Search PubMed
from an AI research console.
PubMed indexes 30+ million biomedical citations. The science runtime exposes it as two callable instruments so an AI agent can search the literature and pull abstracts without ever inventing a citation — every paper comes back pinned to its PMID, DOI and URL.
What the instrument does
PubMed is the National Library of Medicine's index of biomedical literature — abstracts, MeSH terms, DOIs and citation metadata across MEDLINE. It is the first place most research questions land.
On the console, pubmed_search runs a relevance- or date-sorted query and returns structured records; pubmed_abstracts fetches the full abstract text by PMID. Both hand back the identifiers a claim can be traced by.
Tools this powers on the console
pubmed_search · pubmed_abstracts
Research use only. Service names belong to their owners; no partnership or endorsement is implied.
How Perslis integrates PubMed
The failure mode this fixes is the fabricated citation — the plausible paper that does not exist. Because Lois can only report what PubMed returned, every reference on the console carries a live PMID that resolves. There is no path for a made-up DOI to enter the record.
PubMed is also the hub of the literature mesh: a PMID that carries a PMCID hands straight to pmc_fulltext for open-access full text, and the same question fans out to OpenAlex and Semantic Scholar so a search is a triangulation, not a single opinion.
Ask
One question at the console — no per-service query language, no tab-hopping.
Route
Lois calls this instrument alongside the others and gathers what returns.
Pin
Every field keeps its source: database, accession, URL. A result that cannot name its source is refused by construction.
Admit
The PEEL gate decides what enters the lab record. Reported values stay verbatim; unsupported claims stay out.
A real, source-pinned result
Pulled live from the service and pinned to its identifier — the same contract every answer on the console is held to.
| Top result | Sickle Cell Disease. |
|---|---|
| Authors | Pecker LH, Lanzkron S |
| Journal | Ann Intern Med, 2021 |
| PMID | 33428443 |
| DOI | 10.7326/AITC202101190 |
| Total matches | 4,519 |
Source: pubmed · PMID 33428443 · retrieved 2026-09-18
Why route PubMed through one console
For AI agents
A PubMed MCP-style tool an LLM can call directly — structured records in, no scraping, no key juggling.
One question, every source
The same query hits PubMed, OpenAlex and Semantic Scholar at once, then cross-references PMCID to full text.
No hallucinated citations
Every reference resolves to a real PMID. A claim that cannot name its source is refused by construction.
Instruments PubMed talks to
A literature question is never one lookup. PubMed hands off to the full-text and scholarly-graph instruments beside it.
PubMed Central
Open-access full text, fetched by PMCID.
pmc_fulltext
OpenAlex
Open scholarly metadata — works, citation counts, any discipline.
openalex_search · openalex_work
Semantic Scholar
Papers plus the forward citation graph — who built on what.
semantic_scholar_search · semantic_scholar_citations
bioRxiv · medRxiv
Preprint feeds by date window; DOI fetch with the published version when one exists.
biorxiv_recent · biorxiv_fetch