Medium Scraper - Articles, Claps, Comments, Authors & Full Text avatar

Medium Scraper - Articles, Claps, Comments, Authors & Full Text

Pricing

from $0.18 / 1,000 stories

Go to Apify Store
Medium Scraper - Articles, Claps, Comments, Authors & Full Text

Medium Scraper - Articles, Claps, Comments, Authors & Full Text

Medium stories by tag (whole archive or Top lists), search, writer, publication, list or link: claps, clappers, responses, reading time, tags, language, member-only flag, writer and publication stats; full text, responses, writer and tag rows. Monitor mode. $0.18 per 1,000 stories, no start fee.

Pricing

from $0.18 / 1,000 stories

Rating

0.0

(0)

Developer

Top Api

Top Api

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

7 days ago

Last modified

Features

Limited permissions

Share

What does Medium Scraper do?

Medium Scraper extracts Medium stories from medium.com by tag, keyword search, writer, publication, reading list or link, and turns them into structured rows: one row per story with its title, link, dates, reading time, word count, claps, distinct clappers, responses, tags, language, member-only flag, and the stats of its writer and publication. Switch on the full text (plain text, Markdown, HTML), responses and replies, or a profile row per writer, publication, tag and list. Built for content marketers, newsletter curators, AI and RAG teams, and anyone researching writers, topics and engagement on Medium without copying pages by hand.

Medium Scraper · Medium API alternative · Medium articles scraper · Medium tag scraper · Medium author scraper · Medium comments scraper

  • Every story of a tag, not ten: the tag archive is read newest first back to the tag's first story (or oldest first, or Medium's most-read ranking month by month), plus Medium's own Hot, New, Top week / month / year / all-time and Recommended lists
  • Engagement on every row: claps, distinct clappers, claps per clapper, responses, claps per 1,000 writer followers, and (optional) this story's claps against the writer's own median
  • Full text at no extra charge: free stories come complete as text, Markdown (code blocks keep their language) and HTML; member-only stories give the public preview, flagged contentIsPreview
  • Search past Medium's 1,000-result limit: a big search is repeated year by year without duplicates (2,000 stories from one query in a measured run)
  • No login, cookies or Medium API key: it reads what Medium shows any signed-out visitor
  • Pay only for delivered rows: no start fee; duplicates, filtered stories and stories a monitor already delivered cost nothing
  • Export anywhere: JSON, CSV, Excel, XML, RSS; schedules, webhooks, Make, Zapier, n8n, Google Sheets, Slack
  • Works with AI agents: call it through the Apify API or the MCP server

What data can you get from Medium?

📰 Title and subtitle🔗 Link and canonical link (`canonicalUrl`)📅 Published and updated dates
👏 Claps (`clapCount`)🙋 Distinct clappers (`voterCount`)💬 Responses (`responsesCount`)
⏱️ Reading time and word count🏷️ Tags, tag names, Medium topics🔒 Member-only flag (`isLocked`)
🌐 Language🖼️ Preview image, its alt text and size📜 License
✍️ Writer: name, username, bio, avatar👥 Writer followers and following⭐ Member, Partner Program, book author, X handle
🏢 Publication: name, link, own domain, followers📈 Claps per 1,000 followers, claps per clapper🆕 Fresh flag (under 48 hours)
📝 Full text: text, Markdown, HTML🧭 Table of contents, images, outbound links💻 Code-block and paragraph counts
💬 Responses and replies as rows🧑 Writer rows with posting activity and medians📚 Publication, tag and list rows

On a 1,000-story run of the newest artificial-intelligence stories, every row had a title, link, dates, reading time, word count, claps, clappers, responses, language and tags; 98% had a subtitle, 90% a preview image, 86% Medium topics (Medium assigns them a few days after publication), 84% a writer bio, 25% a publication and 17% a writer X handle. 31% of those stories were member-only.

How to scrape Medium with Medium Scraper

  1. Create a free Apify account: $5 of monthly credit, no card needed.
  2. Open Medium Scraper and click Try for free.
  3. Enter Tags, Search terms, Writers, Publications, or paste Medium links, and set Max stories.
  4. Switch on Include full text, Include responses or the profile rows if you need them.
  5. Click Start. Rows stream into the dataset as each page is read.
  6. Open the Storage tab and Export as JSON, CSV or Excel, or let a schedule run it for you.

