pubmed-database · davila7 — Claude Code skill: query PubMed via API
A Claude skill for PubMed access over the E-utilities REST API: MeSH queries, batch fetches, citation matching by PMID or DOI.
npx claude-code-templates@latest --skill scientific/pubmed-databaseWhat pubmed-database does
The skill teaches the agent to talk to PubMed through NCBI's E-utilities. Five endpoints are covered: ESearch returns PMIDs for a query, EFetch pulls full records, ESummary gives short document info, EPost uploads identifiers for batch work, and ELink finds related articles. Queries combine Boolean operators, field tags and MeSH terms, subheadings included.
For Python code the author points to biopython (Bio.Entrez) instead, leaving this skill for raw HTTP calls and custom API clients. Rate limits matter here: 3 requests per second without an NCBI key, 10 with one. Every request should carry a User-Agent header.
Two reference files ship alongside: search_syntax.md for field tags and operators, api_reference.md for endpoint parameters. Typical work is systematic reviews, tracking new publications, or resolving a citation from journal, year, volume and page.
What it helps with
- pull a set of articles on a topic filtered by publication type
- resolve a PMID from a partial citation
- fetch abstracts for many records in one batch
- build a query with MeSH terms and subheadings
- set up recurring monitoring of an author's new papers
Example prompts for the agent
- Find systematic reviews on type 2 diabetes from 2023–2024
- Fetch the abstracts for these PMIDs with EFetch
- Build a query using a MeSH term and the /drug therapy subheading
How to install pubmed-database
1claude-code-templates
npx claude-code-templates@latest --skill scientific/pubmed-database2Skills CLI (Vercel)
npx skills add davila7/claude-code-templates --skill pubmed-database -a claude-code3Manually: put the files in place
~/.claude/skills/pubmed-database/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/pubmed-database/
Manually: put the files in place
.agents/skills/pubmed-database/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/pubmed-database/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install davila7/claude-code-templates pubmed-database2Manually: put the files in place
.github/skills/pubmed-database/SKILL.mdPlatform CLI
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/scientific/pubmed-databaseManually: put the files in place
.devin/skills/pubmed-database/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about pubmed-database
How do I install pubmed-database in Claude Code?
Run npx claude-code-templates@latest --skill scientific/pubmed-database. Or use npx skills add davila7/claude-code-templates --skill pubmed-database -a claude-code.
What does pubmed-database do?
It helps the agent compose PubMed searches and call the E-utilities API: find articles, download records, match citations by identifier.
Do I need an API key for PubMed?
NCBI allows 3 requests per second without a key and 10 with one. The key is optional, but bulk fetches hit the limit fast without it.
Similar skills
All skills for Science and research →pubmed-database by other authors comes first.