pubmed-multi-source-search
Cross-database search using unified_search across academic sources. Triggers: 跨資料庫, multi-source, Semantic Scholar, OpenAlex, CORE, Europe PMC, 綜合搜尋
Best use case
pubmed-multi-source-search is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Cross-database search using unified_search across academic sources. Triggers: 跨資料庫, multi-source, Semantic Scholar, OpenAlex, CORE, Europe PMC, 綜合搜尋
Teams using pubmed-multi-source-search should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/pubmed-multi-source-search/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How pubmed-multi-source-search Compares
| Feature / Agent | pubmed-multi-source-search | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Cross-database search using unified_search across academic sources. Triggers: 跨資料庫, multi-source, Semantic Scholar, OpenAlex, CORE, Europe PMC, 綜合搜尋
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# 多來源綜合搜尋
## 描述
目前公開的多來源搜尋入口是 `unified_search`。它會自動在 PubMed、OpenAlex、Semantic Scholar、Europe PMC、CrossRef 之間分流,並在單次搜尋內做整合與去重。
## 觸發條件
- 「跨資料庫搜尋」
- 「綜合搜尋」
- 「不要只看 PubMed」
- 提到 OpenAlex、Semantic Scholar、Europe PMC、preprint
- 需要跨來源補足 coverage
---
## 核心原則
> 新 workflow 不再直接依賴多個來源別 MCP 工具。對大多數文獻搜尋情境,應優先使用 `unified_search`,而不是舊的來源別工具名稱。
---
## 核心工具
```python
unified_search(
query="machine learning drug discovery",
sources="pubmed,openalex,semantic_scholar,europe_pmc",
limit=25,
ranking="balanced",
output_format="json"
)
```
### `sources` 可選值
- `pubmed`
- `openalex`
- `semantic_scholar`
- `europe_pmc`
- `crossref`
如果不指定,系統會自動選來源。
---
## 常見用法
### 1. 廣泛覆蓋
```python
unified_search(
query="machine learning drug discovery",
limit=30,
ranking="balanced"
)
```
### 2. 只看生醫核心來源
```python
unified_search(
query="sepsis biomarkers",
sources="pubmed,europe_pmc",
limit=25,
ranking="quality"
)
```
### 3. 納入預印本
```python
unified_search(
query="COVID-19 vaccine efficacy",
sources="pubmed,europe_pmc,openalex",
options="preprints",
limit=30,
ranking="recency"
)
```
### 4. 看高影響力跨領域文獻
```python
unified_search(
query="foundation models pathology",
sources="pubmed,openalex,semantic_scholar",
limit=30,
ranking="impact"
)
```
### 5. 程式化後處理
```python
unified_search(
query="CRISPR gene therapy",
sources="pubmed,openalex,semantic_scholar,europe_pmc",
limit=20,
output_format="json"
)
```
---
## 建議工作流程
### 情境 1:跨來源找完整 coverage
```python
unified_search(
query="remimazolam sedation",
sources="pubmed,europe_pmc,openalex,semantic_scholar",
limit=30,
ranking="balanced"
)
```
### 情境 2:先找文獻,再補全文
```python
unified_search(
query="machine learning radiology",
limit=20,
output_format="json"
)
# 對選中的 PMID / DOI 進一步抓全文
get_fulltext(pmid="12345678", extended_sources=True)
```
### 情境 3:跨來源搜尋後做探索
```python
unified_search(
query="CAR-T lymphoma",
limit=15,
ranking="impact"
)
find_related_articles(pmid="12345678")
find_citing_articles(pmid="12345678")
get_citation_metrics(pmids="12345678,23456789")
```
---
## 何時調整 ranking
- `balanced`: 預設,一般探索
- `impact`: 先找高影響力代表作
- `recency`: 先看最新研究與更新動態
- `quality`: 偏向高證據等級與較可靠研究
---
## 何時使用 options
```python
options="preprints, shallow"
```
### 常見組合
- `preprints`: 想補最新未正式出版研究
- `shallow`: 只想快速掃過
- `no_relax`: 不希望零結果時自動放寬
- `no_analysis`: 程式輸出較乾淨
---
## 不是這個 skill 的情境
- 要做 MeSH / 同義詞完整展開:改用 `pubmed-systematic-search`
- 是臨床比較問題:改用 `pubmed-pico-search`
- 想找圖像:用 `search_biomedical_images`
- 想抓全文:用 `pubmed-fulltext-access`
---
## 最後原則
公開 MCP 介面中的文字文獻搜尋,應以 `unified_search` 為主,不再拆成來源別搜尋工作流。Related Skills
pubmed-search-mcp-harness
Cline harness for PubMed Search MCP. Triggers: pubmed search, literature search, unified_search, pipeline, fulltext, release checklist, Cline.
pubmed-systematic-search
Comprehensive search using generate_search_queries and unified_search. Triggers: 系統性搜尋, 完整搜尋, 文獻回顧, systematic search, comprehensive, MeSH expansion, 同義詞
pubmed-quick-search
Quick literature search using unified_search. Triggers: 搜尋, 找論文, search papers, find articles, PubMed, 文獻搜尋, 快速搜尋
pubmed-pico-search
Agent-guided PICO clinical question search using parse_pico handoff and unified_search. Triggers: PICO, 臨床問題, A比B好嗎, treatment comparison, clinical question, 療效比較
pubmed-paper-exploration
Deep exploration from a key paper using related, citing, reference, and citation tree tools. Triggers: 這篇論文的相關研究, 誰引用這篇, 類似文章, related articles, citation tree, paper exploration
pubmed-mcp-tools-reference
Complete reference for all 46 PubMed Search MCP tools. Triggers: 工具列表, all tools, 完整功能, tool reference, 有哪些工具
pubmed-gene-drug-research
Gene, compound, and variant research using search_gene, search_compound, and search_clinvar. Triggers: 基因, gene, 藥物, drug, compound, PubChem, ClinVar, 變異, variant, 臨床意義
pubmed-fulltext-access
Full text access using get_fulltext, figure extraction, and institutional link tools. Triggers: 全文, fulltext, PDF, open access, 免費下載, PMC, 開放取用
academic-figure-drawing-harness
Codex drawing harness integration. Triggers: 繪圖, draw, figure, chart, plot, mermaid, SVG, Gemini, graph, 生成圖表.
zotero-keeper-harness
Cline harness for Zotero Keeper and the bundled VS Code extension. Triggers: zotero keeper, zotero mcp, full check, release checklist, workflow, vsix, Cline.
llm-wiki-builder
Build or refresh Foam-compatible LLM wikis from Zotero, PubMed, documents, and local Markdown notes using a multi-tool workflow.
asset-aware-mcp-harness
Cline harness for this repo (rules + workflows + checks). Triggers: cline harness, full check, release checklist, workflow, 文檔工作流, DFM, citation-ready.