CONTEXT(CLONE)
Sign in

Pre-review a manuscript for method, statistics and ethics

For: Journal managing editor or handling editor triaging submissions before peer review

Pattern: Specialist panelNeeds live modelsDesigned for 5 to 120 agents
Get early accessSign in to use

The pain today

Reviewers are scarce. Manuscripts reach them with missing ethics statements, unreported methods or statistics that do not match the tables, and the review round is wasted.

The ask

“I attached a submitted manuscript, its supplementary files and our reporting checklist. Check it for study method, statistical reporting, ethics and consent statements, data availability and checklist items. Quote the passage behind each concern and say what is missing.”

Plain words, as you would say it to a colleague. Edit it to fit your case before you send it.

What you attach or connect

  • Manuscript
  • Supplementary files
  • Journal reporting checklist
  • Ethics and data availability policy

The unit of work

One worker task per one manuscript section per specialist brief.

Why a swarm fits

Method, statistics, ethics, data availability and reporting are separate checklists over the same sections. Each reviewer needs a section and one checklist, and a queue of submissions multiplies the width.

Not for

Judging novelty or importance, or replacing peer review. It checks reporting against the text, not whether the science is right.

The decision tree

6 typed decisions, each with an action for every answer

At fixed moments in a run, the engine puts one narrow question to a decision model. The decision model never writes text: it answers yes or no with a probability, picks from listed options, or gives a score, about a small slice of the material. The engine then does exactly what this tree says, which is what makes the run auditable. The thresholds are the template's design values, not measured results.

  1. Before workers, before a task runs

    Small worker or strong modelA choice among options

    Before a task runs

    Which treatment does this manuscript section need from the method, statistics, ethics or data availability brief it is paired with?

    Sees only: One section's heading and opening paragraph, and one brief's checklist headings

    Why: Statistical and design passages get the stronger reader; routine checklist items do not.

    • In scope, small workerthenAccept
    • In scope, dense statistics or designthenEscalate to a strong model
    • Out of scope for this briefthenSkip this unit

  2. After workers, the judge checks

    Evidence checkYes or no, with a probability

    After a worker answers

    Does the quoted manuscript passage report the method, test or consent procedure in the incomplete or inconsistent way the concern describes?

    Sees only: One checklist item, one concern and the manuscript passage it quotes

    Why: Editors will forward these concerns to authors, so each must be true to the text.

    • Yes: 0.85 or higherthenAccept
    • Unsure: 0.50 up to 0.85thenEscalate to a strong model
    • No: below 0.50thenReject and retry

  3. Evidence checkYes or no, with a probability

    After a worker answers

    Across the section headings, declarations and supplements shown, is the required ethics, consent or data availability statement absent?

    Sees only: One required statement and the manuscript's headings, declarations and supplement list

    Why: Telling an author a statement is missing when it sits in a supplement wastes a round.

    • Yes: 0.85 or higherthenAccept
    • Unsure: 0.50 up to 0.85thenMark unresolved
    • No: below 0.50thenReject and retry

  4. Reconciler, while merging

    Conflict checkYes or no, with a probability

    While reconciling

    Does the figure reported in this sentence differ from the figure for the same quantity in the quoted table cell?

    Sees only: One sentence from the results and the table row it refers to

    Why: Text and table mismatches are cheap to find and costly to publish.

    • Yes: 0.70 or higherthenMark unresolved
    • Unsure: 0.30 up to 0.70thenEscalate to a strong model
    • No: below 0.30thenAccept

  5. After workers, the judge checks

    Evidence checkYes or no, with a probability

    After a worker answers

    Is the pre-review note free of any recommendation to accept, reject or prioritise the manuscript?

    Sees only: The drafted pre-review note

    Why: The run checks reporting; judging the work is the editor's and reviewers' job.

    • Yes: 0.90 or higherthenAccept
    • Unsure: 0.60 up to 0.90thenReject and retry
    • No: below 0.60thenReject and retry

  6. Accountable person, before anything is settled

    Person decidesYes or no, with a probability

    Before anything is reported as settled

    Does the note contain any concern about ethics, consent or a text and table mismatch, or any comment that might be sent to the authors?

    Sees only: The pre-review note's list of accepted concerns

    Why: The handling editor decides what happens to the manuscript and what authors see.

    Accountable: The handling editor decides whether to send out, return or reject. Authors see only editor-approved comments.

    • Yes: 0.20 or higherthenAsk a person
    • Unsure: 0.05 up to 0.20thenAsk a person
    • No: below 0.05thenAccept

The fleet: who does what

Model tiers by role, not brands: you choose the models. Strong reasoning models plan and reconcile, small fast models do the wide work, and the judge is a decision model from a different family, so it does not share the workers' blind spots.

  1. Planner

    A strong reasoning model writes each brief from the journal's checklist and maps manuscript sections to briefs.

  2. Workers

    Small fast workers from mixed families, one per section and brief, quoting the passage behind each concern.

    Designed for 5 to 120 agents, one worker task per one manuscript section per specialist brief. Each worker receives only its own unit.

    Decisions here:1. Small worker or strong model

  3. Judge, from a different model family

    A decision model from a different family checks that the concern follows from the quoted passage or its absence.

    Decisions here:2. Evidence check3. Evidence check5. Evidence check

  4. Reconciler

    A strong reasoning model writes one pre-review note and keeps conflicting specialist views visible.

    Decisions here:4. Conflict check

  5. Accountable person

    The handling editor decides whether to send out, return or reject. Authors see only editor-approved comments.

    Decisions here:6. Person decides

Checked before anything is accepted

  • Each concern quotes the manuscript or states which section was searched and found silent
  • Figures reported in the text are compared with the same figures in tables
  • Missing statements are confirmed against the whole manuscript and supplements
  • The note is checked to contain no recommendation on acceptance

What comes back

  • Pre-review note per specialist area with quoted passages
  • Checklist items met, missing or unclear
  • Text and table figures that do not match
  • Questions to return to the authors

What to measure

  • Share of concerns the handling editor keeps
  • Review rounds lost to reporting gaps
  • Editor time per submission at triage
  • Cost per manuscript pre-reviewed

Names of measures only. No result is claimed for this template.

Templates open in the workspace chat with the ask filled in. Nothing runs until you send it.

Get early accessSign in to use

More in Education and publishing

Check that every citation supports its sentence

For: Production editor or research integrity officer at a publisher or university press

A monograph or report carries hundreds of citations.

Pattern: Map, verify, reduceNeeds scale6 decisionsDesigned for 40 to 1,200 agents

Map course modules to accreditation learning outcomes

For: Programme director or quality assurance lead preparing an accreditation review

Accreditors ask where each required outcome is taught and assessed.

Pattern: Hierarchical decompositionNeeds scale6 decisionsDesigned for 30 to 700 agents