SuperAPI AI Directory

pubmed-fulltext-access

General & other

View repository

Updated 10-16 09:44

About this skill

Full Text Access Guide

Description

The current full-text workflow is centered on get_fulltext . It automatically tries Europe PMC, Unpaywall, and CORE, and can expand to more sources when necessary. If you need figures, text mining, or institutional subscription links, use other dedicated tools in conjunction.

Triggers

  • "I want to see the full text"
  • "Is there a PDF?"
  • "Is this open access?"
  • "Help me grab the methods or results section"
  • Mentions PMC, full text, PDF, open access

Core Tools

1. get_fulltext

get_fulltext(identifier="PMC7096777")
get_fulltext(pmid="12345678")
get_fulltext(doi="10.1038/s41586-021-03819-2")

Key Parameters

  • identifier: automatically detects PMID, PMCID, DOI
  • pmcid / pmid / doi: can also be passed separately
  • sections: only grab specific sections, for example "introduction,methods,results"
  • include_pdf_links: whether to return a PDF link
  • include_figures: whether to include figure metadata
  • extended_sources: whether to expand to more sources

Most Common Examples

1. Directly grab PMC full text

get_fulltext(identifier="PMC7096777")

2. View only methods and results

get_fulltext(
    pmid="12345678",
    sections="methods,results"
)

3. Use DOI to find OA versions

get_fulltext(
    doi="10.1038/s41586-021-03819-2",
    extended_sources=True
)

4. Full text along with figures

get_fulltext(
    pmcid="PMC7096777",
    include_figures=True
)

Figures and Visual Data

Retrieve article figures

get_article_figures(identifier="PMC12086443")
get_article_figures(pmid="40384072")

Suitable for:

  • Need to extract figure captions and image URLs separately
  • Want to quickly find flowcharts, result figures, microscopy images
  • Need image data that is more structured than the full text

Text Mining

Retrieve text-mined terms

get_text_mined_terms(pmcid="PMC7096777")
get_text_mined_terms(pmid="12345678", semantic_type="CHEMICAL")

Commonly used semantic_type:

  • GENE_PROTEIN
  • DISEASE
  • CHEMICAL
  • ORGANISM
  • GO_TERM

When there is no open access

Institutional Subscription Workflow

list_resolver_presets()
configure_institutional_access(preset="exlibris_sfx", base_url="https://your-library...")
test_institutional_access()
get_institutional_link(pmid="12345678")

This set of tools is suitable for:

  • The institution is known to have a subscription, but the article is not OA
  • Want to convert PubMed/DOI into a library resolver link

Recommended Workflow

Scenario 1: Get the full text directly from an article

fetch_article_details(pmids="12345678")
get_fulltext(pmid="12345678", sections="abstract,results")

Scenario 2: After searching, pick a representative article and read the full text

unified_search(
    query="remimazolam ICU sedation",
    limit=10,
    ranking="quality"
)

# 對選中的 PMID 再做全文抓取
get_fulltext(pmid="12345678", extended_sources=True)

Scenario 3: Grab the full text first, then extract figures and entities

get_fulltext(pmcid="PMC7096777", include_figures=True)
get_article_figures(pmcid="PMC7096777")
get_text_mined_terms(pmcid="PMC7096777", semantic_type="CHEMICAL")

Tool Selection Rules

Only want to get the full text or PDF

First use get_fulltext

Want to grab figures

Use get_article_figures

Want to extract gene / disease / chemical

Use get_text_mined_terms

The article is not OA, but you have a library account

Use configure_institutional_access + get_institutional_link


Final Principle

The full-text workflow has converged into a single public entry point: first use get_fulltext, and when figures or text mining are needed, supplement with dedicated tools.