Manusights

Pre-submission quality control for scientific manuscripts.

[Manusights @manusights]

Manusights reviews your manuscript the way a tough, fair peer reviewer would: pressure-testing the science, the statistics, and the reporting before an editor ever sees it.

It catches what a quick read misses - a mean that's arithmetically impossible, a within-group test passed off as a real effect, a missing CONSORT spine that means a desk-reject (a few real examples below) and then does the part no other tool does: it tells you which journals to actually target, and the gaps to close before you submit.

It's trained on how 35+ active peer reviewers, including current Nature, Cell, and Science reviewers, judge real manuscripts. The best way to see it
is on your own paper.

Try it now → https://t.co/BZIM76RbyO

[Erik @erikinscience]

One thing we heard over and over in conversations with reviewers:

"Don't make me decode your paper."

Reviewers are volunteers. They're often reading manuscripts at night, between meetings, or on weekends.

The easier you make it for them to understand what you did, why it matters, and how the evidence supports the claim, the better your chances of getting a fair review.

A surprising number of review comments come down to confusion, not disagreement.

[Manusights @manusights]

One thing we heard repeatedly from researchers is that choosing the right journal can feel surprisingly opaque.

Most journals publish author guidelines.

Far fewer explain how editors think about fit, contribution, novelty, and audience.

That's one reason we spent so much time interviewing reviewers and editors when building Manusights.

A technically strong paper can still end up at the wrong door.

[Erik @erikinscience]

One thing we heard repeatedly from researchers is that the hardest part of publishing isn't always the science.

It's understanding what journals are actually looking for.

Most journals publish author guidelines. Far fewer publish how editors think about contribution, fit, positioning, and novelty.

That's one reason we spent so much time interviewing reviewers and editors when building Manusights.

[Bo Cowgill @bocowgill]

@auyonomous @manusights The feedback I got from @manusights was both candid (and correct) about the strengths and limitations of this manuscript. I've seen some of the feedback before (the hard way) -- which is itself validating. Seeing this feedback earlier would have been useful.

[Manusights @manusights]

The "hidden curriculum" framing is a great way to put it.

[Bo Cowgill @bocowgill]

I tried @manusights -- different/complementary proposition than @RefineDotInk. Refine seems focused on technical correctness (obviously valuable). Manusights motivated by "hidden curriculum" questions like journal choice, fit, contribution, and positioning/framing. Clear value in both approaches.

[Manusights @manusights]

@bocowgill Really appreciate you trying it!

We spent a lot of time interviewing reviewers and editors because fit, framing, and contribution are often just as important as technical correctness.

[Manusights @manusights]

“Just one more round of revisions!” they said.

#PeerReview

[Manusights @manusights]

exactly why we are building Manusights

[Erik @erikinscience]

Researchers audited 111 million references across 2.5 million papers and estimated that 146,932 hallucinated citations entered the scientific literature in 2025 alone.

The finding that stood out to me is that many of these citations made it through existing moderation and publication processes.

A lot of discussion around AI in science focuses on generation. This feels like a reminder that verification may be the more important problem.

https://t.co/nipNGeANhH

[Erik @erikinscience]

The biggest misconception in AI for science is that writing is the hard part.

Writing is becoming cheap, but verification is becoming expensive.

Does the citation exist?

Does the DOI resolve?

Does the cited paper actually support the claim?

Does the figure support the conclusion?

At Manusights, we built our review process around verification. Every review includes live citation checks, figure analysis, and feedback trained on patterns from 200+ pre-submission reviews.

Curious what your manuscript would look like through that lens?

Comment "manusights" and follow @manusights. We'll give away a free review (normally $39) to a few researchers this week.

[Manusights @manusights]

Save the prompts. Want the version that checks your citations and reads your figures for you? Comment "manusights" and follow, and the $29 review is free.

