About literature-research
A research skill that searches PubMed and can generate a deeper literature review with Claude. It is intended for researchers who need paper metadata in JSON and a Markdown synthesis of findings, themes, and gaps.
Key features
- 01Searches PubMed with a query
- 02Sets a maximum paper count
- 03Exports paper metadata as JSON
- 04Generates a Claude-assisted review
- 05Writes review findings in Markdown
Use cases
- 01Find papers for a research question
- 02Collect PubMed results for later analysis
- 03Summarize findings, themes, and gaps
Install
Copy the command and run it in a terminal.
$
npx skills add https:/After installing, type/literature-researchin Claude Code to confirm it loaded.
Bundled filesscripts/literature_deep_review.pyscripts/pubmed_literature_search.py
Details
From the project READMELiterature Research
Goal
Search academic databases and perform comprehensive literature reviews.
Scripts
./scripts/pubmed_literature_search.py- Search PubMed./scripts/literature_deep_review.py- Deep review with Claude
Usage
Search PubMed
python3 ./scripts/pubmed_literature_search.py \
--query "machine learning cancer diagnosis" \
--limit 50 \
--output .tmp/papers.json
Deep Review
python3 ./scripts/literature_deep_review.py \
--input .tmp/papers.json \
--output .tmp/review.md
Output
- JSON file with paper metadata
- Markdown review with key findings, themes, gaps
Schema
Inputs
| Name | Type | Required | Description |
|---|---|---|---|
query | string | Yes | Research query (e.g., 'machine learning cancer diagnosis') |
limit | integer | No | Max papers to fetch (default: 50) |
Outputs
| Name | Type | Description |
|---|---|---|
papers_file | file_path | JSON file with paper metadata |
review_file | file_path | Markdown review with key findings, themes, gaps |
Credentials
| Name | Source |
|---|---|
ANTHROPIC_API_KEY | .env |
Cost
Free (PubMed API) + Claude for review
Comments
Markdown supported