Claude
572b8acf8c
Make the library multi-platform without duplicating content. Each skills/<name>/SKILL.md body remains the single source of truth; a new generator renders platform-ready exports from it. - scripts/build-exports.mjs — dependency-free Node generator with a PLATFORMS registry so new platforms (Gemini, Cursor, …) are a few lines. Ships ChatGPT exports at exports/chatgpt/<bundle>/<skill>/SYSTEM_PROMPT.md (172 skills), plus generated index READMEs. Supports --platform and --check. - exports/ — generated ChatGPT system prompts, ready to paste into a Custom GPT. - .github/workflows/check-generated.yml — fails a PR if exports or web/skills.json drift from the source skills. - README "Works With" now documents the ready-to-use exports and regen command. - CHANGELOG + SKILL-AUTHORING-STANDARD note the generated artifacts. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016JWn5jRD5tcEFKrubjQ6Px
3.4 KiB
Literature Review Skill
Structures and writes literature reviews — from background sections of a dissertation through to standalone narrative reviews for publication.
Required Inputs
- Topic or research question
- Type of review (narrative / systematic / scoping / integrative / background section)
- Sources provided (paste references, abstracts, or key findings)
- Word count target
- Audience (academic journal / thesis / grant proposal / policy brief)
- Time period to cover
Output Structure
1. Search Strategy Summary (for systematic/scoping reviews)
Databases: [PubMed, EMBASE, PsycINFO, etc.] Search terms: [Key terms and Boolean combinations] Inclusion criteria: Study types, population, date range, language Exclusion criteria: [List] Results: [n] identified → [n] after deduplication → [n] screened → [n] included
2. Literature Review Body
Organised thematically — not chronologically. Each theme = one section.
Structure per thematic section:
[Theme heading]
[Opening: state what this section covers and what evidence shows overall]
[Evidence synthesis: present what multiple studies found, compare and contrast. Do NOT summarise one paper then the next — synthesise across them: "Three studies found X (Smith, 2019; Jones, 2020; Lee, 2021), while two found Y, with the difference attributable to..."]
[Critical analysis: note methodological strengths and weaknesses — sample sizes, study designs, generalisability, risk of bias]
[Closing: transition to next theme]
3. Synthesis Table (systematic/scoping reviews)
| Author, year | Study design | Population | n | Key findings | Quality/Limitations |
|---|
4. Gap Analysis
Well-established: [What literature consistently shows] Contested: [Areas where evidence is mixed and why] Missing: [Gaps the field needs to address] How your study addresses the gap: [If this is for a research proposal]
5. Conclusion Paragraph
[3-5 sentences. Current state of knowledge and what is needed next]
Critical Analysis Framework
For each paper: internal validity, external validity, bias types, effect size significance vs clinical significance, funding conflicts.
Quality Checks
- Organised thematically (not as individual paper summaries)
- Evidence synthesised across papers (not summarised one by one)
- Critical analysis of methodology included for key studies
- Gaps identified — what the field still needs
- All claims cited
Anti-Patterns
- Do not summarise papers one by one — evidence must be synthesised thematically across multiple studies, not presented as a sequence of abstracts
- Do not omit methodological critique — a literature review that only reports findings without assessing study quality is not a critical review
- Do not organise by chronology when thematic organisation is possible — chronological reviews bury the conceptual structure of the field
- Do not present contested findings as settled consensus — where evidence is mixed, name both sides and why the evidence diverges
- Do not skip the gap analysis — identifying what the field still needs is a core deliverable, not an optional addition
Example Trigger Phrases
- "Write a literature review on [topic]"
- "Synthesise the evidence on [topic] from these papers: [paste]"
- "Write the background section for my research proposal on [topic]"