literature-review

What it does

Conduct systematic, comprehensive literature reviews following rigorous academic methodology. Searches multiple literature databases, synthesizes findings thematically, verifies citations for accuracy, and generates professional output documents in markdown and PDF formats. Integrates with tools for database access and provides specialized procedures for citation verification and document generation.

How it works

• Integrates with databases and tooling for searching (e.g., PubMed, arXiv, bioRxiv, Semantic Scholar) and for data handling (gget, bioservices, datacommons-client).
• Requires generating at least 1-2 AI-generated figures using the scientific-schematics skill before finalizing any document.
• Uses a structured multi-phase workflow: planning and scoping, systematic search across databases, screening and selection (deduplication, title/abstract/full-text screening), data extraction, thematic synthesis, critical analysis, and writing the discussion.
• In Phase 2, documents search parameters and exports results (including example markdown search strategies) and aggregates results via a post-processing script (python scripts/search_databases.py).
• In Phase 3, performs deduplication (python search_databases.py results.json --deduplicate), screens titles and abstracts against inclusion/exclusion criteria, and records final included studies; creates a PRISMA-like flow diagram.
• In Phase 4, extracts study data (metadata, design, size, findings, limitations, funding) and assesses quality with domain-appropriate tools.
• In Phase 5, writes a thematic synthesis rather than study-by-study summaries, structures results by themes, compares approaches, and highlights consensus and controversies.
• In Phase 6, verifies all citations using a verification script (python scripts/verify_citations.py my_literature_review.md) and ensures formatting in a chosen citation style (APA, Nature, Vancouver, etc.).
• In Phase 7, generates a PDF (python scripts/generate_pdf.py my_literature_review.md --citation-style apa) and conducts a final quality checklist (including PRISMA diagram, fully documented search methodology, and accurate references).

When to use it

Use this skill for:

• Conducting a systematic literature review for research or publication
• Synthesizing current knowledge across multiple sources
• Performing meta-analyses or scoping reviews
• Writing the literature review section of a research paper or thesis
• Investigating the state of the art and identifying gaps
• Requiring verified citations and professional formatting

What it can touch

• Databases and tools for retrieval and processing: gget, bioservices, datacommons-client
• Local scripts for processing: scripts/search_databases.py, scripts/verify_citations.py, scripts/generate_pdf.py
• Output formats: Markdown and PDF documents, with citations formatted in APA, Nature, Vancouver, etc.
• Visualization assets generated via the scientific-schematics skill using the described command pattern:

  python scripts/generate_schematic.py "your diagram description" -o figures/output.png
  

Caveats

• License is MIT; ensure compliance with any external data usage terms when integrating databases.
• Preprints and non-peer-reviewed sources may require careful verification and caution during synthesis.
• Visual schematics are mandatory and must be generated prior to finalizing documents.
• Citations must be verified for accuracy before final submission.