⬇️ Input

ParameterTypeDescription
🔍 What to scrapeAny mix; at least one source (an empty input reads 25 newest programming stories as a sample)
tagsListTag slugs, names or links: artificial-intelligence, Machine Learning, https://medium.com/tag/python
tagSortSelectnewest (whole archive, default), oldest, most-read, hot, new, top-week, top-month, top-year, top-all-time, recommended, all-top (all ranked lists merged)
searchQueriesListMedium search, one query per line
searchTypeSelectposts (default), people, publications, tags, lists
authorsList@username, username, https://medium.com/@ev, https://name.medium.com
publicationsListbetter-programming, https://medium.com/swlh, levelup.gitconnected.com
startUrlsListAny Medium link: story (also on a writer subdomain or a publication domain), tag page (/tag/python/top/week), profile, publication, search page, list, RSS feed link; story IDs work too
urlsFromFileTextLink of a .txt / .csv file or a shared Google Sheet listing Medium links, story IDs or @handles
🎯 Filters (before any charge)
publishedAfter / publishedBeforeText2025-01-01, an ISO time, or relative: 30 days, 6 months, 24 hours
minClaps / minResponsesIntegerLeave out stories under these counts. Default 0
languageTextISO codes, comma-separated: en or en, es, pt
memberOnlySelectall (default), free, memberOnly
keywords / excludeKeywordsListWords a story's title, subtitle, tags (or text) must / must not contain
➕ Extra data
includeContentBooleanFull text as plain text and Markdown, images, links, table of contents. No extra charge
includeHtmlBooleanThe body as HTML too. No extra charge
includeImagesInMarkdownBooleanKeep images in the Markdown. Default true
includeResponsesBooleanResponses of every delivered story as rows, charged as response
maxResponsesPerPostIntegerDefault 20; 0 = all
includeRepliesBooleanFollow the reply threads too (depth 2, 3 ...)
includeProfilesBooleanA profile row per tag, writer, publication and list you listed, charged as profile
includeArticlesBooleanOff = only those profile rows. Default true
includeAuthors / includePublicationsBooleanA profile row per writer / publication of the delivered stories
maxProfilesIntegerCap on those extra profile rows. Default 100
compareWithAuthorBooleanAdds authorMedianClaps and vsAuthorMedian to every story. No extra charge
🔁 Monitoring
monitorModeBooleanReturn only stories not delivered by earlier runs of the same setup
monitorKey / resetMonitorText / BooleanName a shared memory; start it over
📦 Output
maxItemsIntegerDefault 100; 0 = no limit
maxPostsPerSourceIntegerCap per tag, writer, publication, search or list. Default 0 (none)
dedupeBooleanA story found by several sources is delivered once. Default true
⚙️ Advanced
maxConcurrency, requestTimeoutSecs, proxyConfigurationSources in parallel (default 8), request timeout (default 30 s), an optional proxy (none needed)

Inputs named like other Medium actors' (articleUrls, postUrls, pageAndProfileUrls, searchTerms, authorProfiles, maxResults, maxArticles, limit, tagFeedMode, includeText, includeFullText, onlyNewPosts, skipMemberOnly, publishedWithinDays ...) are accepted too, so an API caller can switch without rewriting its input.

Input JSON

{
"tags": ["artificial-intelligence", "startup"],
"tagSort": "newest",
"publishedAfter": "7 days",
"minClaps": 10,
"includeContent": true,
"maxPostsPerSource": 200,
"maxItems": 400
}

Pick the tag list before anything else

newest walks a tag's complete archive (the programming tag holds 463,000 stories), so a big Max stories keeps going back in time. The ranked lists hold about 100 stories each (Recommended 300+). Do: "tagSort": "top-month" for what readers clapped this month · Don't: expect 1,000 rows from one ranked list (it ends at ~100; use all-top or several tags).

Use a date window instead of reading everything

publishedBefore starts a tag, writer or publication at that date, and publishedAfter stops a newest-first tag or writer there, so {"tags": ["javascript"], "publishedAfter": "2021-03-01", "publishedBefore": "2021-03-31"} reads only March 2021. With tagSort: "most-read" a window is read month by month (about 20 stories a month).

Member-only stories are previews

Every field of a member-only story is filled except the body, which is the public preview a signed-out reader sees (contentIsPreview: true; wordCount is still the full story's). Set memberOnly: "free" when you need complete text.

