medical-imaging-review — Claude Code skill: what it does and setup
Structures medical imaging AI reviews with verifiable citations and qualitative synthesis. Rejects statistical pooling and meta-analyses.
npx skills add luwill/research-skills --skill medical-imaging-review -a claude-codeWhat medical-imaging-review does
The skill organizes evidence review workflows across four distinct routes: narrative reviews, method surveys, scoping reviews, and systematic qualitative reviews. Meta-analysis is ruled out completely. Pooling statistics, generating forest plots, or calculating heterogeneity metrics fall outside the environment's capabilities. Once literature selection starts, the chosen route becomes permanent.
Projects initialize through scripts/init_review_project.py into collision-safe folders stamped with timestamps and unique hashes. Standard generated files include review_config.yaml, claim_ledger.csv, references.bib, and manuscript templates. Search results are imported from the lit-search skill using scripts/import_litsearch.py, which records the PRISMA retrieval count into a tracking log.
Drafting forbids shifting numeric markers like [17]. Instead, each factual sentence connects to persistent citekeys like [@smith2024model] and enters claim_ledger.csv with an exact excerpt from the primary source.
What it helps with
- draft protocols for medical imaging AI surveys and scoping studies
- generate collision-safe project file structures with audit ledgers
- import screening candidate records from
lit-searchruns - link draft claims to first sources inside
claim_ledger.csv
Example prompts for the agent
- Initialize a scoping review project on chest CT lung nodule detection.
- Import the search exports from lit-search run folder run-04.
- Verify draft claims and log them into claim_ledger.csv.
How to install medical-imaging-review
1Skills CLI (Vercel)
npx skills add luwill/research-skills --skill medical-imaging-review -a claude-code2Manually: put the files in place
~/.claude/skills/medical-imaging-review/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/medical-imaging-review/
1Skills CLI (Vercel)
npx skills add luwill/research-skills --skill medical-imaging-review -a codex2Manually: put the files in place
.agents/skills/medical-imaging-review/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add luwill/research-skills --skill medical-imaging-review -a cursor2Manually: put the files in place
.cursor/skills/medical-imaging-review/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install luwill/research-skills medical-imaging-review2Manually: put the files in place
.github/skills/medical-imaging-review/SKILL.md1Skills CLI (Vercel)
npx skills add luwill/research-skills --skill medical-imaging-review -a gemini-cli2Platform CLI
gemini skills install https://github.com/luwill/research-skills --path medical-imaging-reviewManually: put the files in place
.devin/skills/medical-imaging-review/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add luwill/research-skills --skill medical-imaging-review -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about medical-imaging-review
How to install medical-imaging-review in Claude Code?
Run npx skills add luwill/research-skills --skill medical-imaging-review -a claude-code in your terminal.
Can medical-imaging-review run a meta-analysis?
No. The skill rejects requests for statistical pooling, forest plots, and heterogeneity numbers, as it is limited to qualitative synthesis.
How does the skill handle literature imports?
Candidate papers are retrieved by running lit-search, then imported with scripts/import_litsearch.py into a read-only export directory with a PRISMA count record.
