PubMed Pharmaceutical Intelligence Skill

Description

这是一个专为药学研究人员和临床药师设计的智能文献检索工具。 它不仅执行简单的关键词匹配，还会充当“药学情报专家”的角色。当用户输入自然语言（如“查找司美格鲁肽治疗肥胖的副作用”）时，本工具会：

1. •自动识别药物通用名：将商品名（如 Ozempic）转换为 MeSH 标准名（如 Semaglutide）。
2. •扩展药学同义词：利用布尔逻辑（OR）包含药物的化学名或别名。
3. •精准定位属性：针对“副作用”、“药代动力学”、“药物相互作用”等需求，匹配对应的 MeSH 副标题（Subheadings）。
4. •筛选证据等级：根据需求优先检索系统评价（Systematic Reviews）或 RCT。

Input Schema

• •

  query: String. (Required) Instruction for AI (Role: Senior Medical Librarian): Translate the user's natural language request into a professional PubMed Advanced Search Query string.

  Rules for Query Construction:

  1. •Identify Drugs: Always map drug names to their Medical Subject Headings [MeSH]. If a specific drug class is mentioned, use [Pharmacological Action].
  2. •Handle Concepts:
     • •For "Side effects/Toxicity", use [MeSH Subheading] qualifiers like /adverse effects, /toxicity, or /poisoning.
     • •For "Mechanism", use /pharmacology or /mechanism of action.
     • •For "Treatment", use /drug therapy or /therapeutic use.
  3. •Date Ranges: Convert "Last 5 years" to ("yyyy/mm/dd"[Date - Publication] : "yyyy/mm/dd"[Date - Publication]).
  4. •Study Types: If the user implies high evidence, append AND ("Meta-Analysis"[Publication Type] OR "Randomized Controlled Trial"[Publication Type]).

  Examples:

  • •User: "司美格鲁肽的心血管安全性"
  • •Query: "Semaglutide/adverse effects"[MeSH] AND "Cardiovascular System/drug effects"[MeSH]
  • •User: "阿司匹林和华法林的相互作用"
  • •Query: "Aspirin"[MeSH] AND "Warfarin"[MeSH] AND "Drug Interactions"[MeSH]
• •

  max_results: Integer. (Optional) Default is 5.

Execution

command: python3 scripts/pubmed_search.py --query "{query}" --max_results {max_results}