K-Dense-AI/scientific-agent-skills / skills/venue-templates
venue-templates
VerifiedPrepares journal manuscripts, conference papers, research posters, and grant documents using venue-specific formatting guidance and bundled LaTeX scaffolds. Supports selecting an official template, checking current page or anonymity rules, adapting academic writing to a venue, or inspecting a submission PDF.
references/medical_journal_styles.md
View on GitHub# Medical Journal Writing Style
**Reviewed: 2026-10-01.** Resolve journal, article type, study design, and stage
before applying a numerical limit. A publisher family is not one author contract.
## Official author sources
| Target | Source and verification scope |
|---|---|
| NEJM | https://www.nejm.org/author-center — fresh retrieval returned navigation rather than complete article-type instructions; obtain the current full instructions before setting a word limit |
| The Lancet | https://www.thelancet.com/lancet/information-for-authors — the August 2026 [Information for Authors](https://www.thelancet.com/pb-assets/Lancet/authors/tl-info-for-authors-1787150940677.pdf) specifies a five-paragraph summary up to 300 words for original research and requires Research in Context; check the exact article category |
| JAMA | https://jamanetwork.com/journals/jama/pages/instructions-for-authors — original-data abstracts have a 350-word maximum and specified headings |
| The BMJ | https://www.bmj.com/about-bmj/resources-authors/article-types — research has no fixed main-text word limit; abstracts ordinarily use 250–300 words, with up to 400 for CONSORT/PRISMA reporting |
| Annals of Internal Medicine | https://www.acpjournals.org/journal/aim/authors — retrieval returned site navigation, not usable current instructions; verify article type before setting limits |
The previous generic 250/300/350/300/325 abstract table mixed article types and
unverified rules. Do not copy its numbers into a submission. JAMA's original
investigation requirements do not apply to every JAMA Network journal or section.
## Abstract structure
Choose the actual journal's headings; the following are writing patterns, not
universal required labels:
- **NEJM-style research:** Background, Methods, Results, Conclusions.
- **Lancet-style research:** Background, Methods, Findings, Interpretation, Funding.
- **JAMA original data:** Importance; Objective; Design, Setting, and Participants;
Interventions or Exposures; Main Outcomes and Measures; Results; Conclusions
and Relevance; applicable trial registration.
- **The BMJ research:** Objectives, Design, Setting, Participants, Interventions
where relevant, Main Outcome Measures, Results, Conclusions, Trial Registration.
Summary boxes such as The BMJ's “What is already known” and “What this study adds”
are distinct from the structured abstract. Do not import BMJ Open's template or
its separate strengths-and-limitations box into The BMJ without checking.
Use `assets/examples/medical_structured_abstract.md` (relative to the skill root)
for placeholder-based drafting. It intentionally contains no simulated drug
trial outcomes that could be mistaken for published clinical evidence.
## Match claims to evidence
Write the precise population, intervention/exposure, comparator, outcome, time
horizon, and estimand. Randomization supports causal interpretation when its
assumptions and execution hold; it does not automatically remove bias from
nonadherence, missing outcomes, outcome measurement, or selective reporting.
Small trials are not automatically “associational” merely because of size.
For observational analyses, use association language unless a causal estimand,
identification assumptions, design, and sensitivity analysis justify a carefully
qualified causal interpretation. Do not claim benefit from a case report.
A non-significant result does not establish equivalence or “similar safety.”
Describe the interval and which important benefit/harm values remain compatible
with it. Equivalence/noninferiority requires the relevant design, margin, and
analysis. Avoid stock claims about residual confounding as the main limitation
of a randomized intention-to-treat contrast; identify the actual source of bias.
## Numerical reporting
- Give group denominators, event counts, follow-up, effect estimates, and
uncertainty appropriate to the prespecified analysis.
- For binary outcomes, provide absolute risks/differences alongside relative
measures when estimable. A hazard ratio is not a risk ratio.
- State the confidence level and whether intervals/tests are adjusted for
multiplicity. A point estimate plus a P value is insufficient.
- Report exact P values where practical and follow the journal's small-value
convention. Statistical significance alone does not establish clinical value.
- Give NNT/NNH only when meaningful from an absolute effect at a stated time
horizon and baseline-risk context. Do not invert a hazard ratio or a
significance threshold to obtain NNT. If the risk-difference interval crosses
zero, the reciprocal interval is discontinuous; do not invent a finite range.
- Report prespecified outcomes regardless of direction or significance. Label
exploratory and post hoc analyses. Report adverse events systematically.
## Manuscript workflow
1. Establish the clinical question and the gap the design addresses.
2. Describe study design, setting, participants, recruitment, intervention or
exposure, outcomes, sample-size rationale, and prespecified analysis.
3. Report ethics review/consent or the documented exemption, and registration
where required. Never invent an approval number, funding source, or registry ID.
4. Show participant flow and analysis populations, including missing outcomes.
Loss to follow-up does not automatically justify excluding randomized people
from an intention-to-treat analysis.
5. Report results in the prespecified hierarchy, with uncertainty and harms.
6. Discuss strengths, bias, precision, generalizability, and clinical relevance
without extending conclusions beyond the data.
For diagnostic and prediction studies, specify the intended use, reference
standard/outcome timing, validation design, calibration, discrimination, and
threshold-dependent tradeoffs. A high AUC alone does not show patient benefit.
## Reporting guidelines
Use the current guideline and the applicable design extension. Check the
journal's requested checklist version rather than silently substituting one.
| Design | Current primary starting point |
|---|---|
| Randomized trial results | [CONSORT 2025](https://www.consort-spirit.org/), 30-item core checklist plus flow diagram |
| Randomized trial protocol | [SPIRIT 2025](https://www.consort-spirit.org/), 34-item core checklist |
| Observational cohort/case-control/cross-sectional | [STROBE](https://www.strobe-statement.org/), 22 items |
| Systematic review/meta-analysis | [PRISMA 2020](https://www.prisma-statement.org/), plus appropriate extensions |
| Diagnostic accuracy | [STARD 2015](https://www.equator-network.org/reporting-guidelines/stard/); check STARD-AI when applicable |
Reporting completeness does not establish low risk of bias, causal validity,
or adequate precision. These require assessment of what was actually done.
## Venue-specific AI policy
JAMA's current instructions prohibit AI drafting for Opinion, Letters, and
Online Comments, and advise against AI generation or formatting of references
(standard reference managers remain permitted). Check the exact requested work
against that policy; disclosure alone does not resolve a venue prohibition.
For allowed work, follow journal-specific disclosure and human-accountability
requirements. Other journals' policies must be checked independently.
## Final check
- Verify the exact headings and limits from the target article-type page.
- Match abstract numbers to the manuscript, tables, analysis, and registry.
- Keep study-design language, estimand, population, and time horizon consistent.
- Check flow denominators, missing-data reporting, prespecified outcomes, and harms.
- Include required checklist, data-sharing, funding, authorship, and interest statements.
- Inspect the uploaded preview, not just the local draft.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |