Twitter Article Scraper

📰 Turn selected X Articles into usable text

Researchers, journalists, and AI teams can submit public X post URLs or numeric post IDs and get the public Article body as Markdown and plain text. Each dataset row keeps the source post, author, publication dates, links, media references, content blocks, and public engagement counts when X exposes them, so you can cite, search, archive, or analyze the text.

• Convert a public X post into Markdown with X Long Form Article Scraper.
• Read a selected long-form post for a research file with Twitter Long Form Article Scraper.
• Save Article text and source details with Twitter Article Extractor.
• Pull one public Article from its post URL with X Article Scraper.
• Build a structured dataset from submitted Article sources with Twitter Article Scraper.

📄 See the Article, source post, and author

A dataset row groups three top-level objects: article, sourcePost, and author. Optional source fields can be absent when X does not expose them.

Returned data

The article object holds the Article body, structure, media, and links. sourcePost keeps the containing post and its public engagement snapshot. author keeps the public profile fields that X exposes.

▶️ Run the X Article scraper

1. Choose Public post URLs or Numeric post IDs in Find Articles by.
2. Add the matching public sources. The other source section is not used for that run.
3. Start the run and open the default dataset to review the nested Article rows.

What the Actor reads

The Actor reads selected public long-form Articles that are accessible without an X login. It does not search X or discover Articles from a profile, timeline, keyword, or feed. It does not scrape ordinary posts, threads, replies, quote posts, or profiles.

Access limits

Private, deleted, paywalled, geo-restricted, login-gated, and CAPTCHA-protected content is outside the supported scope. Media is returned as source references; the Actor does not download or permanently store media files.

⚙️ Input

Choose one discovery method, then add sources in the matching section. The other section is not used for that run.

Input fields

Field	Type	What it does
discoveryMethod	choice	Required. Choose postUrls to use public post URLs or postIds to use numeric post IDs.
postUrls	array of objects	When discoveryMethod is postUrls, add one or more full X or Twitter status URLs for posts that contain public long-form Articles.
postUrls[].url	string	Full public X or Twitter post URL with a numeric status ID.
postIds	array of strings	When discoveryMethod is postIds, add one or more numeric IDs for public X posts that contain long-form Articles.

Example input

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

{
  "discoveryMethod": "postUrls",
  "postUrls": [
    {
      "url": "https://x.com/tonysimons_/status/2060527240098587085"
    }
  ]
}

🧾 Output

The output schema provides a link to the default Apify dataset. Each saved row follows the nested shape documented below.

Run output

Field	Type	What it does
dataset	URL	Opens the default dataset for the run.

Dataset row

Field	Type	What it does
article	object	Groups the public Article, its content, media, and links.
article.id	string	Stable Article ID assigned by X.
article.url	string	Canonical public URL for the Article.
article.title	string	Article title when the source provides one.
article.previewText	string	Source-provided Article preview when available.
article.publication	object	Source dates for the Article.
article.publication.createdAt	date-time string	Time when the source says the Article was created.
article.publication.firstPublishedAt	date-time string	Time when the source says the Article was first published.
article.publication.lastModifiedAt	date-time string	Time when the source says the Article was last modified.
article.content	object	Holds readable and structured Article content.
article.content.markdown	string	Complete publicly available Article body rendered as Markdown.
article.content.plainText	string	Article body rendered as plain text.
article.content.blocks	array of objects	Source content blocks when available.
article.content.blocks[].type	string	Source-reported type of a content block.
article.content.blocks[].text	string	Text carried by a block when available.
article.content.blocks[].level	integer	Source-reported heading or nesting level when available.
article.content.blocks[].items	array of strings	Text items in a list-like block when available.
article.content.blocks[].inline	array of objects	Inline text runs and formatting in a block when available.
article.content.blocks[].inline[].text	string	Text in an inline run.
article.content.blocks[].inline[].formats	array of strings	Source formatting marks on an inline run.
article.content.blocks[].inline[].url	string	Link destination for an inline run when available.
article.content.references	array of objects	Source references used by structured Article content.
article.content.references[].id	string	Source ID for a content reference when available.
article.content.references[].type	string	Source-reported reference type.
article.content.references[].url	string	Destination URL for a content reference when available.
article.media	object	Source references for cover and inline media.
article.media.cover	object	Cover media reference when available.
article.media.cover.id	string	Source media ID when available.
article.media.cover.url	string	URL where the cover media is available.
article.media.cover.type	string	Source-reported cover media type when available.
article.media.cover.width	integer	Cover media width in pixels when available.
article.media.cover.height	integer	Cover media height in pixels when available.
article.media.inline	array of objects	Source references for media inside the Article body.
article.media.inline[].id	string	Source media ID when available.
article.media.inline[].url	string	URL where the inline media is available.
article.media.inline[].type	string	Source-reported inline media type when available.
article.media.inline[].width	integer	Inline media width in pixels when available.
article.media.inline[].height	integer	Inline media height in pixels when available.
article.links	array of objects	Links embedded in the Article.
article.links[].text	string	Text shown for a link in the Article.
article.links[].url	string	Resolved destination of the link.
sourcePost	object	The public X post that contains the Article.
sourcePost.id	string	Stable ID of the containing post.
sourcePost.url	string	Canonical public URL of the containing post.
sourcePost.engagement	object	Public engagement counts for the containing post when available.
sourcePost.engagement.likes	integer	Reported like count.
sourcePost.engagement.replies	integer	Reported reply count.
sourcePost.engagement.reposts	integer	Reported repost count.
sourcePost.engagement.quotes	integer	Reported quote-post count.
sourcePost.engagement.bookmarks	integer	Reported bookmark count.
sourcePost.engagement.views	integer	Reported view count.
author	object	Public profile shown for the Article author.
author.id	string	Author ID assigned by X when available.
author.name	string	Public author display name when available.
author.username	string	Public author username.
author.profileUrl	string	Public author profile URL when available.
author.isVerified	boolean	Whether X marks the author as verified.
author.description	string	Public profile description when available.
author.location	string	Public profile location when available.
author.websiteUrl	string	Website shown on the public profile when available.
author.profileImageUrl	string	Public profile image URL when available.

