PubMed Article Monitor
Pricing
Pay per usage
PubMed Article Monitor
Monitor new PubMed articles by query, publication type and date via the official NCBI E-utilities (esearch, esummary, efetch). Incremental 'new since last run' monitoring, a markdown digest and an optional webhook. No API key required at 3 requests/second, no scraping.
Pricing
Pay per usage
Rating
0.0
(0)
Developer
Artemi Kupolov
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
2 days ago
Last modified
Categories
Features
Share
Monitor new scientific articles in PubMed by keyword, MeSH term, author, journal or any other PubMed query, through the official NCBI E-utilities. No HTML scraping, no API key required at three requests per second, no bypass of any protection.
Use it as a one-off search or as a scheduled watch that only returns new articles since the last run.
Sources (official):
esearch.fcgi— find PMIDsesummary.fcgi— document summariesefetch.fcgi— full Medline records (title, authors, journal, date, DOI, abstract, MeSH)
Every request sends the tool and email parameters NCBI asks for, and the Actor stays at or below 3 requests per second without a key (or 10 r/s if you provide your own NCBI API key).
Why this Actor
PubMed is the standard bibliographic database for biomedical and life-science literature. This Actor turns a PubMed search into a clean, Google-Sheets-ready dataset:
- Familiar query language — any PubMed term, including field tags such as
[Title],[Author],[Journal]or[MeSH Terms]. - Publication type and date window — filter to journal articles, reviews, trials, preprints and more; default window is the last 7 days on the Entrez entry date (
edat), which is what you want for a "what's new" monitor. - Incremental monitoring —
sinceLastRun+onlyNewstore seen PMIDs and the last date in a key-value store, so a daily schedule only costs you the new rows. - Flat, useful rows — title, authors, journal, date, DOI, PMID, abstract, MeSH terms and the official PubMed link.
- Digest and webhook — an email-friendly markdown digest is saved to the key-value store; optionally POSTed to a URL you provide. The Actor never sends email itself.
Use cases
- Topic watch — schedule a daily run for
mRNA vaccine,CRISPR[Title]or a MeSH term, and forward the digest to Slack or Make viawebhookUrl. - Author or journal alerts — follow a lab (
Smith J[Author]) or a journal (Nature[Journal]) without leaving PubMed's official API. - Evidence reviews — pull 30 or 90 days of systematic reviews and meta-analyses into a spreadsheet, with abstracts and MeSH already filled in.
- RAG / research pipelines — drop new papers into a dataset on a schedule, with DOI and abstract ready for downstream tools.
Input highlights
| Field | What it does |
|---|---|
query | PubMed search term (mRNA vaccine, CRISPR[Title], Smith J[Author]) |
publicationTypes | Journal Article, Review, Systematic Review, Meta-Analysis, Clinical Trial, RCT, Preprint, … |
dateType | edat (entry date, default), pdat (publication date), mhdat, mdat |
datePreset | last_24_hours, last_7_days (default), last_30_days, last_90_days, ytd, custom, since_last_run |
sinceLastRun + onlyNew + monitorKey | Incremental watch; use a distinct monitorKey per schedule |
webhookUrl | HTTPS endpoint that receives the digest of new articles |
maxItems | 1–10 000, default 100 |
tool / email | Official E-utilities identification (defaults are filled in) |
ncbiApiKey | Optional; only needed above 3 requests per second |
The default date window is the last 7 days, so a run never asks PubMed for the entire archive.
Example: new mRNA-vaccine papers this week
{"query": "mRNA vaccine","publicationTypes": ["Journal Article", "Review"],"datePreset": "last_7_days","dateType": "edat","sinceLastRun": false,"maxItems": 50,"includeDigest": true,"monitorKey": "mrna-vaccine"}
Example: incremental daily watch
{"query": "CRISPR[Title]","publicationTypes": ["Journal Article"],"datePreset": "since_last_run","dateType": "edat","sinceLastRun": true,"onlyNew": true,"maxItems": 200,"webhookUrl": "https://example.com/hooks/pubmed","monitorKey": "crispr-title-daily"}
Output
Each dataset item is one article. The most useful fields:
| Field | Contents |
|---|---|
pmid | PubMed identifier |
title | Article title |
authors / authorsText | Author list, and the same names joined with ; |
journal / journalAbbrev | Full journal title and ISO abbreviation |
publicationDate | Best available date (YYYY-MM-DD when the day is known) |
doi / doiUrl | Digital Object Identifier and https://doi.org/… |
abstract | Abstract, with labelled sections kept (BACKGROUND: …) |
meshTerms / meshHeadings | MeSH descriptors, and descriptor/qualifier pairs |
publicationTypes | e.g. Journal Article, Review |
link / pubmedUrl | Official record, https://pubmed.ncbi.nlm.nih.gov/{pmid}/ |
pmcid / pmcUrl | PubMed Central id and link, when the article is in PMC |
isNew | true if this PMID was not in the monitor's seen list |
A markdown digest of new articles is written to the key-value store as digest.md.
Pricing and fair use
This Actor is published free. A later pay-per-event price of $3 per 1,000 articles is documented only as a proposal and is not enabled.
The Actor uses the official NCBI E-utilities. Please keep scheduled runs reasonable (a daily or hourly watch on a focused query is fine; do not harvest the entire database). Large jobs should follow NCBI's usage guidelines: identify yourself with tool and email, stay within the rate limit, and prefer evenings and weekends in US Eastern time for very large downloads.
Changelog
- 0.1 — First public version. esearch + efetch, filters, onlyNew / sinceLastRun, digest, webhook.