From jama-skills
Tightens prose and removes spin from JAMA manuscripts to match the AMA Manual of Style. Polishes language and framing without altering analysis or reported numbers.
How this skill is triggered — by the user, by Claude, or both
Slash command
/jama-skills:jama-writing-styleThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- Prose is verbose, hedged, or jargon-heavy for a general clinician audience
Spin is language that makes results sound better than the data support. JAMA reviewers and editors actively police it.
| Spin pattern | Fix |
|---|---|
| Null primary outcome, upbeat conclusion | State the primary result plainly; do not pivot to secondaries |
| Secondary/post hoc outcome framed as main finding | Foreground the pre-specified primary outcome |
| "Trend toward significance" for a non-significant result | Report the estimate + CI; avoid the phrase |
| Causal verbs ("reduced," "caused") for observational data | Use associational verbs ("was associated with") |
| "Safe and well tolerated" beyond what harms data show | Report harms quantitatively; calibrate the claim |
| Over-generalizing beyond the studied population | Bound the conclusion to the sample/setting |
At the Journal of the American Medical Association, manuscript editors and the AMA Manual of Style copy desk hold prose to a general-clinician readability standard and treat unsupported optimism as a fixable defect. The bar: a non-specialist clinician reads the Conclusions, knows what changes at the bedside, and trusts the wording does not exceed the data. Recurring fixes: Conclusions warmer than the Results get rewritten to the primary outcome; stigmatizing labels become person-first; a leading-zero "p=0.043" becomes exact P = .04 with the 95% CI; causal verbs on a cohort become "was associated with"; and generic limitations name the specific bias, population bound, and missing data.
Vignette (illustrative): a prospective cohort, N = 12,500 adults, finds a dietary pattern associated with lower incident heart failure; adjusted hazard ratio 0.88 (95% CI, 0.79-0.98).
The rewrite swaps the causal verb for associational language, keeps the estimate with its CI, and bounds the recommendation.
Calibration anchors (hedge where uncertain): person-first language, exact-P / no-leading-zero notation, and estimate-plus-CI pairing are durable AMA conventions; reference-formatting and race/ethnicity rules evolve — confirm against the latest AMA Manual of Style.
【Spin found】none / instances: ...
【Causal-vs-associational language correct】yes / fixes: ...
【Harms calibrated】yes / no
【Limitations specific】yes / no
【AMA style conformance】numbers/units/abbrev/refs: ...
【Next skill】jama-cover-letter
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin jama-skillsStructures and tightens NEJM Original Articles into terse IMRAD format with short main text, limited references, claim-first results, and sober discussion with limitations.
Polishes prose for Cancer Cell (Cell Press) manuscripts: refines Summary/Introduction/Results/Discussion, calibrates claims to evidence strength, and enforces Cell Press nomenclature and house style.
Polishes prose for JMR manuscripts: enforces 200-word third-person abstract, AMA author-year citations, exact p-values, and front-loaded research argument. For late-stage polish after contribution is settled.