Genuine current-beta row

The row below is shortened. The markdown, plainText, blocks, references, and inline media values use "..." as omission markers. The remaining values are from a successful current-beta run.

{
  "article": {
    "id": "2052796100608974848",
    "url": "https://x.com/trq212/status/2052809885763747935",
    "title": "Using Claude Code: The Unreasonable Effectiveness of HTML",
    "previewText": "This is now also on the Claude Blog.\n\nMarkdown has become the dominant file format used by agents to communicate with us. It’s simple, portable, has some rich text capability and is easy for you to",
    "publication": {
      "createdAt": "2026-05-08T17:56:30.000Z",
      "firstPublishedAt": "2026-05-08T17:56:30.000Z",
      "lastModifiedAt": "2026-05-19T21:04:23.000Z"
    },
    "content": {
      "markdown": "...",
      "plainText": "...",
      "blocks": "...",
      "references": "..."
    },
    "media": {
      "cover": {
        "id": "2052796450510348288",
        "url": "https://pbs.twimg.com/media/HHz_ftzaIAAwkQs.jpg",
        "type": "ApiImage",
        "width": 2400,
        "height": 960
      },
      "inline": "..."
    },
    "links": [
      {
        "text": "Claude Blog",
        "url": "https://claude.com/blog/using-claude-code-the-unreasonable-effectiveness-of-html"
      },
      {
        "text": "https://thariqs.github.io/html-effectiveness",
        "url": "https://thariqs.github.io/html-effectiveness/"
      },
      {
        "text": "https://x.com/trq212/status/2017024445244924382",
        "url": "https://x.com/trq212/status/2017024445244924382"
      },
      {
        "text": "https://thariqs.github.io/html-effectiveness/",
        "url": "https://thariqs.github.io/html-effectiveness/"
      }
    ]
  },
  "sourcePost": {
    "id": "2052809885763747935",
    "url": "https://x.com/trq212/status/2052809885763747935",
    "engagement": {
      "likes": 17841,
      "replies": 1098,
      "reposts": 2292,
      "quotes": 1938,
      "bookmarks": 34604,
      "views": 14717658
    }
  },
  "author": {
    "username": "trq212",
    "id": "352806502",
    "name": "Thariq",
    "profileUrl": "https://x.com/trq212",
    "isVerified": true,
    "description": "Claude Code @anthropicai. prev YC W20, @spc, @medialab",
    "location": "SF via Toronto",
    "websiteUrl": "http://thariq.io",
    "profileImageUrl": "https://pbs.twimg.com/profile_images/1976939058741039104/r3GgzqRh_200x200.jpg"
  }
}

💳 Pricing

This Actor uses pay-per-event pricing. The X Article event is charged only when one public long-form Article is extracted and saved. Check the Actor page for the current event price.

🔌 Integrations

Results are available in the default Apify dataset and through the Apify API. Use the dataset export tools or the dataset API URL to pass Article rows to your own research, archive, or text workflow.

For a short walkthrough of using an Actor in an Apify workflow, see:

❓ FAQ

Can I submit URLs and numeric IDs in one run?

No. Choose one discovery method and fill in its matching section for each run.

What if a post is an ordinary tweet or its Article is not public?

The Actor is for public long-form X Articles only. The source must be readable without an X login, and ordinary posts and restricted content are outside the supported scope.

Do I need an X login, developer API key, or cookies?

No. You do not need to give the Actor an X login, developer API key, or cookies.

Does the Actor find Articles from a profile or keyword?

No. Submit the public post URL or numeric post ID for each Article you want to read. The Actor does not perform profile discovery or X search.

Which post URL should I submit when an Article was reposted?

Use the public post that contains the Article. An unrelated repost is not a reliable Article source.

What are content blocks and references for?

They preserve source-provided structure and reference details for workflows that need more than the Markdown or plain-text rendering.

Are engagement counts historical?

No. They are public counts captured with the Article and can change after extraction. Optional counts may be absent when X does not expose them.

Does the Actor download Article images?

No. It returns source media references, including available URLs and dimensions, but it does not create permanent media copies.

Can it read private, subscriber-only, or paywalled Articles?

No. The Actor does not bypass login, payment, geographic, deletion, or other access restrictions.

📝 Changelog

v0.0 (18-09-2026)

• 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

• X Articles to Markdown Scraper - Convert selected public X Articles to Markdown and plain text with a sibling Article workflow.
• Twitter Scraper - Collect ordinary X posts, profiles, and timelines when the source is not an Article.
• Webpage Text Extractor - Extract Markdown and article text from public web pages outside X.
• URL to BibTeX Converter - Turn Article URLs into BibTeX citations for reference lists.
• Substack API Scraper: Posts, Authors & Newsletters - Collect public newsletter posts when the long-form source is on Substack.

Made with ❤️ by Maxime Dupré