Medium Articles Scraper

📝 Turn public Medium reading into structured data

Content researchers, writers, SEO teams, and developers can use this Actor to find public Medium articles and save structured rows. Each row can include the title, URL, author, publication details, tags, publication date, preview, reading time, clap and response counts, access status, and optional full text. That gives you a searchable source for topic research, author or publication review, content planning, and other work that starts with public Medium data.

Use cases

• Collect titles, authors, and dates from public pages with Medium Article Scraper.
• Find posts that match a phrase with Medium Blog Search.
• Look up matching pages with Medium Article Search.
• Check member-only status with Is Medium Free?.
• Review a topic collection with Medium Blog.

📚 Public Medium article records

What you get

Each saved row is one public Medium article. The Actor saves the first eligible occurrence when the same article appears again through another keyword, URL, handle, ID, or similar submitted value. Later matches are ignored. matchedInput shows the one value that first found the row.

Public access

It works with supported public Medium sources without a Medium account or authentication. It does not access private data or bypass paywalls. Member-only and paywall status is reported when the source exposes it.

Use the data

Use article metadata to compare topics, authors, publications, tags, dates, reading time, and public engagement. Request full text when you need public article text and it is available.

▶️ Choose a Medium target and run

Choose a target

Select exactly one target for a run. Keyword search accepts a list of same-kind keywords. Tag, author, publication, and article URL use their matching field. Fields for another target are ignored.

Keep a run focused

Date and preview filters apply to keyword, tag, author, and publication targets. Article URL ignores them.

Set the amount

Use maxItems to stop after a chosen number of article records. Leave it empty to return all available results until the source is exhausted. The public source may expose fewer matches than requested.

Reduce review work

For a small check, use a small maxItems value and keep includeFullText off. Use previewKeyword or date filters when you need a narrower set.

⚙️ Input

Input fields

Field	Type	What it does
target	string	Choose keyword, tag, author, publication, or articleUrl. The default is keyword.
keywords	array of strings	For keyword, search one or more words or phrases. Add each one as a separate item. The default is programming.
tag	string	For tag, enter one public Medium tag, such as programming.
authorUrl	string	For author, enter one public Medium author profile URL, such as https://medium.com/@username.
publicationUrl	string	For publication, enter one public Medium publication URL, such as https://medium.com/publication-name.
articleUrl	string	For articleUrl, enter one public Medium article URL, such as https://medium.com/@author/article-title-1234567890.
includeFullText	boolean	When true, add public full text when it is available. This does not bypass member-only access. The default is false.
includeMemberOnly	boolean	When true, keep member-only articles in the metadata results. Set it to false to remove them. The default is true.
maxItems	integer	Stop after this many article records. Leave it empty to return all available results until the source is exhausted. The default is 1.
publishedAfter	date string	For keyword, tag, author, and publication, keep articles published on or after this UTC date in YYYY-MM-DD form. articleUrl ignores this field.
publishedBefore	date string	For keyword, tag, author, and publication, keep articles published on or before this UTC date in YYYY-MM-DD form. articleUrl ignores this field.
previewKeyword	string	After discovery, keep articles with this word or phrase in the title or public preview. articleUrl ignores this field.

Successful default input

This example is copied from the public input of a successful current-beta run.

{
  "target": "keyword",
  "keywords": [
    "programming"
  ],
  "includeFullText": false,
  "includeMemberOnly": true,
  "maxItems": 1
}

🧾 Output

Run output

Field	Type	What it does
dataset	URL	Links to the default dataset view with the article rows from this run.

Article row fields

There is one article-row shape. Optional fields are omitted when they are not available or were not requested. Every row requires title, url, author.name, tags, and publishedAt.

Field	Type	What it does
title	string	Title of the Medium article.
url	URL	Public URL of the article.
author	object	Public author details.
author.name	string	Public display name of the author.
author.profileUrl	URL	Public Medium profile URL when available.
author.bio	string	Public author bio when available.
author.title	string	Public author profile title when available.
publication	object, optional	Public publication details when the article is linked to one.
publication.name	string	Public publication name.
publication.url	URL	Public Medium URL for the publication.
publication.description	string	Public publication description when available.
publication.followerCount	integer	Public follower count when available.
tags	array of strings	Public Medium tags attached to the article.
publishedAt	date-time string	Date and time when the article was published.
preview	string, optional	Public preview or subtitle when available.
readingTimeMinutes	integer, optional	Estimated reading time in minutes when available.
clapCount	integer, optional	Public clap count when available.
responseCount	integer, optional	Public response count when available.
access	object, optional	Publication, visibility, and member-only access details exposed by Medium.
access.isPublished	boolean	Whether Medium marks the article as published.
access.visibility	string	Visibility status exposed by Medium.
access.isMemberOnly	boolean	Whether Medium marks the article as member-only.
access.isPaywalled	boolean	Whether Medium marks the article as paywalled.
fullText	string, optional	Public article text when requested and available.
matchedInput	string, optional	The keyword, URL, or other submitted value that first found the article.

