An AI agent skill to search and retrieve biomedical literature and scientific papers from PubMed and other NCBI databases using the E-Utilities API.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install search-pubmed
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install search-pubmed using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is PubMed Literature Search?
The PubMed Literature Search skill is a powerful extension designed for AI coding agents to interact directly with the National Center for Biotechnology Information (NCBI) databases. By wrapping Biopython's Bio.Entrez module, this skill allows agents to programmatically query PubMed, PMC, and nucleotide databases, bringing peer-reviewed scientific literature, genomic insights, and medical publications straight into the workspace.
Integrating this tool with Openclaw Skills empowers your AI agent to independently perform literature reviews, look up specific PubMed IDs (PMIDs), resolve scientific queries, and explore biological research gaps. Whether you are analyzing gene functions, investigating drug interactions, or researching clinical papers, this skill acts as a bridge between LLMs and authoritative biomedical data sources.
PubMed Literature Search Use Cases
- Scientific Literature Retrieval: Quickly find papers, clinical trials, and biological reviews using advanced keyword queries and boolean logic.
- Direct PMID Lookup: Fetch citation metadata, authors, DOI links, and abstract details directly using individual or multiple PMIDs.
- Research Gap Identification: Systematically search the literature to discover unstudied areas or confirm the absence of published research on specific cross-topics.
- Cross-Database Exploration: Programmatically switch query contexts to search other NCBI databases, such as Gene, PMC, or Nucleotide, for comprehensive life-science analysis.
How PubMed Literature Search Works
- Triggering: The AI agent detects a literature search prompt (e.g., 'find papers on CRISPR' or 'lookup PMID 41185614') and invokes the bundled Python script.
- Query Formulation: The underlying script uses Entrez syntax to query NCBI E-Utilities APIs via the Biopython library.
- Result Retrieval: The script processes NCBI response packages, mapping fields such as title, author names, journal source, publication date, and DOIs.
- Structured Output: The agent receives a clean, structured text output, making it easy to parse, present, or cite within subsequent development or research workflows.
PubMed Literature Search Setup
To run this skill, ensure you have Python 3.6+ and the required library installed:
pip install biopython
Once installed within the Openclaw Skills directory structure, you can invoke the search script from your workspace root. Here is the standard execution command:
python .claude/skills/search-pubmed/scripts/search_pubmed.py "your query here"
Quick Start Examples
Search biomedical terms:
python .claude/skills/search-pubmed/scripts/search_pubmed.py "immunotherapy and oncology"
Fetch metadata and abstract for specific PMIDs:
python .claude/skills/search-pubmed/scripts/search_pubmed.py --pmid 41185614 --full
PubMed Literature Search Data Schema & Taxonomy
The script queries NCBI Entrez databases and outputs consistent, structured metadata for each record returned.
Key Metadata Fields
| Field | Description | Format |
|---|---|---|
| PMID | Clickable identifier linking to PubMed | https://pubmed.ncbi.nlm.nih.gov/[PMID] |
| Title | Full title of the scientific article | Text string |
| Author | Author representation listing first and last author | `First Author |
| Source | Journal name or publication abbreviation | Text string |
| Date | Date the article was published | YYYY-MM-DD or publication year |
| DOI | Digital Object Identifier for easy citation tracking | String format (if available) |
Bundled Skill Structure
scripts/search_pubmed.py: The executable python script managing connections and queries.references/entrez-help.md: Detailed reference manual outlining advanced NCBI Entrez syntax.
PubMed Literature Search Advanced Features
- Boolean & Field-Qualified Queries: Supports complex logic operators (
AND,OR,NOT) as well as field tags (e.g.,[Title/Abstract],[Author],[Date - Publication]) for granular search control. - Multi-DB Support: Easily pivot searches across databases beyond
pubmed, targeting NCBI databases likepmc(PubMed Central) andnucleotidevia the--dbparameter. - API Key Optimization: Includes an
--api-keyparameter that automatically elevates the NCBI rate limit from 3 requests/sec up to 10 requests/sec, ideal for heavy programmatic usage. - Openclaw Skills Integration: Native design allows seamless integration with other Openclaw Skills and multi-agent systems searching for genomic, proteomic, and literary correlations.
SKILL.md
Loading
Related Openclaw Skills
AI-Generated Code Format Review
A dedicated Java formatting review skill designed to automatically audit and correct naming, comments, structure, and style discrepancies in AI-generated code.
AgentNOMOS Governance Preflight
A read-only governance preflight tool that evaluates proposed AI agent actions for authorization, boundaries, risks, and required evidence before execution.
PrepKit: Informatics Olympiad Lesson Preparation Framework
PrepKit is an advanced instructional framework based on the ADDIE model, specifically tailored for AI agents to design and develop elite training materials for informatics competitions like NOI, CSP, and GESP.
agent-token-usage
A utility to accurately calculate, aggregate, and display cumulative LLM token usage across all OpenClaw agents, featuring a Python CLI and an optional Control UI analytics button.
China Company Search via WenDaoYun
An Openclaw Skills tool interfacing with the WenDaoYun API to retrieve comprehensive corporate data, financial status, and risk metrics for Chinese enterprises.
AI Bookmark and Desktop Dock Assistant (Chinese)
The AI Bookmark and Desktop Dock Assistant turns prompts, local directories, browser bookmarks, and application shortcuts into highly searchable and actionable instruction cards.
Featured*
