What problem does it solve?
This Skill automates the process of harvesting literature based on keyword search, building candidate tables, downloading PDFs or full texts, and deduplicating results, saving you time and effort in research projects.
Core Features & Use Cases
- Keyword-based Search: Search scholarly APIs (PubMed, Europe PMC, Crossref, OpenAlex) using any topic keywords.
- Candidate Table Generation: Create a table of candidate literature based on search results.
- PDF and Full Text Download: Download accessible PDFs or HTML/XML full texts from the candidate table.
- HTML-to-PDF Second Pass: Attempt to extract PDFs from saved HTML pages.
- Deduplication: Deduplicate downloaded files to ensure only unique documents are stored.
- Use Case: Imagine you are conducting a literature review on "genetic engineering". Use this Skill to automatically search for relevant papers, download the PDFs, and organize them into a deduplicated set.
Quick Start
Run the keyword-research-harvest skill with the following command:
keyword-research-harvest run --output-root "my_literature_runs" --config "config.json" --run-name "genetic_engineering_20231001"