u9401066/pubmed-search-mcp · Archived

pubmed-fulltext-access

Full text access using get_fulltext, figure extraction, and institutional link tools. Triggers: ?

First seen Jul 15, 2026

Installation

$ npx skills add u9401066/pubmed-search-mcp --skill pubmed-fulltext-access

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from u9401066/pubmed-search-mcp · top by installs.

npx skills add u9401066/pubmed-search-mcp

Browse all from u9401066/pubmed-search-mcp

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

Stars 27
License LICENSE
Default branch master
Open issues 2
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 4,133 B
  • docs SUMMARY.md 193 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

全文取得指南

描述

目前的全文 workflow 以 get_fulltext 為核心。它會自動嘗試 Europe PMC、Unpaywall、CORE,必要時可擴展到更多來源。若需要圖表、文字探勘或機構訂閱連結,則搭配其他專用工具。

觸發條件

  • 「我要看全文」
  • 「有 PDF 嗎?」
  • 「這篇有 open access 嗎?」
  • 「幫我抓方法或結果段落」
  • 提到 PMC、全文、PDF、開放取用

核心工具

1. get_fulltext

get_fulltext(source={"kind":"pmcid","value":"PMC7096777"})
get_fulltext(source={"kind":"pmid","value":"12345678"})
get_fulltext(source={"kind":"doi","value":"10.1038/s41586-021-03819-2"})

參數重點

  • source: 必填的 discriminated object;kind 明確指定 pmidpmciddoivalue 放對應識別碼
  • sections: 只抓特定段落,例如 "introduction,methods,results"
  • includepdflinks: 是否回傳 PDF 連結
  • include_figures: 是否一起帶 figure metadata
  • extended_sources: 是否擴展到更多來源

最常用範例

1. 直接抓 PMC 全文

get_fulltext(source={"kind":"pmcid","value":"PMC7096777"})

2. 只看方法與結果

get_fulltext(
    source={"kind":"pmid","value":"12345678"},
    sections="methods,results"
)

3. 用 DOI 找 OA 版本

get_fulltext(
    source={"kind":"doi","value":"10.1038/s41586-021-03819-2"},
    extended_sources=True
)

4. 全文連同 figures

get_fulltext(
    source={"kind":"pmcid","value":"PMC7096777"},
    include_figures=True
)

圖表與視覺資料

取得文章圖表

get_article_figures(source={"kind":"pmcid","value":"PMC12086443"})
get_article_figures(source={"kind":"pmid","value":"40384072"})

適合用在:

  • 要單獨抽 figure caption 與 image URL
  • 想快速找到流程圖、結果圖、顯微圖
  • 需要比全文更結構化的圖像資料

文字探勘

取得 text-mined terms

get_text_mined_terms(source={"kind":"pmcid","value":"PMC7096777"})
get_text_mined_terms(source={"kind":"pmid","value":"12345678"}, semantic_type="CHEMICAL")

常用 semantic_type:

  • GENE_PROTEIN
  • DISEASE
  • CHEMICAL
  • ORGANISM
  • GO_TERM

沒有 open access 時

機構訂閱工作流

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

這一組工具適合:

  • 已知機構有訂閱,但文章不是 OA
  • 想把 PubMed/DOI 轉成圖書館 resolver 連結

建議工作流程

情境 1:從文章直接拿全文

fetch_article_details(pmids="12345678")
get_fulltext(source={"kind":"pmid","value":"12345678"}, sections="abstract,results")

情境 2:搜尋後挑代表性文章讀全文

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

# 對選中的 PMID 再做全文抓取
get_fulltext(source={"kind":"pmid","value":"12345678"}, extended_sources=True)

情境 3:先抓全文,再抽圖表與實體

get_fulltext(source={"kind":"pmcid","value":"PMC7096777"}, include_figures=True)
get_article_figures(source={"kind":"pmcid","value":"PMC7096777"})
get_text_mined_terms(source={"kind":"pmcid","value":"PMC7096777"}, semantic_type="CHEMICAL")

工具選擇規則

只想拿全文或 PDF

先用 get_fulltext

想抓圖表

getarticlefigures

想抽 gene / disease / chemical

gettextmined_terms

文章不是 OA,但你有圖書館帳號

configureinstitutionalaccess + getinstitutionallink


最後原則

全文工作流已經收斂成單一公開入口:先用 get_fulltext,需要圖表或 text mining 時再補用專用工具。