⬆️ Output

Every story becomes one item with type: "post"; responses are type: "response", profile rows author, publication, tag or list. The Dataset tab has a view for each (Stories, Full text, Responses, Writers, Publications, Tags and lists). A real story row from a 1,000-row run (abbreviated):

{
"type": "post",
"id": "8a855a04c955",
"url": "https://briefs.aiadvances.org/how-to-run-microsofts-4b-computer-use-agent-without-giving-it-your-machine-8a855a04c955",
"title": "How to Run Microsoft’s 4B Computer-Use Agent Without Giving It Your Machine",
"publishedAt": "2026-11-06T18:20:11.730Z",
"readingTimeMinutes": 8.85,
"wordCount": 2199,
"clapCount": 100,
"voterCount": 2,
"responsesCount": 0,
"isLocked": true,
"language": "en",
"tags": ["programming", "artificial-intelligence", "machine-learning", "cybersecurity", "ai-agent"],
"authorName": "MohamedAbdelmenem",
"authorFollowers": 3216,
"authorIsPartnerProgram": true,
"publicationName": "AI Advances Briefs",
"publicationDomain": "briefs.aiadvances.org",
"clapsPer1kFollowers": 31.09,
"source": "tag:artificial-intelligence",
"rank": 66
}

A response row:

{
"type": "response",
"id": "fc4d7dc89d5b",
"text": "The “none” affiliation includes Jitendra Malik and Michael Jordan xD",
"publishedAt": "2017-05-26T00:09:41.924Z",
"clapCount": 0,
"repliesCount": 0,
"depth": 1,
"rootPostUrl": "https://karpathy.medium.com/icml-accepted-papers-institution-stats-bad8d2943f5d",
"authorName": "David Krueger",
"authorFollowers": 2
}

A writer row (with includeProfiles, abbreviated):

{
"type": "author",
"name": "Andrej Karpathy",
"username": "karpathy",
"followerCount": 62184,
"followingCount": 183,
"twitter": "karpathy",
"customDomain": "karpathy.medium.com",
"lastPostDate": "2017-11-11T22:18:53.751Z",
"medianClaps": 1800.5,
"medianResponses": 14,
"topPostTitle": "Software 2.0",
"topPostClaps": 61073,
"topTags": ["machine-learning", "artificial-intelligence", "deep-learning", "programming", "software-development"]
}

Fields

FieldWhat it isFill (1,000-story run)
title, url, publishedAt, updatedAtStory title, link, first and latest publication100%
readingTimeMinutes, wordCountMedium's reading time and word count100%
clapCount, voterCount, responsesCountClaps, distinct clappers, responses100% (31% had at least one clap)
clapsPerVoter, clapsPer1kFollowersDerived engagement (null without clappers / followers)31% / 85%
isLocked, visibility, language, licenseMember-only flag, visibility, detected language, license100%
tags, tagNames, topicsTags and Medium topics100% / 100% / 86%
subtitle, previewImageUrlSubtitle, preview image98% / 90%
authorName, authorUsername, authorFollowers, authorFollowingWriter100%
authorBio, authorTwitterWriter bio and X handle84% / 17%
authorIsMember, authorIsPartnerProgram, authorIsBookAuthorWriter badges100%
publicationName, publicationUrl, publicationFollowersPublication (stories outside one have none)25%
contentText, contentMarkdown, contentHtml, tableOfContentsBody (with full text on)100% of stories (60% of them member-only previews in two Top-list runs)
source, sourceType, sourceInput, rankWhere the row came from and its position there100%

The dataset schema documents all 118 fields of all six row types.

How can I use Medium data?

  • 🎯 Content marketers: pull the Top-month stories of your topics with claps and responses to see which angles, titles and lengths work
  • 📨 Newsletter curators: schedule publishedAfter: "1 day" on a few tags and get each day's new stories with their text
  • 🤝 Creator and outreach teams: search people for a topic and get writer rows with followers, bio, X and Mastodon handles, Partner Program status and posting activity
  • 🤖 AI and RAG teams: collect clean Markdown with code blocks and their language, tags and canonical links; keep memberOnly: "free" for complete texts
  • 🏢 Competitor and blog monitoring: watch a publication or an engineering blog on its own domain with monitor mode and receive only the new stories
  • 📈 Analysts: measure a topic's output and engagement over time from the tag archive and its date windows

How much does it cost to scrape Medium?

Medium Scraper is pay-per-event, with no start fee:

RowPrice
Story (post)$0.18 per 1,000
Response or reply (response)$1.30 per 1,000
Writer, publication, tag or list row (profile)$1.30 per 1,000
StoriesCost
10$0.0018
100$0.018
1,000$0.18
10,000$1.80

Full text, HTML, the writer comparison and the activity metrics cost nothing extra. Duplicates, filtered stories and stories a monitor delivered before are never charged. Apify platform usage is included. Measured on the default 256 MB: 1,000 stories in 51-52 s, 5,000 in 258-300 s, 2,000 search results in 48 s.