Genuine publication row

This complete row came from a successful current-beta publication run. It is not shortened.

{
  "title": "Support Vector Machines: A Progression of Algorithms",
  "url": "https://medium.com/data-science/support-vector-machines-a-progression-of-algorithms-841d63574825",
  "author": {
    "name": "Jimin Kang",
    "profileUrl": "https://medium.com/@jimin.kang821",
    "bio": "Lifelong student"
  },
  "tags": [
    "Machine Learning",
    "Support Vector Machine",
    "Support Vector Classifier",
    "Classification Algorithms",
    "Statistical Learning",
    "machine-learning",
    "data-science"
  ],
  "publishedAt": "2024-06-11T01:33:09.176Z",
  "access": {
    "isPublished": true,
    "visibility": "PUBLIC",
    "isMemberOnly": false,
    "isPaywalled": false
  },
  "publication": {
    "name": "TDS Archive",
    "url": "https://medium.com/data-science",
    "description": "An archive of data science, data analytics, data engineering, machine learning, and artificial intelligence writing from the former Towards Data Science Medium publication.",
    "followerCount": 828220
  },
  "preview": "MMC, SVC, SVM: What’s the difference?",
  "readingTimeMinutes": 11,
  "clapCount": 69,
  "responseCount": 0,
  "matchedInput": "https://medium.com/towards-data-science"
}

Genuine full-text row

This complete row came from a successful current-beta article URL run with full text requested. It is not shortened.

{
  "title": "The Freedom of Pet Projects",
  "url": "https://medium.com/@asvid/the-freedom-of-pet-projects-eadfe9cde099",
  "author": {
    "name": "Adam Świderski",
    "profileUrl": "https://medium.com/@asvid",
    "bio": "Software Engineer"
  },
  "tags": [
    "Software Engineering",
    "Career Development",
    "Programming",
    "Side Project",
    "Kotlin"
  ],
  "publishedAt": "2026-08-27T19:53:23.580Z",
  "access": {
    "isPublished": true,
    "visibility": "PUBLIC",
    "isMemberOnly": false,
    "isPaywalled": false
  },
  "preview": "In June, I learned how to change diapers and convinced myself that 5 hours of sleep is a luxury. Somewhere in between, I also shipped iOS…",
  "readingTimeMinutes": 4,
  "clapCount": 0,
  "responseCount": 0,
  "fullText": "The Freedom of Pet Projects\nIn June, I learned how to change diapers and convinced myself that 5 hours of sleep is a luxury. Somewhere in between, I also shipped iOS for Kombine — which was absolutely not on the roadmap.\nThe roadmap said: Android → Backend → Webapp → iOS sometime later. It was a good plan. Reasonable order. And that’s exactly the kind of plan a pet project doesn’t have to follow.\nI got bored, so I stopped\nThe web app worked. Golang backend, HTMX, a bit of Alpine.js — clean stack, no JS framework nonsense, exactly what I wanted to build.\nAnd I couldn’t make myself care.\nIt took me a while to realize why, but I get it now. While I created a core scaffold for the mobile app and oversaw the backend build, I completely delegated the web app to AI. No, it’s not slop. Not entirely. It’s just… I don’t know what is going on there. I get the principle-server-side HTML generation based on templates, plus some JS for handling real-time events. It doesn’t feel like mine. And when I was reviewing the web app after it was feature-ready, I had no idea how to debug it, or even how to guide the LLM to fix things effectively. I became the vibe-coder meme — “make me a website” kind of guy — and I didn’t like it.\nI’m a proud engineer. I enjoy my craft. I’ll take any new tool that makes me faster, as long as I’m still in control. Here I wasn’t. And it struck me after weeks of “developing” this service.\nWhat initially felt like boredom was actually hidden procrastination, under an “I have no idea what I’m doing here” sign.\nSo I stopped. No sprint was at risk. No stakeholder was waiting. No performance review depended on it.\nKombine started two years ago as a rough Android scaffold with no deadline — the kind of project that sits when work eats your energy. What’s changed since then is Claude Code: rebuilding a module went from a week to an afternoon, so nothing I built ever got too precious to abandon. That’s why boredom-driven pivots are affordable now. I shipped Android, built the webapp far enough to test pomodoro sync on a real second client, then lost interest — same as always. I went back to Android. The iOS migration looked too complex to touch. A month later I tried it anyway, and went from quick spike to App Store release without really planning to. And I was really curious whether Kotlin Multiplatform with Compose would actually work on iOS. It does.\nBoredom is a signal, not a problem\nThe best decision I made wasn’t the KMP migration. It was listening to the feeling that told me the web app wasn’t worth my Saturday.\nPet projects have one job: bring joy. That’s the whole deal — you don’t owe them a timeline, an order of operations, or a roadmap. The moment you’re working through something that stopped being interesting, you’ve lost the one thing that made it worth doing at all.\nPivot again?\nThis project began its life because I had a thought: how hard can it be to sync a pomodoro timer perfectly across multiple devices? I got that sorted. I added a task manager, started using it daily, moved from Todoist, and haven’t looked back.\nRecently I implemented a simple container for horizon goals: weekly, monthly, and so on. Something I used to do on paper. Sometimes. And I like it — it actually helps me track what I want to achieve and keeps me focused. Paper is great for writing and planning, but not for tracking. So instead of the automatic activity tracker I had on the roadmap, I’m exploring this direction instead.\nObviously.\nStay foolish\nI’d encourage anyone to have a pet project like this — to fool around, make stupid decisions, and invest time in them, because that’s all that’s at stake: your time-exchanged for learning, experience, and the joy of building. I had projects before this one that all died in a sad and lonely place when I got bored — too complex, too simple, stuck on something trivial, momentum gone. Every software engineer has a graveyard like that.\nI would never randomly pivot the roadmap in an organization with multiple teams and other devs depending on it. Don’t try it at work. Do try it at home. I went where my curiosity dragged me, and the worst case was wasting some time learning KMP isn’t what I want on iOS — and since I am not a fan of React, I’d rewrite it natively. OK, sure, Claude Code would.\nI didn’t optimize for time or money. I didn’t follow the roadmap. I got carried away by curiosity.\nThat’s a pretty good compass.\nDon’t ever hire me as a project manager. Just leave me with the technology.\nOriginally published at https://swiderski.tech.",
  "matchedInput": "https://medium.com/@asvid/the-freedom-of-pet-projects-eadfe9cde099"
}

