medsci-skills 4.1.0

package/skills/find-journal/references/journal_profiles/The_BMJ.md

@@ -0,0 +1,31 @@
+# The BMJ
+
+## Identity
+- **Abbreviation:** BMJ
+- **Publisher:** BMJ Group (British Medical Association)
+- **ISSN:** 0959-8138 / 1756-1833
+- **Homepage:** https://www.bmj.com/
+- **Author guidelines:** https://www.bmj.com/about-bmj/resources-authors
+
+## Scope
+The BMJ publishes original research, reviews, and education articles across all areas of medicine with emphasis on studies relevant to clinical practice, public health policy, and evidence-based medicine. It prioritizes large observational studies, trials, and systematic reviews with broad international applicability.
+
+## Scope Keywords
+clinical trials, evidence-based medicine, public health, epidemiology, primary care, health policy, observational studies, systematic reviews, patient outcomes, medical education, health services research, global health, disease prevention, clinical guidelines, population health
+
+## Article Types Accepted
+- Research (Original Article)
+- Analysis
+- Clinical Review (commissioned)
+- Education (commissioned)
+- Practice
+- Editorials
+- Correspondence (Rapid Responses)
+
+## Classification
+- **Tier:** Q1
+- **Open Access:** Full OA (all research articles)
+- **Field:** General medicine
+
+## Special Notes
+The BMJ is the 3rd highest-impact general medical journal (IF 93.7). It uses open peer review with prepublication history posted alongside accepted papers. Requires a "What is already known / What this study adds" summary box and mandatory patient/public involvement reporting. AI policy: all tasks permitted with disclosure, AI images must be declared.

package/skills/find-journal/references/journal_profiles/The_Lancet.md

@@ -0,0 +1,31 @@
+# The Lancet
+
+## Identity
+- **Abbreviation:** Lancet
+- **Publisher:** Elsevier (The Lancet Group)
+- **ISSN:** 0140-6736 / 1474-547X
+- **Homepage:** https://www.thelancet.com/
+- **Author guidelines:** https://www.thelancet.com/publishing-with-the-lancet
+
+## Scope
+The Lancet publishes original research, reviews, and commentary on all aspects of human health, with particular emphasis on large clinical trials, global health, and studies that inform health policy. It prioritizes practice-changing evidence with immediate international relevance.
+
+## Scope Keywords
+clinical trials, global health, health policy, randomized controlled trials, epidemiology, infectious disease, chronic disease, mortality, therapeutics, public health, maternal health, child health, health systems, non-communicable diseases, clinical outcomes
+
+## Article Types Accepted
+- Article (Original Research)
+- Review
+- Seminar
+- Series
+- Comment
+- Correspondence
+- Viewpoint
+
+## Classification
+- **Tier:** Q1
+- **Open Access:** Hybrid
+- **Field:** General medicine
+
+## Special Notes
+The Lancet is the 2nd highest-impact general medical journal. It requires a mandatory "Research in Context" panel with systematic search strategy, and all figures are redrawn in-house into Lancet style. Uses midline dot (·) as decimal separator. AI policy: language editing only, dual disclosure required, AI images banned.

package/skills/find-journal/references/journal_profiles/The_Lancet_Digital_Health.md

@@ -0,0 +1,29 @@
+# The Lancet Digital Health
+
+## Identity
+- **Abbreviation:** Lancet Digit Health
+- **Publisher:** Elsevier (The Lancet Group)
+- **ISSN:** 2589-7500 / 2589-7500
+- **Homepage:** https://www.thelancet.com/journals/landig/home
+- **Author guidelines:** https://www.thelancet.com/publishing-with-the-lancet
+
+## Scope
+The Lancet Digital Health publishes original research, reviews, and commentary on digital technologies in health and medicine, including AI/ML clinical applications, telemedicine, wearable devices, electronic health records, and digital public health interventions. It emphasizes studies demonstrating real-world clinical impact of digital health tools.
+
+## Scope Keywords
+digital health, artificial intelligence, machine learning, deep learning, clinical decision support, telemedicine, wearable devices, electronic health records, medical imaging AI, diagnostic algorithms, mobile health, health informatics, digital therapeutics, prediction models, computer-aided diagnosis, natural language processing
+
+## Article Types Accepted
+- Article (Original Research)
+- Review
+- Comment
+- Correspondence
+- Viewpoint
+
+## Classification
+- **Tier:** Q1
+- **Open Access:** Full OA (Gold, CC BY 4.0)
+- **Field:** Medical informatics / Digital health
+
+## Special Notes
+Ranked #1 in medical informatics by Impact Factor (~24.1). Follows Lancet Group conventions including Research in Context panel, midline dot decimal separator, and in-house figure redrawing. Requires CLAIM or TRIPOD+AI checklists for AI/ML studies and encourages code availability statements. AI policy: language editing only, Methods disclosure required per ICMJE.