Run it on a schedule

  • Schedules: save your input as a task, open Schedules and run it hourly or daily. With monitorMode: true each run returns only stories that earlier runs did not deliver; newest-first tags and writers stop reading at the newest story already delivered, and a poll that finds nothing writes nothing and charges nothing.
  • Example: {"tags": ["product-management"], "authors": ["@lenny"], "monitorMode": true, "monitorKey": "pm-daily", "includeContent": true} on a daily schedule.
  • Webhooks: fire a URL when a run finishes and push the rows to your CRM, Slack or a sheet.
  • No-code: Make, Zapier, n8n, Google Sheets, Airtable, Slack through Apify integrations.
  • API: call it from Node.js or Python with apify-client; code is on the API tab and in the Apify API reference.
  • MCP / AI agents: connect Claude, Cursor or any MCP client through the MCP tab and ask for Medium stories in plain language.

More scrapers from TopApi

📨 Substack Scraper📰 WeChat Articles Scraper📄 Website Content Crawler
🐦 X Profile Scraper🦋 Bluesky Scraper🧭 All TopApi actors

Limits & empty runs

  • Ranked lists are short: Hot, New and each Top list end at about 100 stories, Recommended at about 300, Most read at about 20 a month; the newest-first archive has no such limit.
  • Search: Medium answers up to 1,000 results per query; more are read year by year (newest year first), so a search for more than 1,000 is ordered by year, then by relevance.
  • Publications list stories in the order they were added to the publication, which is close to, but not always, their publication order.
  • Member-only bodies are the public preview (contentIsPreview: true).
  • Activity metrics (posting frequency, medians, top story) are on the profile rows of writers and publications you list; rows from searches or found writers carry the profile fields only.
  • Speed: each source pages one request after another (about 20 stories a second on one tag); several sources run in parallel.
  • Empty dataset: the status message names the cause: a tag, writer, publication, story or list that does not exist; filters that removed every story read (with the count per filter); or a monitor poll with nothing new. Nothing is charged.
  • Not available: reading counts and earnings (only the writer sees them), private and draft stories, the full body of member-only stories.

FAQ

How many stories can Medium Scraper collect in one run?

As many as you set in Max stories. A tag's newest-first archive goes back to its first story (5,000 stories in 258-300 s in our runs); ranked lists end at about 100 each; a search goes past 1,000 year by year. Use several tags or writers in one run for more.

Can I get the full text of Medium articles?

Yes. Turn on Include full text: free stories come complete as text and Markdown (and HTML with Include HTML). Member-only stories give the public preview, flagged contentIsPreview.

Can Medium Scraper get comments (responses)?

Yes. Include responses adds each delivered story's responses as rows with text, writer, claps and reply count; Include replies follows the threads.

Can I scrape a Medium writer's or publication's whole history?

Yes. List them under Writers or Publications (or paste their links). They are read newest first back to their first story; publishedBefore starts at an older date.

Can I run Medium Scraper on a schedule to get new stories only?

Yes. Turn on Only new stories since the last run and add a schedule. Each poll delivers and charges only stories no earlier poll delivered; an empty poll charges nothing.

Does Medium Scraper find writers and publications?

Yes. Set Search for to People, Publications, Tags or Lists, or turn on the profile rows to get one row per writer and publication you list or find.

Can I integrate Medium Scraper with other apps?

Yes: Make, Zapier, n8n, Google Sheets, Slack and more through Apify integrations, or webhooks when a run finishes.

Can I use Medium Scraper with the Apify API or MCP?

Yes. Use apify-client for Node.js or Python, or the MCP tab for Claude, Cursor and other MCP clients.

Do I need a Medium account, cookies or an API key?

No. Medium Scraper reads only what Medium shows a signed-out visitor.

Why did my run return no results?

The status message says why: the tag, writer, publication, story or list does not exist (check the spelling or link), the filters removed every story read (it lists how many each filter removed), or a monitor poll found nothing new.

How is Medium Scraper different from other Medium scrapers?

It reads whole tag archives and every ranked list rather than the ten stories of an RSS feed, carries distinct clappers and responses on every row, searches past the 1,000-result limit, adds writer medians and claps per follower, and costs $0.18 per 1,000 stories.

Your feedback

Found a bug or missing a field? Open an issue on the Issues tab.