Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting systematic literature reviews, meta-analyses, research synthesis, or comprehensive literature searches across biomedical, scientific, and technical domains. Creates professionally formatted markdown documents and PDFs with verified citations in multiple citation styles (APA, Nature, Vancouver, etc.).
.claude/skills/mkurman-literature-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 103% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 95% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 33% | 0% |
--------|-------------------|----------------| | 0-3 years | 20+ citations | Noteworthy | | 0-3 years | 100+ citations | Highly Influential | | 3-7 years | 100+ citations | Significant | | 3-7 years | 500+ citations | Landmark Paper | | 7+ years | 500+ citations | Seminal Work | | 7+ years | 1000+ citations | Foundational |
Prioritize papers from higher-tier venues:
Prefer papers from:
For any topic, identify foundational work by:
Complete workflow for a biomedical literature review:
bash# 1. Create review document from template cp assets/review_template.md crispr_sickle_cell_review.md # 2. Search multiple databases using appropriate skills # - Use gget skill for PubMed, bioRxiv # - Use direct API access for arXiv, Semantic Scholar # - Export results in JSON format # 3. Aggregate and process results python scripts/search_databases.py combined_results.json \ --deduplicate \ --rank citations \ --year-start 2015 \ --year-end 2024 \ --format markdown \ --output search_results.md \ --summary # 4. Screen results and extract data # - Manually screen titles, abstracts, full texts # - Extract key data into the review document # - Organize by themes # 5. Write the review following template structure # - Introduction with clear objectives # - Detailed methodology section # - Results organized thematically # - Critical discussion # - Clear conclusions # 6. Verify all citations python scripts/verify_citations.py crispr_sickle_cell_review.md # Review the citation report cat crispr_sickle_cell_review_citation_report.json # Fix any failed citations and re-verify python scripts/verify_citations.py crispr_sickle_cell_review.md # 7. Generate professional PDF python scripts/generate_pdf.py crispr_sickle_cell_review.md \ --citation-style nature \ --output crispr_sickle_cell_review.pdf # 8. Review final PDF and markdown outputs
This skill works seamlessly with other scientific skills:
Scripts:
scripts/verify_citations.py: Verify DOIs and generate formatted citationsscripts/generate_pdf.py: Convert markdown to professional PDFscripts/search_databases.py: Process, deduplicate, and format search resultsReferences:
references/citation_styles.md: Detailed citation formatting guide (APA, Nature, Vancouver, Chicago, IEEE)references/database_strategies.md: Comprehensive database search strategiesAssets:
assets/review_template.md: Complete literature review template with all sectionsGuidelines:
Tools:
Citation Styles:
bashpip install requests # For citation verification
bash# For PDF generation brew install pandoc # macOS apt-get install pandoc # Linux # For LaTeX (PDF generation) brew install --cask mactex # macOS apt-get install texlive-xetex # Linux
Check dependencies:
bashpython scripts/generate_pdf.py --check-deps
This literature-review skill provides:
Conduct thorough, rigorous literature reviews that meet academic standards and provide comprehensive synthesis of current knowledge in any domain.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 36,892 | 36,340 | -1% | 1 | 1 | 0% | 6,232 | 8,398 | +35% | 0 | 0 | — |
case-07 | fail→pass | 10,189 | 6,824 | -33% | 1 | 1 | 0% | 1,617 | 3,276 | +103% | 0 | 0 | — |
case-02 | pass→fail | 36,229 | 6,715 | -81% | 1 | 1 | 0% | 6,212 | 2,601 | -58% | 0 | 0 | — |
case-03 | fail→fail | 37,112 | 34,817 | -6% | 1 | 1 | 0% | 6,202 | 8,367 | +35% | 0 | 0 | — |
case-04 | pass→pass | 11,644 | 4,774 | -59% | 1 | 1 | 0% | 1,860 | 2,971 | +60% | 0 | 0 | — |
case-05 | fail→pass | 11,086 | 2,307 | -79% | 1 | 1 | 0% | 1,860 | 2,558 | +38% | 0 | 0 | — |
case-06 | pass→pass | 12,332 | 3,876 | -69% | 1 | 1 | 0% | 1,963 | 2,817 | +44% | 0 | 0 | — |
case-08 | fail→pass | 13,302 | 12,913 | -3% | 1 | 1 | 0% | 2,163 | 4,211 | +95% | 0 | 0 | — |
case-09 | pass→pass | 12,371 | 1,495 | -88% | 1 | 1 | 0% | 2,020 | 2,380 | +18% | 0 | 0 | — |
case-10 | fail→pass | 14,459 | 4,850 | -66% | 1 | 1 | 0% | 2,190 | 2,918 | +33% | 0 | 0 | — |
case-11 | fail→pass | 12,525 | 3,497 | -72% | 1 | 1 | 0% | 2,066 | 2,741 | +33% | 0 | 0 | — |
case-12 | pass→pass | 14,386 | 6,724 | -53% | 1 | 1 | 0% | 2,305 | 3,238 | +40% | 0 | 0 | — |
case-13 | pass→pass | 11,122 | 6,589 | -41% | 1 | 1 | 0% | 1,754 | 3,219 | +84% | 0 | 0 | — |
case-14 | pass→pass | 13,329 | 8,960 | -33% | 1 | 1 | 0% | 2,107 | 3,497 | +66% | 0 | 0 | — |
case-15 | fail→pass | 17,433 | 4,030 | -77% | 1 | 1 | 0% | 2,709 | 2,826 | +4% | 0 | 0 | — |
case-16 | fail→pass | 13,560 | 3,832 | -72% | 1 | 1 | 0% | 2,123 | 2,864 | +35% | 0 | 0 | — |
case-17 | fail→pass | 11,581 | 2,797 | -76% | 1 | 1 | 0% | 1,804 | 2,582 | +43% | 0 | 0 | — |
case-18 | pass→pass | 9,396 | 2,429 | -74% | 1 | 1 | 0% | 1,471 | 2,507 | +70% | 0 | 0 | — |
case-19 | pass→pass | 13,581 | 4,946 | -64% | 1 | 1 | 0% | 2,201 | 2,931 | +33% | 0 | 0 | — |
case-20 | fail→pass | 11,577 | 2,903 | -75% | 1 | 1 | 0% | 1,901 | 2,638 | +39% | 0 | 0 | — |
case-21 | fail→pass | 12,115 | 3,100 | -74% | 1 | 1 | 0% | 1,960 | 2,692 | +37% | 0 | 0 | — |
case-22 | pass→pass | 11,480 | 3,192 | -72% | 1 | 1 | 0% | 2,005 | 2,666 | +33% | 0 | 0 | — |
case-23 | fail→fail | 7,393 | 13,052 | +77% | 1 | 1 | 0% | 1,311 | 4,749 | +262% | 0 | 0 | — |
case-24 | fail→fail | 25,879 | 30,637 | +18% | 1 | 1 | 0% | 4,032 | 7,705 | +91% | 0 | 0 | — |
case-25 | fail→fail | 18,671 | 21,254 | +14% | 1 | 1 | 0% | 2,829 | 5,232 | +85% | 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. 25 cases were attempted, and 24 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 24 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.