itself (sciwrite-manuscript) or for grant applications or responding to grant or fellowship reviewers (sciwrite-proposals).

dvdsosa · Static checks on file

Installation

Read the source and dependencies before running a command. Copying does not install or execute anything.

npx skills add dvdsosa/scientific-writing-skills --skill sciwrite-publish

About this skill


name: sciwrite-publish description: > Navigate publication: choosing a journal, the submission process, responding to reviewers after a revise-and-resubmit, performing a peer review yourself, authorship decisions and order, plagiarism and self-plagiarism, and spotting predatory journals. Based on Stanford's Writing in the Sciences. Use this skill whenever someone mentions submitting a paper, a decision letter, reviewer comments, a rebuttal or response letter, "revise and resubmit", "my paper was rejected", being invited to review a manuscript, who should be an author or in what order, whether a passage counts as plagiarism, reusing their own text from a previous paper, an unsolicited invitation from a journal or conference, or whether a journal is legitimate. Also use it when someone asks how to choose where to submit. Do NOT use it for writing the manuscript itself (sciwrite-manuscript) or for grant applications or responding to grant or fellowship reviewers (sciwrite-proposals).

Publication, ethics and peer review

Answer in whatever language the user writes to you in.

One number is worth keeping in mind throughout, because it reframes what the rest of this plugin is for: roughly 60 % of reviewer criticisms concern the quality of the writing and the tables and figures, and only about 40 % the quality of the science. Good writing and good data presentation are not decoration on top of getting published — they are most of it.


Choosing where to submit

Do this before you write, not after. It determines formatting, length, reference style, and often the framing.

  • Identify who actually cares about these data. That may not be the most prestigious venue; a niche journal with the right readership often serves the work better.
  • Aim high, but be realistic. If you are junior, ask someone senior where the paper belongs.
  • Read the online instructions for authors in full. Length limits, reference counts and formats, table and figure requirements.
  • Pull a published paper from that journal and copy its table formatting, footnote symbols and section structure. This costs ten minutes and makes the submission look professional.
  • Check the journal is legitimate — see references/predatory-journals.md.

The submission process and what the letter means

Submission itself takes a couple of hours of online form-filling, plus time to collect conflict-of-interest and copyright forms signed by every author. Budget for it.

Four possible outcomes:

OutcomeReality
Accepted as isalmost never happens
Accepted with minor revisionsrare, and excellent
Rejected, resubmission possible ("revise and resubmit")this is what you are aiming for
Rejected outrightsubmit elsewhere — but fix it first

Revise-and-resubmit letters read like rejections. They open with something like your manuscript is not acceptable for publication, and then, further down, invite you to revise and resubmit. That second part is the actual message: the editor thinks the paper is publishable if you address the concerns. Most papers that eventually appear in journals go through this. Editors want a round of revision, because it is what makes the paper defensible.

On an outright rejection, the temptation is to send it elsewhere unchanged. Do not. You do not owe the reviewers a response, but the problems they found are real and the next set of reviewers will find them too. Fix them first.

After acceptance, review the proofs carefully. Errors are introduced during formatting and layout almost every time.


Responding to reviewers

A response letter is its own genre. See references/response-letter.md for the template and worked phrasing.

The essentials:

  • Be polite and appreciative, and mean it by the time you write. Your first instinct on reading criticism will be defensive. That is normal; get past it quickly, because the faster you can treat criticism as information rather than as an attack, the easier your career gets.
  • Respond point by point, numbered, mirroring the reviewers' numbering. For each: what you did, and exactly where in the manuscript it now appears.
  • You do not have to make every change. Small and inconsequential requests: just do them. Requests you genuinely disagree with: explain your reasoning, respectfully. Reviewers are sometimes wrong.
  • Include a tracked-changes copy. Most journals require it, and reviewers check that the changes were actually made.

One thing worth knowing, because it changes how you read a bad review: when a reviewer misidentifies a problem, there is usually still a problem. A reviewer who thinks your analysis is wrong may have been misled by a poor explanation of a correct analysis. In one course example, a reviewer appeared to misunderstand the background literature — and the real cause was that the authors had numbered their references wrong, attributing a fact to the wrong paper. The reviewer was wrong about the diagnosis and right that something was broken.


Doing a peer review

If you are invited, accept. Editors actively want junior reviewers: they are closer to current methods, and they are more careful — review quality demonstrably deteriorates the longer someone has been doing it. You will also see the back end of publication, read other reviewers' comments on the same paper, and discover that your judgement is better than you thought.

Full method, tone guidance and structure: references/peer-review.md.

The two things that matter most:

Tone. You are looking for problems, so a critical tone creeps in without your noticing. Criticise the work, not the authors. The authors should delete Table 5. Not only is it completely irrelevant, but it reveals their utter lack of statistical understanding and *Table 5 contains unnecessary information; for example, a Pearson's correlation coefficient may no

Read the complete source ↗