Use when targeting Radiology (RSNA) or deciding whether a medical-imaging study fits this venue. Encodes the journal's fit, the diagnostic-accuracy and imaging-methodology bar, STARD/CLAIM reporting and reproducibility expectations, RSNA house style, official-submission re-check, and desk-reject heuristics. Venue-fit aid only, not clinical advice.
SkillScore92Maintained
Automatic
◆Research pack
Last verified 9/1/2026
What it changes
Passed 22 of 24 test cases with the skill, and 15 of 24 without it.
Without skill63%
With skill92%
Checks it passed that the model alone missed
Rates fit as High
Specifies STARD as the reporting guideline
Specifies STARD as the required reporting guideline
and 6 more in the full report
8 fixed14 passed either way1 still failing1 broke
Pass-rate lift
+29 pp
Tokens
+13%
Agent turns
0%
Response time
-35%
Model gemini-3.6-flashMeasured 9/1/2026v122 of 24 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
We conducted a prospective multi-center trial across 4 medical centers assessing ultra-fast diffusion-weighted MRI for hyperacute ischemic stroke diagnosis against standard 24-hour MRI/CT follow-up reference standard in 600 emergency patients. Evaluate fit and provide structured feedback fields.
What was checked, without and with the skill
Checked
Without skill
With skill
Rates the overall fit as High
Passed
Passed
Specifies STARD as the required reporting guideline
Missed
Passed
Identifies prospective multi-center design as a methodological strength
Passed
Passed
case-06 · gemini-3.6-flash · 9/1/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md116 linessingle file
Radiology (radiology)
Journal positioning
Radiology is the flagship journal of the Radiological Society of North America (RSNA),
publishing original research across diagnostic and interventional imaging — imaging
physics and technique, diagnostic accuracy, image-guided intervention, and imaging
artificial intelligence — with a strong emphasis on rigorous design, adequate sample
size, and clinical relevance. The defining expectation is a methodologically sound
imaging study with a clinically meaningful question and an appropriate reference
standard, not a small retrospective series or an AI model evaluated on a single
internal dataset. 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. Before submitting, re-check the live Radiology author instructions.
When to trigger
The author names Radiology for a diagnostic-imaging, imaging-physics, interventional,
or imaging-AI study and wants a fit/framing check.
An imaging study must be re-framed around a clinically meaningful diagnostic or
outcome question with a valid reference standard.
The author is choosing between Radiology, a subspecialty imaging journal, and a
general clinical journal.
The author needs the journal's diagnostic-accuracy reporting and reproducibility
expectations (STARD, CLAIM for AI).
Scope & topic fit
Diagnostic-accuracy studies across modalities (CT, MRI, ultrasound, PET, radiography)
with an appropriate reference standard.
Imaging physics, acquisition, reconstruction, and quantitative-imaging biomarker
development and validation.
Image-guided and interventional procedures with outcome data.
Artificial intelligence and machine learning for imaging, with rigorous training/
validation/test design and external validation.
Prognostic and screening imaging studies with clinically meaningful endpoints.
Method & evidence bar
Diagnostic-accuracy studies need an adequate, representative sample, a valid and