[Erik @erikinscience]

Three prompts you can use to surface problems in your paper before submission:

Prompt 1, the skeptic: "Act as Reviewer 2 for [target journal]. You want to reject this. List the 3 things most likely to trigger a desk rejection or major revisions, most severe first, and quote the exact sentences."

Prompt 2, find the contribution: "Read only my abstract. In one sentence, what is the novel contribution? If it is not clear in the first two sentences, tell me where it finally shows up."
If the model has to hunt for it, so will the editor.

Prompt 3, steelman the rejection: "State my main claim in one sentence. Now make the strongest argument that my evidence does not support it. Do not be nice. I am fixing this before a reviewer does."

Even well-prompted, a general LLM still will not verify your citations or read your figures. If you would rather not run all of these prompts and babysit the output, try @manusights: pre-submission review trained on 35+ real peer reviewers, with live citation checks and figure analysis. It is normally a $29 review. Comment "manusights" and follow @manusights, and I will DM you one free.

[Erik @erikinscience]

Three prompts you can use to surface problems in your paper before submission:

Prompt 1, the skeptic: "Act as Reviewer 2 for [target journal]. You want to reject this. List the 3 things most likely to trigger a desk rejection or major revisions, most severe first, and quote the exact sentences."

Prompt 2, find the contribution: "Read only my abstract. In one sentence, what is the novel contribution? If it is not clear in the first two sentences, tell me where it finally shows up."
If the model has to hunt for it, so will the editor.

Prompt 3, steelman the rejection: "State my main claim in one sentence. Now make the strongest argument that my evidence does not support it. Do not be nice. I am fixing this before a reviewer does."

Even well-prompted, a general LLM still will not verify your citations or read your figures. If you would rather not run all of these prompts and babysit the output, try @manusights: pre-submission review trained on 35+ real peer reviewers, with live citation checks and figure analysis. It is normally a $29 review. Comment "manusights" and follow @manusights, and I will DM you one free.

[nature @Nature]

Tens of thousands of publications from 2025 might include invalid references generated by AI, a Nature analysis suggests

https://t.co/SiIxVgJRZ7

[Erik @erikinscience]

the peer review system is breaking down and the field is having trouble saying it out loud.

it was designed for a world of maybe 100,000 papers a year. we now publish 4-5 million. there aren't enough qualified reviewers, and the ones who exist are already full-time researchers doing it for free, for journals that charge $5,000-$15,000 in APCs and return nothing.

what that produces: sloppy reviews, slow turnarounds, conflicts of interest that never get disclosed, and a selection effect where the most contrarian or surprising findings get blocked by entrenched reviewers while incremental work that confirms existing beliefs sails through.

the cancer immunotherapy timing paper. the GRAIL trial design. the entire replication crisis in nutrition, psychology, and social science. these aren't random failures. they're a system operating past its design limits.

the answer isn't "more peer review." it's transparency, post-publication scrutiny, registered reports, and actually compensating reviewers for the work they're doing. the current system is a prestige laundering machine and we keep acting like the laundering is the point.

[Manusights @manusights]

Here's what "not novel enough" actually means: your paper reads like 5 others they saw this week. Same framing, same angle, same claims. The work might be new but the pitch isn't

[Manusights @manusights]

You know what's wild? Half the papers I review bury their actual contribution on page 7. Meanwhile, the editor already decided by paragraph 2.

[Manusights @manusights]

@acagamic Messy preliminary data often contains the most interesting questions

[Manusights @manusights]

The dirty secret about "not novel enough" rejections? In 9 out of 10 cases, the novelty was there. Authors just buried it under three paragraphs of literature review. Editors won't dig for gold

[Manusights @manusights]

POV: Your paper's been "with editor" for 6 months

Month 1: They're being thorough

Month 3: Maybe they're on sabbatical?

Month 6: Pretty sure my paper is now structural support for their coffee mug

[Manusights @manusights]

@forallcurious So we're debugging cancer like it's faulty code that just needs the right patches - honestly pretty elegant.