package/skills/find-journal/references/journal_profiles/World_Journal_of_Hepatology.md

@@ -0,0 +1,53 @@
+# World Journal of Hepatology
+
+## Identity
+- **Abbreviation:** World J Hepatol
+- **Publisher:** Baishideng Publishing Group Inc., Pleasanton, CA, USA
+- **ISSN:** 1948-5182 (online-only)
+- **Homepage:** https://www.wjgnet.com/1948-5182/index.htm
+- **Author guidelines (BPG common):** https://www.wjgnet.com/bpg/gerinfo/204
+- **Submission system:** F6Publishing — https://www.f6publishing.com
+
+## Scope
+World Journal of Hepatology is one of 47 peer-reviewed open-access journals of the Baishideng Publishing Group, dedicated to clinical and basic research on hepatobiliary diseases. Topical coverage includes liver cirrhosis, viral hepatitis (HBV, HCV), NAFLD/MASLD, alcohol-associated liver disease, autoimmune hepatitis, hepatocellular carcinoma, liver transplantation, and emerging biomarkers / molecular pathophysiology. Receptive to MASLD natural-history cohort studies, retrospective cohort analyses, and observational epidemiology with East-Asian or international generalizability.
+
+## Scope Keywords
+hepatology, liver disease, NAFLD, MASLD, alcoholic liver disease, ALD, MetALD, viral hepatitis, hepatitis B, hepatitis C, HCC, hepatocellular carcinoma, cirrhosis, liver fibrosis, FIB-4, retrospective cohort, observational study, biliary disease, liver transplantation, hepatobiliary
+
+## Article Types Accepted
+- Editorial (3.1)
+- Frontier (3.2)
+- Guidelines (3.3)
+- Field of Vision (3.4)
+- Opinion Review (3.5)
+- Expert Consensus (3.6)
+- Evidence Review (3.7)
+- Review (3.8)
+- Minireview (3.9)
+- Case-Control Study (3.10)
+- **Retrospective Cohort Study (3.11)** ← retrospective-cohort fit
+- Retrospective Study (3.12)
+- Clinical Trials Study (3.13)
+- **Observational Study (3.14)** ← retrospective-cohort fit
+- Prospective Study (3.15)
+- Randomized Controlled Trial (3.16)
+- Randomized Clinical Trial (3.17)
+- Basic Study (3.18)
+- Evidence-Based Medicine (3.19)
+- Systematic Review (3.20)
+- Meta-Analysis (3.21)
+- Scientometrics (3.22)
+- Case Report (3.23)
+- Correspondence / Letter to the Editor (3.24)
+- Correction / Retraction Note (3.25, 3.26)
+
+## Classification
+- **Tier:** Q3–Q4 hepatology (specialty open access; Q-tier varies year on year, verify at JCR)
+- **Open Access:** Full OA, **CC BY-NC 4.0** (no commercial reuse)
+- **Field:** hepatology — clinical and basic, broad Asian/international scope
+
+## Special Notes
+Baishideng publishes 47 OA journals; World Journal of Hepatology is the dedicated hepatology venue and accepts a highly granular taxonomy of study designs (each design has its own writing guidelines page, e.g. Retrospective Cohort Study at https://www.wjgnet.com/bpg/GerInfo/210, Observational Study at /200). All articles are externally peer-reviewed and selected by an in-house editor. CC BY-NC license means commercial reuse and AI-training reuse are explicitly restricted ("All rights are reserved, including those for text and data mining, AI training, and similar technologies"). 17-step editorial process is documented (initial check → peer review → editor evaluation → decision → revision → re-review → acceptance → PDF proof → publish online → peer-review report disclosure → contributor disclosure → author evaluation → article quality tracking → reader comments). APC verified at https://www.wjgnet.com/bpg/gerinfo/242 (changes; check current rate). Reference Citation Analysis (RCA) is a Baishideng-internal citation/quality tracking tool — not a substitute for JCR.
+
+## Acceptance Rate (estimated)
+~30–40 % (Baishideng OA family historically broad; World J Gastroenterol — the flagship — has reported acceptance ~30–35 % in recent years; World J Hepatol is typically slightly higher given lower submission volume).

