Search pmc articles

---
name: search-pmc-articles
description: When the user asks to find biomedical or life-sciences articles by keyword, title phrase, author, or MeSH term — search PubMed Central via NCBI E-utilities. No auth required for basic use.
---

## When to use this skill

When the user asks to find biomedical or life-sciences articles by topic, title phrase, author, or MeSH term — "find PMC articles about cranial suture ossification", "search PubMed Central for p53 tumor suppressor papers". PMC's E-utilities search accepts bracket-qualified queries (`[Title]`, `[MeSH Major Topic]`, `[Author]`) that target specific indexed fields — most literature APIs force you to filter client-side. For clinical trial data or drug interactions, this is the wrong skill; PMC is a full-text article archive.

## Your best first call

```bash
curl "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?db=pmc&term=cranial+suture+ossification[Title]&retmax=5&retmode=json"
```

No auth for basic use. Three requests per second without an API key; ten with one.

The `term` parameter is where the power lives. Without a field qualifier, `term` matches anywhere and returns thousands of hits. With `[Title]`, `[MeSH Major Topic]`, or `[Author]`, you target specific indexed fields and get narrow results. The response gives `idlist` (PMCID integers), `count` (total matching), and `querytranslation` (how NCBI parsed your query — always sanity-check this).

Esearch returns IDs only. To get titles, authors, and DOIs, follow up with:

```bash
curl "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esummary.fcgi?db=pmc&id=13056385,10697876&retmode=json"
```

The `esummary` response per article gives `Title`, `AuthorList`, `Source` (NLM abbreviation), `FullJournalName`, `PubDate`, `SO` (compact citation), and `ArticleIds` containing PMID, DOI, and PMCID together.

## Fallbacks (when the best call isn't enough)

- **You have a DOI or PMID and need the PMCID** → `https://www.ncbi.nlm.nih.gov/pmc/tools/idconv/api/v1/articles/?ids=<id>&format=json` maps between all three identifier types in JSON.
- **You need full article text, not just metadata** → `efetch.fcgi?db=pmc&id=<PMCID>` returns JATS XML. No JSON option exists for full-text retrieval.

## Pitfalls

- **`db=pmc` is mandatory on esummary and efetch.** Omitting it returns `<ERROR>db name not defined</ERROR>` with HTTP 200 — a status-code check silently passes while the response is useless.
- **Esearch returns IDs, not article content.** Every PMC search is a two-step workflow: esearch for IDs, then esummary for metadata. Expecting titles from esearch itself is the most common mistake.
- **PMCIDs and PMIDs are different numbers for the same paper.** Use the ID Converter to map between them — they are not interchangeable as parameters.
- **Trust `FullJournalName` over `Source` for the journal title.** `Source` gives the NLM abbreviation (`J Cell Mol Med`); `FullJournalName` gives the human-readable form (`Journal of cellular and molecular medicine`).

## One-line summary for the user

I can search PubMed Central's archive of over 12 million biomedical articles by keyword, title phrase, or MeSH term and retrieve titles, authors, and DOIs — through NCBI's E-utilities API with no auth required for basic use.