💳 Pricing

How charging works

Pricing is pay per event. The Article event charges once for each article saved to the dataset. Prices vary by Apify tier and are shown on the Store page. There is no separate run-start charge listed for this Actor.

🔌 Integrations

Dataset access

Read the dataset in Apify Console or through its API. Export the rows in the formats supported by Apify, or send the dataset to your own workflow.

❓ FAQ

What happens when the same article appears more than once?

The first eligible occurrence is saved. Later matches for that article are ignored, and matchedInput keeps the value that first found it.

Can I search more than one keyword in a run?

Yes. Choose the keyword target and add each word or phrase to keywords. The other target types use their matching single value, and fields for another target are ignored.

Can I retrieve the article text?

Set includeFullText to true. The Actor returns public full text when it is available. It does not bypass member-only access or paywalls.

What happens to member-only articles?

Set includeMemberOnly to true to keep their metadata rows, or set it to false to remove them. The access fields show the status exposed by Medium.

Can I use date or preview filters with an article URL?

No. publishedAfter, publishedBefore, and previewKeyword apply to keyword, tag, author, and publication targets. The articleUrl target ignores them.

What does leaving maxItems empty do?

It returns all available results until the source is exhausted. The public source can still provide fewer rows than expected.

Why did the run return fewer articles than my limit?

The public Medium source may expose fewer matching articles, or the date and preview filters may remove matches. The Actor does not promise an exact result count or unlimited historical coverage.

Do I need a Medium account?

No. The Actor uses supported public Medium sources and does not require a Medium account or authentication.

Can I combine a tag, author, and publication search in one run?

No. Choose one target per run. Start another run for a different target or filter set.

📝 Changelog

0.0: Initial release

🆘 Support

For issues, questions, or feature requests, file a ticket and I'll fix or implement it in less than 24h 🫡

🔗 Related Actors

• Webpage Text Extractor extracts readable article text and metadata from public web pages when you already have page URLs.
• RSS Feed Reader turns public RSS and Atom feeds into rows with titles, links, dates, authors, and content.
• Naver Blog Review Scraper collects public Naver Blog posts with content, author details, and keyword context.
• MarketWatch Scraper collects public financial articles with summaries, text, authors, keywords, and tickers.
• arXiv Papers Scraper finds public research papers with titles, authors, abstracts, categories, dates, and source URLs.

Made with ❤️ by Maxime Dupré