From scimesh
Use when generating PRISMA flowcharts and synthesis for systematic literature review. TRIGGERS: slr export, synthesize, PRISMA, generate report, synthesis, generate report, export review
How this skill is triggered — by the user, by Claude, or both
Slash command
/scimesh:synthesizingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate PRISMA flowcharts and synthesis reports for systematic literature review.
Generate PRISMA flowcharts and synthesis reports for systematic literature review.
Generate final synthesis including PRISMA flowchart, included/excluded paper tables, and export options.
Prerequisite: Screening must be complete (scimesh:screening). Optionally, extraction too (scimesh:extracting).
# Screening statistics
uvx scimesh workspace stats {review_path}/
Output:
Total papers: 120
included: 31 (25.8%)
excluded: 85 (70.8%)
maybe: 4 (3.3%)
unscreened: 0 (0.0%)
Progress: 100%
# Generate PRISMA synthesis with mermaid flowchart and tables
uvx scimesh workspace prisma {review_path}/ -o {review_path}/synthesis.md
This generates a complete synthesis document with:
# Export included papers to BibTeX
uvx scimesh workspace export {review_path}/ --status included -f bibtex -o included.bib
# Export to RIS
uvx scimesh workspace export {review_path}/ --status included -f ris -o included.ris
# Export to CSV (spreadsheet-friendly)
uvx scimesh workspace export {review_path}/ -f csv -o all_papers.csv
# Export to JSON
uvx scimesh workspace export {review_path}/ --status included -f json -o included.json
# Export to YAML
uvx scimesh workspace export {review_path}/ -f yaml -o papers.yaml
# Screening statistics
uvx scimesh workspace stats {review_path}/
# List papers (table format)
uvx scimesh workspace list {review_path}/
# List unscreened papers
uvx scimesh workspace list {review_path}/ --status unscreened
# List included papers
uvx scimesh workspace list {review_path}/ --status included
# Generate PRISMA synthesis
uvx scimesh workspace prisma {review_path}/ -o synthesis.md
{review_path}/
├── index.yaml # Protocol + stats
├── searches.yaml # Search history
├── papers.yaml # Paper list with search_ids
├── synthesis.md # PRISMA + synthesis (generated)
├── included.bib # BibTeX of included papers
└── papers/
└── {year}/ # Organized by publication year
└── {paper-slug}/
├── index.yaml
├── fulltext.pdf
└── condensed.md
Before writing a narrative synthesis, check how many included papers have extraction data (condensed.md files). Present this to the user:
{
"question": "Extraction status: {extracted}/{included} papers have condensed.md. How to proceed?",
"header": "Synthesis",
"options": [
{"label": "Synthesize now (Recommended)", "description": "Use available data (abstracts + any extractions)"},
{"label": "Extract first", "description": "Run scimesh:extracting before synthesis"},
{"label": "PRISMA only", "description": "Generate PRISMA flowchart without narrative synthesis"}
],
"multiSelect": False
}
To count extraction status, check each included paper directory for a condensed.md file:
# List included papers
uvx scimesh workspace list {review_path}/ --status included
Then use Glob to check which of those paper directories contain condensed.md:
Glob(pattern="{review_path}/papers/**/condensed.md")
After PRISMA generation, write a narrative synthesis that ties the included papers together. The approach depends on whether extraction was done.
title, abstract, year, and any screening notes fields.Use this structure for the narrative synthesis section appended to synthesis.md:
## Narrative Synthesis
### Overview
Brief paragraph summarizing the scope: how many papers, what time period, what broad questions they address.
### Theme 1: {Descriptive Theme Name}
Discussion of papers in this theme. Every claim cites at least one paper.
Compare and contrast findings. Note agreements and contradictions.
### Theme 2: {Descriptive Theme Name}
...
### Theme 3: {Descriptive Theme Name}
...
### Cross-Cutting Discussion
How do the themes relate? What is the overall trajectory of the field?
Where do findings converge or diverge across themes?
### Gaps and Limitations
What questions remain unanswered? What methodological limitations are common?
What populations, domains, or conditions are underrepresented?
Do NOT just list papers sequentially. The synthesis must be organized by themes, not by paper.
The narrative synthesis must meet ALL of the following criteria:
{
"question": "How do you want to export the review?",
"header": "Export",
"options": [
{"label": "Full synthesis (Rec)", "description": "PRISMA + tables + narrative"},
{"label": "BibTeX only", "description": "Export citations for reference manager"},
{"label": "CSV summary", "description": "Spreadsheet-friendly format"}
],
"multiSelect": True
}
npx claudepluginhub gabfssilva/scimesh --plugin scimeshConducts systematic literature reviews across PubMed, arXiv, bioRxiv, Semantic Scholar and other academic databases. Generates professionally formatted markdown and PDF documents with verified citations in APA, Nature, Vancouver and other styles.
Conducts systematic literature reviews across PubMed, arXiv, bioRxiv, and Semantic Scholar, producing markdown and PDF output with verified citations. Use for meta-analysis, research synthesis, or broad literature searches in biomedical and scientific domains.
Use this skill for "write a literature review", "synthesize papers", "review the literature", "summarize research findings", "identify research trends", "gap analysis", "thematic review", "systematic review", "scoping review", "narrative review", "compare studies", "research synthesis", or when the user wants to synthesize multiple papers into a cohesive literature review.