package/skills/find-journal/references/journal_profiles/npj_Digital_Medicine.md

@@ -0,0 +1,29 @@
+# npj Digital Medicine
+
+## Identity
+- **Abbreviation:** npj Digit. Med.
+- **Publisher:** Nature Portfolio (Springer Nature)
+- **ISSN:** 2398-6352 (online only)
+- **Homepage:** https://www.nature.com/npjdigitalmed/
+- **Author guidelines:** https://www.nature.com/npjdigitalmed/submission-guidelines
+
+## Scope
+High-impact Nature-family journal publishing original research on digital technologies transforming healthcare, including AI/ML diagnostics, wearables, mobile health, digital biomarkers, and clinical decision support. Emphasizes translational impact with rigorous validation and reproducibility (code/data availability mandatory).
+
+## Scope Keywords
+digital medicine, artificial intelligence, machine learning, wearable sensors, digital biomarkers, mobile health, clinical decision support, remote patient monitoring, deep learning, digital therapeutics, health data science, predictive analytics, electronic health records, federated learning, computer vision in medicine, natural language processing
+
+## Article Types Accepted
+- Article
+- Review
+- Brief Communication
+- Perspective
+- Comment
+
+## Classification
+- **Tier:** Q1
+- **Open Access:** Full OA (CC BY 4.0)
+- **Field:** Digital health / Medical AI
+
+## Special Notes
+Nature Portfolio flagship for digital health; desk reject rate >50%. Methods section placed after Discussion (Nature convention). Data and Code Availability statements mandatory. Preprints encouraged. AI policy: LLMs cannot be authors; document AI tool use in Methods section.

package/skills/find-journal/skill.yml

@@ -0,0 +1,34 @@
+schema_version: 2
+name: find-journal
+layer: D
+owner_domain: journal_recommendation
+
+when_to_use: "Recommend journals for a manuscript via two-pass matching against the public profile library plus any user-local private profiles."
+when_NOT_to_use: "Adding a journal to the database (use add-journal); building the submission package (use sync-submission)."
+
+inputs:
+  - "manuscript abstract / scope"
+  - "study type"
+outputs:
+  - "ranked journal recommendations with scope rationale and AI-policy notes"
+side_effects:
+  - reads_journal_profiles
+downstream_consumers:
+  - add-journal
+  - write-paper
+  - sync-submission
+forbidden_actions:
+  - fabricate_impact_factor_or_apc
+  - recommend_journals_outside_the_profile_library
+
+# v2.1 quality card
+purpose: "Rank candidate journals by scope fit from the curated profile library, with AI-disclosure policy and homepage links."
+safety_boundaries:
+  - "No cached impact-factor/APC values are asserted; users verify current metrics at journal sites."
+  - "Recommendations are drawn from the curated profile library, not invented venues."
+known_limitations:
+  - "Coverage is limited to profiled journals; an unprofiled fit may be missed (add via add-journal)."
+  - "Scope-fit ranking is advisory, not an acceptance prediction."
+validation_commands:
+  - "verify shortlisted journals' current scope and metrics at their official sites"
+evidence_surface: manual_workflow

