Medium Article Discovery & Metadata Extractor

Medium is full of useful writing, but browsing it by hand makes it hard to compare coverage across topics, authors and publications. This actor turns five common discovery paths — tag browsing, author feeds, publication feeds, keyword search and direct article URLs — into one consistent article dataset you can sort, filter and join with your own tools.

Every record describes a single article: who wrote it, where it was published, when it went live, how long it takes to read, how much engagement it has, whether it sits behind Medium's member paywall, and which topics and tags it carries.

Discovery modes

Mode	What it collects	Best for
byTag	Articles tagged with a topic, newest first or by top stories from the week/month/year	Trend tracking and editorial research
byUser	An author's articles, including posts distributed to publications	Monitoring a writer's output
byPublication	Articles published under a publication	Following a publication's latest work
search	Articles matching a keyword query	Broad topic discovery
byUrl	Metadata for one or more article URLs	Enriching a list of links you already have

Tag, author, publication and search modes walk the source's pagination until maxItems is reached or the source runs out, so a run can return thousands of records rather than a single page.

Output fields

Field	Description
articleId	Medium's short article identifier
title	Article headline
subtitle	Article subtitle or standfirst, when present
previewText	Subtitle or search-style preview text, capped at 300 characters
authorName	Author display name
authorUsername	Medium handle without @
authorBio	Author bio, when the profile has one
authorImageUrl	Author avatar URL
authorUrl	Author profile URL
publicationName	Publication name, when the article belongs to one
publicationSlug	Publication slug
publicationDomain	Custom publication domain, when one is used
publicationUrl	Publication home URL
publishedDate	First publish time, ISO-8601 UTC
updatedDate	Latest update time, when it differs from publishedDate
tags	Article tag names
topics	Medium topic names attached to the article
readingTimeMinutes	Estimated reading time in whole minutes
wordCount	Article word count
language	Detected language code
clapCount	Clap count reported by Medium
responseCount	Number of responses on the article
isPaywalled	true when the article is member-only
isSeries	Present and true when the article belongs to a series
articleUrl	Canonical article URL from Medium
canonicalUrl	Canonical URL when it differs from articleUrl
imageUrl	Article preview image URL
recordType	Always article
siteName	Always Medium
scrapedAt	When the record was collected, ISO-8601 UTC

Empty fields are omitted rather than written as null. A field appears only when Medium exposes that value for the article.

Input fields

Field	Type	Default	Used by
mode	select	byTag	All modes
tag	select	programming	byTag
customTag	text	–	byTag (overrides tag)
sortBy	select	new	byTag
username	text	–	byUser
publication	text	–	byPublication
query	text	artificial intelligence	search
articleUrl	text	–	byUrl
fromDate	text	–	All modes
containsKeyword	text	–	All modes
excludePaywalled	boolean	false	All modes
maxItems	integer	5	All modes

Example: newest Python articles

{
  "mode": "byTag",
  "tag": "python",
  "sortBy": "new",
  "maxItems": 25
}

Example: an author's recent work

{
  "mode": "byUser",
  "username": "towardsdatascience",
  "maxItems": 50
}

Example: a publication's archive

{
  "mode": "byPublication",
  "publication": "better-programming",
  "maxItems": 100
}

Example: keyword search with a date cut-off

{
  "mode": "search",
  "query": "large language models",
  "fromDate": "2025-01-01",
  "excludePaywalled": true,
  "maxItems": 40
}

Example: enrich links you already have

{
  "mode": "byUrl",
  "articleUrl": "https://medium.com/@author/example-article-abc123def456\nhttps://medium.com/p/abcdef123456",
  "maxItems": 10
}

What you can do with it

• Build a topic monitoring sheet that shows which authors and publications are covering a subject each week.
• Compare top and newest stories for a tag to separate evergreen hits from fresh coverage.
• Filter out member-only posts before sending articles to a newsletter or internal digest.
• Enrich a list of saved links with author, publication, date, engagement and topic metadata.
• Track an author's or publication's output over time with scheduled runs.

FAQ

Do I need a Medium account or cookies? No. The actor reads publicly available article metadata without logging in.

Do I need to configure a proxy? No. The actor picks a working connection automatically, so you never supply proxy settings, cookies or an API key.

Can I get the full article text? This actor focuses on discovery and metadata. It returns titles, previews, author and publication details and engagement metrics, not the article body.

How far back can it go? It follows the source's own pagination until maxItems is reached or the source runs out. Very large requests can span many pages, so raise maxItems deliberately.

Why are some fields missing on some records? Medium does not expose every field for every article. A personal post may have no publication, an older post may have no subtitle, and some articles have no tags. Missing values are omitted instead of filled with placeholders.

How are paywalled articles handled? isPaywalled marks member-only articles, and excludePaywalled removes them. Only public metadata is collected; the actor never tries to read member-only body content.

What counts as an article URL? Standard Medium links, medium.com/p/... short links, and articles hosted on a custom Medium domain such as a publication's own website.

Can I combine filters? Yes. fromDate, containsKeyword and excludePaywalled work together in every mode.

Is this affiliated with Medium? No. This is an independent third-party actor that reads publicly available article metadata. It is not operated by or affiliated with Medium.