Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting Critical Care Medicine or deciding whether an intensive/critical-care study fits this venue. Encodes the journal's ICU-focused fit, the clinical-evidence and translational bar, reporting-guideline and registration requirements, SCCM house style, official-submission re-check, and desk-reject heuristics. Venue-fit aid only, not clinical advice.
.claude/skills/brycewang-stanford-critical-care-medicine/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 84% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 3% | 0% |
Critical Care Medicine is the flagship journal of the Society of Critical Care Medicine (SCCM), publishing clinical and translational research centered on the care of the critically ill across the whole ICU — sepsis, ARDS, resuscitation, shock, multiorgan failure and organ support, and the systems and processes of critical-care delivery. Its defining expectation is a clinically important advance in intensive-care management or critical-illness mechanism that informs how clinicians care for ICU patients, not a narrow single-center series with no outcome relevance or a basic experiment without critical-illness anchoring. Unlike the broader pulmonary/critical-care flagship, Critical Care Medicine is ICU-discipline-focused and spans the whole critically ill patient, not just the lung. This skill is a fit / venue-selection / re-framing aid; it is not clinical or regulatory advice and does not replace the journal's current instructions for authors. Before submitting, re-check the live Critical Care Medicine author instructions.
study and wants a fit/framing check.
critical-illness mechanism with outcome relevance.
critical care), and The Lancet Respiratory Medicine.
expectations.
studies.
in the ICU context.
delirium management.
long-term/post-ICU outcomes.
to critically ill patients or relevant models.
(mortality, organ-failure-free or ventilator-free days, functional outcome); surrogate physiologic endpoints need justification.
for observational work, PRISMA for systematic reviews, ARRIVE for animal studies.
expected, and pragmatic/cluster designs need appropriate analysis.
selection bias, and missing data; causal language must match the design.
patients or validated models.
re-check current article types (Clinical Investigation, etc.) and limits on the live guide.
implication and bounds generalizability to ICU practice.
reports ARRIVE-aligned detail.
the protocol, full statistical methods, and additional analyses.
../../resources/source-basis.md and../../resources/official-source-map.md; start from the ICMJE/EQUATOR and SCCM anchors, then cite the current Critical Care Medicine page you checked.
the current version.
data/code-availability, and protocol/SAP submission.
animal-care/IACUC approval, ICMJE authorship and conflict-of-interest disclosure, funding, and AI-use disclosure.
win.
american-journal-of-respiratory-and-critical-care-medicine.the-lancet-respiratory-medicine.anesthesiology.journal-of-the-american-society-of-nephrology / kidney-international.jama / NEJM / The Lancet in the natural-science bundle).text[Fit] High / Medium / Low (one-line reason) [Target] Critical Care Medicine (SCCM) [Specialty tags] <sepsis / ARDS / resuscitation / organ support / ICU systems> [Study design / reporting guideline] <RCT-CONSORT / cohort-STROBE / review-PRISMA / animal-ARRIVE> [Method/evidence] <power, ICU endpoint, confounding control, registration> [Top risk] <the single most likely reason for rejection> [Official items to re-check] <article type / registration / checklist / consent (deferred) / ethics / disclosures> [Re-route suggestion] <if not a fit, a better-matched venue>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,108 | 17,469 | +44% | 1 | 1 | 0% | 1,999 | 3,679 | +84% | 0 | 0 | — |
case-02 | fail→pass | 21,785 | 13,282 | -39% | 1 | 1 | 0% | 2,743 | 2,970 | +8% | 0 | 0 | — |
case-03 | fail→pass | 28,056 | 15,348 | -45% | 1 | 1 | 0% | 3,094 | 3,383 | +9% | 0 | 0 | — |
case-04 | pass→pass | 20,454 | 34,254 | +67% | 1 | 1 | 0% | 2,530 | 3,318 | +31% | 0 | 0 | — |
case-05 | pass→fail | 13,978 | 17,889 | +28% | 1 | 1 | 0% | 1,512 | 3,870 | +156% | 0 | 0 | — |
case-06 | pass→pass | 8,245 | 11,152 | +35% | 1 | 1 | 0% | 1,455 | 3,703 | +155% | 0 | 0 | — |
case-07 | fail→pass | 21,457 | 11,975 | -44% | 1 | 1 | 0% | 2,606 | 2,757 | +6% | 0 | 0 | — |
case-08 | pass→pass | 19,125 | 9,804 | -49% | 1 | 1 | 0% | 2,164 | 3,008 | +39% | 0 | 0 | — |
case-09 | fail→pass | 21,891 | 12,586 | -43% | 1 | 1 | 0% | 2,808 | 2,884 | +3% | 0 | 0 | — |
case-18 | pass→pass | 17,645 | 14,173 | -20% | 1 | 1 | 0% | 2,110 | 3,247 | +54% | 0 | 0 | — |
case-10 | fail→pass | 29,762 | 9,673 | -67% | 1 | 1 | 0% | 2,844 | 2,511 | -12% | 0 | 0 | — |
case-11 | fail→pass | 16,394 | 12,499 | -24% | 1 | 1 | 0% | 2,653 | 2,839 | +7% | 0 | 0 | — |
case-12 | pass→pass | 15,946 | 13,571 | -15% | 1 | 1 | 0% | 2,778 | 3,037 | +9% | 0 | 0 | — |
case-13 | pass→pass | 15,944 | 11,481 | -28% | 1 | 1 | 0% | 2,197 | 2,798 | +27% | 0 | 0 | — |
case-14 | pass→pass | 16,018 | 12,200 | -24% | 1 | 1 | 0% | 2,269 | 2,862 | +26% | 0 | 0 | — |
case-15 | pass→pass | 17,538 | 12,804 | -27% | 1 | 1 | 0% | 2,079 | 2,873 | +38% | 0 | 0 | — |
case-16 | pass→pass | 15,362 | 6,847 | -55% | 1 | 1 | 0% | 2,752 | 2,764 | +0% | 0 | 0 | — |
case-17 | fail→pass | 18,719 | 6,859 | -63% | 1 | 1 | 0% | 2,392 | 2,875 | +20% | 0 | 0 | — |
case-19 | pass→pass | 12,578 | 17,078 | +36% | 1 | 1 | 0% | 2,082 | 3,246 | +56% | 0 | 0 | — |
case-20 | pass→pass | 17,511 | 7,546 | -57% | 1 | 1 | 0% | 2,218 | 2,980 | +34% | 0 | 0 | — |
case-21 | fail→fail | 22,027 | 13,770 | -37% | 1 | 1 | 0% | 2,925 | 2,814 | -4% | 0 | 0 | — |
case-22 | pass→pass | 19,449 | 14,368 | -26% | 1 | 1 | 0% | 2,557 | 3,166 | +24% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.