package/skills/fulltext-retrieval/SKILL.md

@@ -0,0 +1,174 @@
+---
+name: fulltext-retrieval
+description: Batch download open-access PDFs by DOI using legitimate OA APIs (Unpaywall, PMC, OpenAlex, Crossref). Optional PDF→Markdown conversion for token-efficient LLM analysis.
+triggers: PDF download, fulltext retrieval, open access PDF, batch download papers, meta-analysis PDF, PDF to markdown, convert PDF
+tools: Read, Write, Edit, Bash, Grep, Glob
+model: inherit
+---
+
+# Fulltext Retrieval Skill
+
+Batch download open-access full-text PDFs from a DOI list using legitimate OA APIs only.
+
+## Pipeline
+
+```
+DOI list → Unpaywall → PMC (Europe PMC / OA FTP / web) → OpenAlex → Crossref → landing page
+```
+
+Each DOI goes through these sources in order until a valid PDF (≥10 KB, `%PDF-` header) is found.
+
+## Quick Start
+
+```bash
+# Prepare a DOI list (one per line)
+cat > dois.txt << 'EOF'
+10.1007/s00330-010-1783-x
+10.1002/mp.12524
+10.1148/radiol.13131265
+EOF
+
+# Run
+python fetch_oa.py dois.txt --output pdfs/ --email your@email.com
+
+# Verbose mode for debugging
+python fetch_oa.py dois.txt -o pdfs/ -e your@email.com --verbose
+```
+
+## Input Formats
+
+**Plain text** — one DOI per line:
+```
+10.1007/s00330-010-1783-x
+10.1002/mp.12524
+```
+
+**TSV with header** — must contain a `DOI` column, optional `PMID` column:
+```tsv
+ID	Title	DOI	PMID	Year
+1	Some paper	10.1007/s00330-010-1783-x	20628747	2010
+```
+
+When a PMID is available, the PMC lookup is more reliable (PMID → PMCID conversion).
+
+## PMC Download (JS-Challenge Resistant)
+
+PMC web pages may block automated downloads with JavaScript proof-of-work challenges. This tool uses three fallback methods:
+
+### Method A: Europe PMC REST API (most reliable)
+
+```bash
+PMCID="PMC9733600"
+curl -sLo output.pdf \
+  "https://europepmc.org/backend/ptpmcrender.fcgi?accid=${PMCID}&blobtype=pdf"
+```
+
+### Method B: PMC OA FTP Service
+
+```bash
+curl -s "https://www.ncbi.nlm.nih.gov/pmc/utils/oa/oa.fcgi?id=${PMCID}" | \
+    grep -oE 'href="[^"]*\.pdf"' | head -1 | \
+    sed 's/href="//;s/"//' | xargs curl -sLo output.pdf
+```
+
+### DOI/PMID → PMCID Conversion
+
+```bash
+# Works with both DOI and PMID
+curl -s "https://www.ncbi.nlm.nih.gov/pmc/utils/idconv/v1.0/?ids=${DOI}&format=json" | \
+    python3 -c "import sys,json; print(json.load(sys.stdin)['records'][0].get('pmcid',''))"
+```
+
+## Output
+
+- PDFs saved as `{DOI_safe}.pdf` (slashes replaced with underscores)
+- `manual_needed.txt` — DOIs that could not be retrieved via OA
+- Summary with OA/PMC/fail/skip counts
+
+## Requirements
+
+- Python 3.10+ (stdlib only, no pip dependencies)
+- Contact email (required by Unpaywall Terms of Service)
+
+## API Policies
+
+| Source | Rate Limit | Notes |
+|--------|-----------|-------|
+| Unpaywall | 100 req/sec | Email required |
+| NCBI PMC | 3 req/sec without API key | Add `&api_key=` for higher limits |
+| OpenAlex | 100k req/day | Polite pool with email in User-Agent |
+| Crossref | 50 req/sec with email | Plus service with `mailto:` in UA |
+| Europe PMC | No documented limit | Be polite, ≤1 req/sec recommended |
+
+The script uses 0.3–0.5 second delays between requests.
+
+## PDF → Markdown Conversion (Optional)
+
+After downloading PDFs, convert them to LLM-friendly Markdown for token-efficient repeated analysis. Uses [pymupdf4llm](https://github.com/pymupdf/RAG) — optimized for academic papers with two-column layout handling and table preservation.
+
+### Quick Start
+
+```bash
+# Install (one-time)
+pip install pymupdf4llm
+
+# Convert all PDFs in a directory
+python pdf_to_md.py pdfs/
+
+# Convert with verbose output
+python pdf_to_md.py pdfs/ -v
+
+# Custom output directory
+python pdf_to_md.py pdfs/ -o markdown/
+
+# First 10 pages only (useful for long supplements)
+python pdf_to_md.py pdfs/ --pages 0-9
+
+# Overwrite existing conversions
+python pdf_to_md.py pdfs/ --force
+```
+
+### Combined Workflow
+
+```bash
+# Step 1: Download PDFs
+python fetch_oa.py dois.txt -o pdfs/ -e your@email.com
+
+# Step 2: Convert to Markdown (only successful downloads)
+python pdf_to_md.py pdfs/ -v
+```
+
+After conversion, `.md` files sit alongside `.pdf` files. Claude Code can then use `Read` for full content or `Grep` for targeted extraction — significantly more token-efficient than re-reading PDFs.
+
+### When to Convert
+
+| Scenario | Recommendation |
+|----------|---------------|
+| Screening/triage (read once) | Skip — read PDF directly |
+| Data extraction from k≥5 studies | Convert — repeated reads save tokens |
+| Meta-analysis full pipeline | Convert — papers referenced across multiple phases |
+| Single paper deep review | Optional — marginal benefit |
+
+### Academic Paper Defaults
+
+- **Images**: Skipped (saves tokens; figures referenced by caption text)
+- **Tables**: `lines_strict` strategy (preserves grid-line tables accurately)
+- **Layout**: Two-column academic layout handled automatically
+- **Headers/footers**: Removed by pymupdf4llm
+
+### Dependency Note
+
+`pdf_to_md.py` requires [pymupdf4llm](https://pypi.org/project/pymupdf4llm/) (AGPL-3.0). This is an **optional** dependency — `fetch_oa.py` remains stdlib-only with zero external dependencies. The AGPL license applies to pymupdf4llm itself, not to this skill.
+
+## Limitations
+
+- Only retrieves **open-access** articles. Paywalled articles require institutional access.
+- Landing page scraping may fail on publisher-specific JavaScript-heavy pages.
+- Some recent articles may not yet be indexed by OA sources.
+- PDF→Markdown quality depends on the PDF's text layer. Scanned-only PDFs may produce poor output.
+
+## Anti-Hallucination
+
+- **Never fabricate file paths, URLs, DOIs, or package names.** Verify existence before recommending.
+- **Never invent journal metadata, impact factors, or submission policies** without verification at the journal's website.
+- If a tool, package, or resource does not exist or you are unsure, say so explicitly rather than guessing.