Product
Seven agents read your manuscript end to end and hand it back as native Word tracked changes, every edit cited to a real rule.
adolescent-mh-mena.docx
Peer Review · Reviewed · 38 findings
cohort-attrition-v3.docx
Peer Review · Evidence agent reading
pilot-study-draft.docx
Peer Review · Reviewed · 12 findings
How it works
There is nothing to learn. You hand over a Word file and you get a Word file back.
Drop the manuscript and name the journal you are aiming at. No conversion, no plugin, no reformatting first.
Drop your manuscript
.docx up to 40 MB
Target journal
The Lancet Regional Health
Every paper gets the same read: a full Peer Review on our best models, weighed against your target journal's own rules. There is nothing to configure.
Seven agents · best models
Peer Review · MAX MODE
The same review on every paper
Structure, Clarity, Citations, Method, Evidence, Integrity, and Journal fit each read the whole manuscript and write their own edits.
One .docx carrying tracked changes, and the Waraq score across four lanes so you know where the paper is weak before you read a finding.
These are real tracked changes. Take the ones you agree with, reject the rest, and the paper stays in your voice.
The results utilise use a mixed-methods design.
Run the review again on the revised file and clear it against the journal's own rules before the desk ever sees it.
Ready to resubmit after 14 accepted changes
Inside the app
A walk through the four surfaces you will actually use at app.waraq.io.
Desk verdict
Revisions before submission
One number and four lanes. The lanes tell you where the paper is weak before you open a single finding, and the verdict tells you what a desk editor would decide about it today.
Attrition is not reported for the control arm.
The abstract runs 327 words; this journal allows 300.
Two references are listed but never cited in the text.
No finding arrives on its own authority. Every one names the checklist item, style rule, or journal requirement it came from, and points at the place in your manuscript it belongs to.
on: attrition is not reported for the control arm
Why does this need to change? The attrition is in the supplement.
Method agent
PRISMA item 16a asks for attrition per arm in the main text. Table S2 has the numbers; the edit moves them into Methods and keeps the supplement reference.
Disagree with an edit and ask why. The agent that wrote it answers from your paper and the rule it cited, not from a general opinion about writing.
The Lancet Regional Health
Elsevier
BMC Public Health
Springer Nature
Frontiers in Public Health
Frontiers
Your manuscript ranked against 15,000 journals on real scope and method fit, then checked line by line against the one you choose.
One review, done in full
Every plan runs the same review: a full Peer Review in MAX mode on our best models. Plans differ only by how many papers you review each period.
Seven agents · every paper
Every paper gets the same pass over structure, formatting, references, and the journal's own limits, returned as tracked changes in the same .docx.
Written for
Drafts between revisions, a co-author's chapter, and the last look before you press submit.
Max mode · best models · a few minutes
A full read by all seven agents on the strongest models, weighing method and evidence the way a senior reviewer does before the paper reaches a desk editor.
Written for
First submissions, revisions after a rejection, and any paper whose argument no senior reader has been through yet.
Every review is a full read, cited to your target journal's rules. Your plan refills automatically each period.
FAQ
Chatbots produce text. Waraq produces a reviewed .docx with native tracked changes, every change cited to a journal rule, plus a plan, a score, and a community. Different category.
Structure, methods, citations, clarity, and journal fit, from the same rule engine that reviews your paper. It is a readiness measure, not a judgment.
No. Waraq edits your writing and marks it as tracked changes. You accept or reject every change.
A maintained library of major publishers, plus any journal via pasted guidelines or a link to the author guidelines page.
Microsoft Word .docx in and out. The returned file opens in Word with native tracked changes and comments.
Manuscripts are processed for your review only, never used to train models, and deletable at any time.
Yes. Waraq reviews Arabic and English manuscripts, and the interface runs fully in both.
Upload the file you are working on now. Sign in with your email and a one-time code.
Waraq edits your writing. It does not write for you.